• Home

  • Documents

  •  · . ' ANDREW PHASE II PROJECT OBJECTIVES . ' • ·UPDATE MODELING AND ANALYSIS...
81
. ' ' II ' " , ANDREW PHASE II ' JOINT1 INDUSTRY PROJECT ' ' ' ' PROGRESS MEETING NUMBER 1 ' ' • '. NOVEMBER 16, 1994 ' PMB ENGINEERING, ' ' .

 · . ' ANDREW PHASE II PROJECT OBJECTIVES . ' • ·UPDATE MODELING AND ANALYSIS RECIPE· JOINTS ' ' ·BRACES WAVE PROFILE FOUNDATIONS . ' ' CONDUCTORS· • ANALVZE PLATFORMS USING

  • Upload
    hatruc

  • View
    224

  • Download
    0

Embed Size (px)

Citation preview

Page 1:  · . ' ANDREW PHASE II PROJECT OBJECTIVES . ' • ·UPDATE MODELING AND ANALYSIS RECIPE· JOINTS ' ' ·BRACES WAVE PROFILE FOUNDATIONS . ' ' CONDUCTORS· • ANALVZE PLATFORMS USING

II

ANDREW PHASE II

JOINT1 INDUSTRY PROJECT

PROGRESS MEETING NUMBER 1

bull

NOVEMBER 16 1994

PMB ENGINEERING INC~

bull

ANDREW PHASE II JOINT INDUSTRY PROJECT PROGRESS MEETING NUMBER 1

NOVEMBER 16

AGENDA

INTRODUCTION amp OVERVIEW OF PROJECT (R L SMITH) 900

UPDATE RECIPE (RW LITTON) 930

BREAK 1030

I

MSL PRESENTATION ON JOINTS amp CONTINUE UPDATE RECIPE 1045

ANDREW STORM CONDITIONS UPDATED (RW LITTON) 1145

LUNCH 1200

PLATFORM SELECTION (DK DOLAN) 100

BAYESIAN CALIBRATION (RAJIVAGGARWAL) 200

DISCUSSION(R L SMITH) 330

REVISIT WORK PLAN (R L SMITH) 400

middotADJOURN 430

I ~

1 It

I t I I

ANDREW PHASE II

UST OF PARTICIPANTS

bull

AMOCO PRODUCTION COMPANY

bull

CHEVRON PETROLEUM TECHNOLOGY COMPANY

bull

EXXON PRODUCTION RESEARCH COMPANY

bull MINERALS MANAGEMENT SERVICES

bullmiddot

MOBIL RESEARCH AND DEVELOPMENT CORPORATION

I bull PHILLIPS PETROLEUM COMPANY

bull bullSHELL OIL COMPANY

bull UNOCAL CORPORATION

i I I

j

ANDREW PHASE I

OBJECTIVES AND RESULTS

bull DEVELOPED A DATABASE OF PLATFORMS AFFECTED BY ANDREW

bull DEVELOPED AN ULTIMATE CAPACITY ANALYSIS RECIPE amp

PROCEDURE

bull IDENTIFIED DEFICIENCIES IN ULTIMATE CAPACITY ANALYSIS PROCESS

bull DEVELOPED A BAYESIAN UPDATING PROCEDURE

l

bull ESTABLISHED A GLOBAL BIAS FACTOR FOR PLATFORMS

bull PROVIDED EXAMPLES FOR THE USE OF THE BIAS FACTOR

ANDREW PHASE II

PROJECT OBJECTIVES

bull middotUPDATE MODELING AND ANALYSIS RECIPEmiddot

JOINTS

middotBRACES

WAVE PROFILE

FOUNDATIONS

CONDUCTORSmiddot

bull ANALVZE PLATFORMS USING UPDATED RECIPE

PLATFORMS SELECTED FROM FAMILY middotOF EXISTING STRUCTURES USED IN PHASE I

POSSIBLE middotREPLACEMENT IF MORE MEANINGFUL PLATFORM(S) middotFOUND

bull EXPAND BAYESIAN CALIBRATION

TWO BIAS FACTORSmiddot DEVELOPED JACKET AND FOUNDATION

UTILIZE RESULTS OF THE APlMMS FOUNDATION PROJECT

PRODUCE EXAMPLE USING EXPANDED PROCEDURE

-middot

bullbull

SUBSEQUENT PROJECTS

bull MMS INSPECTION PROJECT

bull APlMMS CAISSON PROJECT

bull APlMMS FOUNDATION PROJECT

bull TRIALSBENCHMARK JIP

I I It

ANDREW PHASE II

PROJECT TEAM

bull R L SMITH PROJECT MANAGER

bullbull F middot

J PUSKAR PROJECT ADVISOR

bull Rw LITTON TECHNICAL MANAGER

bull RAGGARWAL TECHNICAL DEVELOPER

bull D K DOLAN TECHNICAL ADVISOR middot

bull C A CORNELL RELIABILITY CONSULTANT

bull F MOSES RELIABILITY CONSULTANT

bull middotMSL ENGINEERING LTD TUBULAR JOINT CONSULTANT

q gtI

ANDREW PHASE II

UPDATE MODELING middotAND ANALYSIS RECIPE

bull CONDUCTOR RESISTANCE

bull WA VE FORCE PROFILE

bull BRACE MODELING

middotbull FOUNDATIONSmiddot

bull JOINT MODELING

~ I I

-ANDREW PHASE II

CONDUCTOR RESISTANCE

shybull RESISTANCE NOT INCLUDED IN PHASE I

bull CONTRIBUTION TO OVERALL RESISTANCE A PHASE II

STUDY ISSUE

middotbull middotCOMPARE FOR 4-LEG AND 8-LEG PLATFORM

bull CONSIDER CONDUCTOR POSmON IN GUIDES

- 13 LEFT 13 CENTER 13 RIGHT

Project ST151K Model Conductor version 1 Mon Nov 14 143140 1994

- ---- ---

--

117 s

111

7i

31

0 -5 -~ - ~()

- iltgt --lo

-(pO

-so

bull bullf OC)

-120

-ISo

-ISO

z

CAgt ~x Singl_e Conductor Model

bull

=I=

-_

-_

-gtshy

~ -

~

I

I~

- gt

~-----------____________________

Project STlSlK Model dir_2 version 7 Mon Nov 14 124200 1994

177 s

Ill

74

37

0- - e -zo - ~o

- Jo

-100

-10

-ISO

z

UV 1-+x

Project STlSlX Model Conductor Version 1 Mon Nov 14 140620 1994

bullmiddot

I

I

I I

I

I I

I I I I I

1 ==

-L

-L - L

-

- L

-

-

L

-

Uniform Displacement step 6 Inelastic Events Legend

Elastic

Strut Buckling Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield Beam Clmn FUlly Plastic Fracture

--

i~~~~~~~~~~~---~--~~-J___--JJ~_middot~~~=-

Project ST151K Model Conductor Version1 Mon Nov 14 142735 1994

---

~

I

I

middot

I

I

I

- shy- shy

- shy

- shy

4x z

CIV Uniform Displacement step 33 Inelastic Events Legend

bull Elasti~ Strut Buckling Strut Residual Strut Reloading Plastic StrutNLTruss Beam Clmn Initial Yield Beam Clmn-Fully Plastic Fracture

gt bull

Project STlSlK middot Model Conductor Version 2 Mon Nov 14 142025 1994

-----------------

bullf

I

I

I

II I I

I I I I

I ==

-

Variable Displacement step 9 Inelastic Events Legend

Elastic Strut Residual

bull

Strut Buckling Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield Beam Clmn Fully Plastic Fracture

--

Project ST151K Model Conductor Version 2 Mon Nov 14 142145 1994

I I I I

I I

I

I

I - - -

-Ishy-

-

Variable Displacement step 35 InelasticEvents Legend

Elastic Strut Buckling Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield Beam Clmn Fully Plastic Fracture

Project ST151K Model Conductormiddot

Version 1

Clgt - Cut Plane Force Fx Mon Nov 14 140051 1994

- -_ vii I fOIZM

0 _

~

N ID 0 bull 0

x 0

middotrt ~ () bull u 0 ~ 0 Ir

N bull

0

0 1 2 3 4 5 X Deck Disp [Ft)

I I 1 1 WbullTI- bullbull

lltcxi lr-C

011lt~~~~~~~---~~~~~~~~L~~~~~~~--~~~~~~~~i~~~~~~~--

Uniform Node Displacement

shy

Ji

r

Comparison of Pushover Curves for ST 151K Variable Load Pattern and Monotonic Load Pattern (Andrew)

Soll Improved 100

6000

5000 -i 4000

f

llS

3000 Ul

J 2000

10 c11 c1lgtolts 1000

0 v

bull ~

f

_ ----- --- ---shy

bull --Variable Lold

- - Andrew Load

I I I I I I

4 5 62 30 1

Deck Dlsplacement (ft)

bullbull r 1 II

~

ANDREW PHASE Il

WA VE FORCE PROFILE

bull DECK WA VE FORCE ~

PHASE I USED SECTION 17 (1056H degCd)

PHASE Il USE REVISED SECTION 17 (1000H Cd)

bull PHASE I USED SINGLE FORCE PROFILE FOR PUSHOVER

- USED WAVE HEIGHT CLOSE TO ULTIMATE CAPACITY

bull PHASE J

Il WILL STUDY THE

USE OF VARIABLE

PROFILE ~

- IF FORCE LEVEL INCREASES (OR DECREASES) BY A FACTOR EQUIVALENT TO A 2 FOOT CHANGE IN WA VE HEIGHT THE INCREMENTAL FO~CE PA1IERN IS CHANGED ACCORDINGLY (CAP FEATURE)

- INITIAL EXAMPLE

f~

a

middotn

a

I N -a a a amiddot

+

~ a

l

a w gt 0c 0 gt a I shyc -CJ wc w gtc(

== c( I shyZ w E W a 0 z- 8

ii sect

~ ~ bullT 0

I if

x ~

bullbullbull

II ~------------________

ANDREW PHASE II

WA VE FORCE PROFILEmiddot

CAP AUTOMATIC APPLICATION FEATURE

STAT

fR1 0 lt f lt1

STAT

STAT

f(R3-R2) middot f lt1

Project ST151K Model Wave_Loading Version 1 rue Nov 8 164107 1994 PT

E~ nss FT

SL 14amp FT

a 111 FT

EL 14 li

E-1 37 FT

El o FT

1middot1middot ______________ _ ________ __ _ __

0

200 400 600 800 1000 12000 Force at Elevatloii (kip)

200

a ll_

Wave Force at Major Elevations for Discrete Wave Heights

180

bullHmiddot40ft

Hmiddot495ft

H bull 51 ft

Hmiddot55ft_

H-59ft I

Andrew (H bull 60 86 ft) II

i I

_11---__

- middot- Hmiddot30ft

-

- bull -

middot bull bull bull middot

---Hmiddot53ft

bullbullbullbullHmiddot57ft

-----H-63ft

Elevation Above Mudllne 100

(ft)

180

140

120

80

60

40

20

Loading at Major Elevations

---H-49511

-deg-Hmiddot608611

0 200 400 600 800 1000 1200

Force It Elevatlon (kip)

--

Base Shear vs Wave Height ST151K

6000

5000 1---------Ult1mateCapacitybull 4890 kip (100 Improved Soll)-----------~--

4000a a lii ~ 3000

81 ampI

2000

1000

0 +-~~~--i~~~~~~~~~-+-~~~--lf--~~~~~~~~-+-~~~--l

30 35 40 45 50 55 60 65

Wave Height (ft)

bull middot

~

bull

II

2

g

~

It)

0

cCD E 0 E a c E euro~ CDshygtItgtmiddoto ~ti c -a Gic CD gt ~

-= c -Cll

I z

bullO 0i i ~ g 0

() CI -bull(q-ll) iuawow llu1wniiaAO

11

shy

116

114

112

g 110

bullc = 108

~ ~ 106 -sect 104

102

100

98

96

-

6555 60

Wave Height vs Center of Action ST151K

30 35 40 45 50

Wave Height (ft)

t

Comparison of Pushover Curves for ST 151K Variable Load Pattern and Monotonic Load Pattern (Andrew)

Soll Improved 100

--- --------- --- ----6000

5000

- 4000 8 - I

1 3000 Ul

81 m 2000

_J -shy

1000

---Variable Load

- - Andrew Load

0 f-~~~~~r--~~~~-t-~~~~~-t-~~~~~+-~~~~----t~~~~~-1

0 1 - 2 3 4 5 6

Deck Displacement (ft)

-middotshy

1 t I I ~

ANDREW PHASE II

BRACE MODELING

bull MARSHALL STRUTS USED FORPHASE I

bull PHASE II WILL ASSESS DIFFERENCE IN USING BUCKLING BEAMS AND MARSHALL STRUTS

middot bull COMPARE FOR 2D FRAME

bull COMPARE FOR PLATFORM 177 WHERE LOWER

BAY EXHIBITS LARGE BRACE MOMENTS

Project andrew Model beamframe Version 1 Mon Nov 14 145447 1994

~

middot Frame Test buckling beam-column model

-

Project andrew Model strutframe Version 1 Mon Nov 14 145754 1994

bull

Frame Test strut model

middot-middot _____________--_

_middotmiddotr_~ --ll--~-middot__

Buckling Beam bull

Strut (K=O 65)

Lateral Displacement

ANDREW PHASE II

FOUNDATIONS

bull ~LICIT NONLINEAR API FOUNDATION USED INiPHASE I

bull FOUNDATIONS

JIP SELECTED MINI-VANE DATA FOR

STRENGTH PROFILE

bull COMMENTS ON BENCHMARK COMPARISONS

- SIGNIFICANT VARIABILITY DUE TO FOUNDATION MODELING METHODS

II Ii

z

ctV 4x SPlO caissonfy-42s soil-api(lf-097step-114)

Inelastic Events Legend Elastic Strut Buckling

Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield

--------shyBeam Clmn Fully Plastic Fracture

~--~~~~~~~~~--~----~__i~~---~~~~__--~

Pro)ect apipile Model South Pelto 10 Version Wed May 18072235 1994 - -

1bullt

+

t

f i bullbullbull

bullbull bullbull

I I

5

~x SPlO Caissonfy=42s soil=api(lf=lOOstep=137)

Inelastic Events Legend

Elastic Strut Bucklilg

Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield

earn Clmn Full~ Plas1c Fracture

ANDREW PHASE II middot middot

JOINT MODELING

middotbull API RIGID-PLASTIC USED IN PHASE I

a ELASTIC-BRITTLE USED IN PHASE IB

amiddot MSL WILL PROVIDE NONLINEAR FORCE-DEFORMATION

AND MOMENT-DEFORMATION RELATIONS FOR PHASE II

a MODELING OPTIONS FOR PHASE II -

l

- UNCO~LEDmiddotONE DIMENSIONAL MEMBER END JOINT MODELING

J Ii

ANDREW PHASE Il

JOINT MODELING OPTIONS DISCUSSION

bull JD COUPLED INTERACTION SURFACE

- IDEAL

- DATA NOT AVAILABLE TO CALIBRATE

bull 2D COUPLED INTERACTION SURFACE

- SIGNIFICANT MODELING IMPROVEMENT

- DATAAVAILABLE

- IMPLEMENTATION PRACTICAL BUT NEEDS DEVELOPMENT AND CALIBRATION

bull 2D UNCOUPLED INTERACTION SURFACE

- SIGNIFICANT MODELING IMPROVEMENT

- DATA AVAILABLE

- IMPLEMENTATION PRACTICAL AND AVAILABLE

ANDREW PHASE II

JOINT MODEJJNG OPTIONS ILLUSTRATIONS

P

e

ANDREW PHASE II

JOINT MODELING PHASE II METHOD

e

I I I

ANDREW PHASE II

ANDREW STORM CONDmONS UPDATED

bull -PHASE I USED OCEANWEATHER HINDCAST-(Nov 1992)

bull middotmiddotPHASE II WILL USE THE REVISED HINDCAST (Nov 1994)

bull lt c lt I lt

2-2

l---4-~~~+-~~-+-~~~1--~~-+-~~--l--~T--+-~~--i24

f-tJVtMfiz~ 1Cfl (r STvb(

Plollol bullbull 5middotNOYmiddot9214bull1J Irbullbull Illbull [ANDREVl~NDREVCUSTbullMAXll 5-NOY-1992 10bull54 - 9208

HEIGHT lml

I~ MAXI

I L 1~middotamp SZ~- ~A I 13c

I J) ( r I X~0iWJ~0gtJ gt L gt gt I 1c I 120

-- N

~ deg

I I I I I I e C I ~=-41 122 -94 -92 -90 -88 -86 -84 -82 -80

-

Figure 7 contours of maximum hindcast significant wave height

_

~

~tl ~

NOltNttIJ 17lt ltNS S iJJ)y

Maximum Significant Wave Height - Hurricane Andrew 1992

30 ~

26 ~ I ~-~~~- == -shy -~ 35-_

24deg1--shy --------~- middot-middot-------middotmiddotmiddot----middot------middot-middot~middot- - - ~-----1---middot

22middot ~~~~~~~-~ -94 -92 -90 -88 -84deg-86 -82 -somiddot

c

-

Platform Selection bullbullbull1111111111111

0 BasisCriteria for Selection middot

0 Review of Phase I Platforms middot 0 Alternative Platforms

0 PMB Recommendations

0 Participants Comments

gt

L

l I

BasisCriteria for Selection bullbullbull1111111111111

0 Benefit for Calibration - Classification (UnexpectedExpected)

- Failure Mode (StructuresFoundations)

- Damaged StructuresKnown response

- Better Representation of Full ~opulation

0 Availability of Data - Existing Models

- Inspection Data

- Damage Data

PREDICTED BEHAVIOR

-

OBSERVED

BEHAVIOR

-SURVIVED DAMAGED FAILED

SURVIVED

ST134W WD90A

-MC311 -

MC397 -

--

-

-

ST151K ST130Q

-

-

-

DAMAGED - -

-

-

--

-

-

ST52 SS139 ST177B -

ST161A

-_

-

FAILED

-

-

-

ST151H ST130A ST72

Operator - Platfonns - Platfonns Survived Damagedamp

- Falled -

bull

-

-

Jle eBDIGIPllDbl -Amocoshy 30 1 Chevron -

143 12 Exxon 72 Mobll 22 1 Phllllps 7 bull

-Shell 39 Unocal 25 1

Total Partlclnts Platfonns 338 15

Other Operators bull- 310 13

-

Total Platfonns - -

648 28

fbml

Platforms Considered 6 7

tie of Total Platfonns 1 25

OVERVIEW OF PLATFORMS SURVIVEDDAMAGEDFAILED

-

shy

v

Platforms Evaluated in Phase I

-

-Platfonn Water Year Number Analysis Avallabillty of Availability Platform

- Name Depth rl of Model DetaDed rl ClassUlcatlon

Installation legs Inspection Soll Report Based on Information Phase I Calibration - -ft

- -- Suaill flalfnrm ~BSi -STISIK 137 1963 8 CAP Yes Yes Un eel Sarvlval

-STIJOO 170 1964 4 CAPmiddotshy Yes ST134data u ST134W 137 1981 4 CAP Yes Yesmiddot E

-

edSarvlval -

middotsurvival

WD90A 184 - 1964 8 CAP Yes E -

- Sarvlvill

MC3ll 343 1978 8 KARMA

MC397 468 1991 4 KARMA -

bull ~reglmmiddot- -

Yes Sure Survival

bull Yes - Sare Sanlval-

-T23STS2 63 1969 4 CAP

T2SSS139 62 1969 4 CAP -

Yes

Yes

Yes Ex

Yes Une

bull SurvivalLikely to DamalI

middot bull SanlvalLlkely to DamalI

ST161A 118 1964 8 USFOS

Eail11a pound1al(bullllm Clss ST177B 142 --1965 8 CAP

STISIH 137 1964 - 8 CAP C

STIJOA 140 19S8 8 CAP - -

Yes

Yes

ollamed

Yes

- Yes E

-

at about 1 mUe

Yes

ST134data

bull SafvlvalLlkelv to Dam- bull- -

E middotrahre

Emeried Failure -

Likely FaUare

T21ST72 61 1969 4 CAP Collamed No Likely FaUure

Platforms Identified for Evaluating Their Likelihood of Falling Under UnexpeCted Survival Category

CllEllATOa

AllB

llLOCll

Sl1lvcnJllE

NAME

YEAR

INSTAUJCD

WATER

IJBPJB

middotmiddotDISTANCE

ftOMSllOllE

(MIUIS)

Ill

(M)

AMOCO BllW

STllI

WDI075

A

bull D

4

n uo 17J

44

32

bull 20

10

bull CllEVllOH STtllO c I 1bull 21 11

STl130 D 0 161 21 11 STllJ4 I 0 137 D - 11

STIUI G I 137 32 11

STIUI I 0 121 32 11

STIUI

J 0 140 32 11

STIUI It 4 137 32 11

STllSI PROl)I 0 137 32 11

STllSI PROD-2 0 137 32 11

STll7 A 0

140 35 11

WDl117 c 65 214 34 10

DXON STshy E 51 14 bull STl165 A t3 bull 34 WD0073

WD0073

WD0073

A _ c

4

62

4

IA IA

17J

22

22

22

bullbullbull WD003-shy E 65 161 Z7

A 0 IN 31 MOBD Bl12I

ssn A bull

50

50

31

7

bull 7

MIJllPllY SS114 B 7 50 7 7

PlllUIPS SSl14t A 55 33 7

SAMEDAN ST17J A 4 117 7 bull 7

SJIELL WD0103 A 65 223 27 WDllOJ bull 65 221 27 WDll04 c 65 221 27 WDOI04 D 67 2 27

UNOCAL

SSl20I

SSl20I

bull SSl20I

SS0209

SS0209

SS0209

SS0209

SS 0253

SS 0253

E

I

PtJllE

A

bull D

G A

c

0

4

4

7J

61

0

0

69

103

97

97

9S

100

95

100

165

175

34

34

34

34

34

34

34

54

54

bullbullbullbullbullbullbullbullbull

Platforms Recommended for Phase II

llalr y_ w- Aot Av- _ I 1 Doplla Model nm bull bull _ bull I I H11 s18-1 Irht lbwl~ bullbullbull(-illtdlel OM lllMtD

II ---

-STISIIC 137 1963 I CAP Yeo Yeo I Yeo

STtshy -170 1964 4 CAP Yeo ST134dlU I Yeo

STl34W 137 Yeo Yeo -1981 4 CAP D Yeo

WD90A 114 1964 I CAP Yeo - D

MC311 30 1971 I KARMA Yeo -sre-

MC397 middot 461 1991 4 ICARMA Yeo Sireshy - - r- -

__ T13ST5l 63 1969 4 CAP Yeo Yeobull middot- ~n- D Yeo _ Tl5SSl39 62 1969 4 CAP Yeo Yeo I Yeo

_ STIT711 142 1965 I CAP Yeo I Yeo middot-1shySTl61A Ill 1964 I USFOS Yeo Yeo - D bulloshy

- r-

STISIH 137 1964 I CAP y D y -_ STl31A 140 1951 I CAP Yeo STl34dlU I Yeo

T21Sll 61 1969 4 CAP No rar-middot I Yeo

~-rmiddot-middot SS209A 95 1967 16 CAP Yeo D -SSU4R

11 UI A 41 -in D SS 2741AIM CAP bull

_ _ -shy - -shy- -shy

CALIBRATION

bull PHASEIAPPROACH

bull CALIBRATION ISSUES

bull PHASE II WORK SCOPE

bull SOFlWARE AVAILABILITY

bull EFFECT OF ADDITIONAL PLATFORMS

bull WEIGHTING APPROACH

bull EXAMPLE APPLICATION middot

(l=Ofce Mode] Wind

~

fcurrent l -middot

~SW ~leldl4

ultimate resistance force I first

design load Illgt

i _

deflection

-

bull

4

Historical Hurrlcane Dariiage and Survival of Platforms

Components Required for Calibration (Petrauskas 1992)

PRIOR Distribution

PLATFORM EXPERIENCE

BAYESIAN Survivals and Failures

UPDATING

Waves Winds and Currents bull

POSTERIOR Distribution

- A_middotJA ~ Survival Data Failure Data

Result Result

Load Resistance

middotfl

-

Bayesian Updaing middotGeneral Description (Petrauskas 1992)

L Imiddot

WHERE TO INTRODUCE BIAS FACTOR (B

On Resistance (Rmue - B (RPREDICreD I

middotOn Loads (Smue middot - B (SPRE01cTEo

On Safety Margin (RSmue - B (RSPREDICTED

CALIBRATION METHODOLOGY

gt

r

FIELD OBSERVATIONS

Observed Behavior

survivals Damages

Failures

r

bull

CAPACITY ANALYSIS

Predicted Load and Resistance Predicted Failure Modes

Survival Damage Failure cases

RELIABILITY ANALYSIS

- Probability of Failure ~

Ukellhood Failure

BAYESIAN UPDATING

Prior Distnbution of B Combined Likelihood Function

Posterior Dlstnbution of B

BIAS FACTOR

Mean Value cov bull

Represents Modeling Uncertainties

Prior Distribution of B Likelihood Functionmiddot Observing Success case STISIK and Posterior Distribution of B

2 00

uo

160

140

120

e 100 L $ uoIshy

0 60 ~ ~ 040

020

0 00

0 00 0 50

_ I 00

b

I 50 2 00

-shy lk [bl STISIK)

--shy Pnor-f(b)

--shy Pos1enor r(bl ISIK)

Prior or B Mean bull 10 COV bull 030

Posterlor of Bbull Meanbull 136COV 0 16

2 50

Likelihood Function and Posterior Distribution Example Platform STISIK

- -

~

- Prior and Posterior Distributions or B Failure cases (ST177B STlSlH ST130A T21) Damage cases (T23 T25 ST161A) Success cases (STlSlK ST130Q ST134W WD90A MC311 MC397)

400

350

300

2 so 0-

I

bull bull

bull bull gt

- bullbull ~ bull bull I bull

_ bull bull bullbull I bull bull bullII V -

bull bull

I bull ~ bull

bullbull bullI bull bullI bull bull bull

middot -~

bull__ ~ middotI middotmiddot

L 2 00

e ~ I SO

-I 00

- 0 so

000 gt

0 00 020 0 40 060 0 80 I 00 120 140 160 110 200

b -

-

shy

Prior-r

- - - - Posterior fLlll failure cases

bull bull bull bull bull bull bull Posterior flall damage cases

- bull bull - bull middot Postenor flall success cases

Posterior flall success fulure damage cases

-

Posterior of B Mean= 119 COV 0101

Posterior Distribution or Blas Factor (8) bull Different Categories and All Cases Combined

shy

l

PHASE I - KEY CONCLUSIONS

bull GLOBAL CORRECTION FACTOR

bull 15 TO 20 CONSERVATISM MEAN)

bull UNEXPECTED SURVIVALS HAVE GREATEST IMPACT

bull DAMAGED CASES MORE INFORMATIVE

bull CALIBRATION ISSUES PHASE II

bull

EFFECT OF IMPROVED CAPACITY ANALYSIS RECIPE

bull APPLICABILITY TO MULTIPLE FAILURE MODES

bull NO FOUNDATION DAMAGE CASE INCLUDED

bull IMPROVED REPRESENTATION OF FULL ANDREW DISTRIBUTION

-bull NUMBER OF PLATFORMS CONSIDERED

I 1 I

I t ) bull ~

PHASE II WORK SCOPE (CALIBRATION)

bull

bull ESTABLISH MULTIPLE BIAS FACTORS

Use Nine Base Case Platforms

middot - Capacity Analysis for Revised recipe

Foundation Blas Factor from Foundation JIP

bull DEVELOP WEIGHTING APPROACH

bull UPDATE MULTIPLE BIAS FACTORS

bullEffect of Platforms Not In the Nine Base Cases

bull EXAMPLE APPLICATION OF BIAS FACTORS I

bullbull

PHASE I PHASE II

CAPACITYbull ANALYSISRu bull

---------------------1----middot

Jc1 - iJbullraquoJ dbull

FttlS frt[f~l ratdhJJ

-

LIKELIHOOD FUNCTION

DEVELOPMENT

lk Jb Jfall1N) bull Jb)

111 Ill Isurvtvoq bull 1 - bl

-middot -------------------shy ~---------------------middot-middot

r111ir8 Cb1 111 Cbl BAYESIAN

UPDATING

~cove

R1R2 bull middot---------------------~----

Jblbull P(g cOJ

1 bull JbRI (MBSI I bull 1

~middot

llcJllJfalluro) bull P(gcOJ Ill 111 b21fallun) bull P ((g1 CO) U 1112 c OD

middot--------------------shy ----------------------shy bull

r111111r8 Cbl 111 Ill i-1nro1

deg ~1b2)laquottfrr (b1b21-lnfo) re Ill bull 1-11middot b2) bull lodegamp

middot

i1 covamp i2 cov~i

-

shy

--

Likelihood Function vs b All Success Case STISI K STl30Q STl34W WD90A MC311 MC397

I 0000

0 9000

OIOOO

0 7000

~ 0amp000

ta osoOO

~ 0 4000

0 3000

0 2000

0 1000

bull 0 0000

bull

I

middot

middot bull I

Ibull

-sibull ~ -middot-middot-middot-middot -middot-

----7----+-----~ I

Ibull bull bull I

rbull

--- Jk ISIK[blsuccess)

- bull - bull - lk 130Q[blsuccess) bull

lk 134W[blsuccess)

Jk-WD90A(blsuccc~I

- - - lk MC397[blsucccss)

--ltgt-- lk middotMC31 l[blsucccss)

--- Jk [bl 6 out of6succcsHascs)

05 I 5 2 250

b

Likelihood Functionsmiddot Success Cases - Phbullsc L

shy

bullbull bull bull bull bull bull -------------------~

bull bull bullbull bullbull bullbull bullbull bullbull bullbull bull

bull bullbull

bullbullbullbull

bullbull

bullbullbullbullbullbullbull

bullbullbullbullbullbull

bullbullbullbullbullbullbull

z 0

t z ) LL

c 0 0c i ~ i

J

~ sect ~--------------~ e l5

I

I

SOFTWARE

bull PF PROGRAM

bull CALREL

bull PF +CALREL

bull RELACS

bull

gt i

l

SENSITIVITY STUDIES

POSTERIOR DISTRIBUTION OF B

bull EFFECT OF CHANGE IN DISTRIBUTION OF PRIOR OF B

bull EFFECT OF REMOVAL OF ONE PLATFORM FROM EACH CATEGORY

bull EFFECT OF REMOVAL OF THREE DAMAGE CASES

bull EFFECT OF ADDITIONAL 13 SIMILAR PLATFORMS

-I

Prior and Posterior Distributions of B Variation In COV of PriorofB fromOl to04 -- -

4500

-4000

-

- -

3500 --middot - bull bull - bull bull Prior-r ICOV =02

3000

- - - - Prior-r ICOV =03 -

bull -~ e 2soo -

bull bull bull bull bull bull bull Pnor-r ICOV =o 4t frJ ~ 2000 Posterior rlCOV =0 2

bull bull Postenor rlCOV =0 3 bull bull 1500 bull middot bull - shy -middot- Posterior rlCOV =04

middot~ I 000 - bull

-middotmiddot ~ ~middotmiddot M -~

- middot 0 500

wbullbullbull _ I middot~ middotmiddotmiddotmiddotmiddot middotmiddotmiddotmiddotmiddot - -~ ~ __ middot_ _ bull -shy0 000

0 000 0200 0 400 0 600 0 800 IOOO 1200 1400 1600 1800 2000 -

- - b

-

=shy

Posterior Distribution of Bias Factor (B) bull Effect of Variation In COV of Prior of B

bull

bullPrior and Posterior Distributions of B Evaluating Effect of Removal of One Platform from Each Category from the Base Case of 13 Platforms

400

~ ~ 350

j IA i

3 00 bull I bull I ---Prior-r I bull l

250 1 I 1

I bull bull ---Posterior r lall 13 casesshy- ~E 2 00 bull bull bull bull bull bull bull Posterior r without STISIK

- ~ 1 bull - bull I - bull bull - bull bull Posterior r Iwithout T2I I 50 I

J-tt~i - bull - bull - Posterior r Iwithout T2S

tOO [-

I I ~ ~ I 0 50

0 00

___ r J i

l~T

~--

0 00 0 20 0 40 0 60 080 I 00 120 140 160 110 2 00

b

zshy

Posterior Distribution of Bias Factor (8) bull Effect of Removal of middotPlatform from Any Category

bullPrior and Posterior Distributions or B Evaluating Effect or Removal or 3 Damage Cases from the Base Case or 13 Platrorms

4

35

3

25

e L 2 ~ bulle shy

I 15

bull I

05

0

0 02 04 06 I 2 I 4 16 IS 2

I I

I

I

I Ibull

J -tmiddot L--

bullI

~

I

fl

II ~

I I v I

middot bull bull

bull L

~ ll- I-~lt

OS I

b

I

bull

1middot bullbullbullbull

~ __

bull

- - - - Prior-r

- - - - Postenor flall failure cases

Postenor flall success cases

--- Posterior flall 13 cases

---Posterior fI success failure cases only

IPosterior or B Mean =t11 COV =0121

-

_I

Posterir Distribution or Bias Factor (B) bull Effect or Removal or 3-Damage Cases

-

bullPrior and Posterior Distributions or B Evaluating Effect or Additional 13 Similar Observations as

Used in Andrew JIP The POsterior or B from Andrew-JIP Considered as New Prior or B

- 500

450

400

350

bull bull

bull bull bull bull J bull

bull bull f ~ I

I ~ Ibull bull

bullbull bull

l bull I -bullbull bull bull

I

I bull bull

bull I I~~bull bull bull

bull ~

I bull bull

~ ~_ ~~~

- - New Prior r bull

Posterior 114 new failure cases ~ 300- bull bull bull bull bull bull bull Posterior f 13 new damage casesl 150 ~

- bull bull - bull bull Posterior 116 new success cases i 2 00

Postenor 1113 new success failure -150 damage cases

I 00

0 50 -

0 00 0 00 0 20 0 40 0 60 0 80 I 00 120 I 40 160 110 2 00

Posterior ol B Mean= 120 COV= 0071 b

shy

shy

_rshy

Posterior Distribution of Bias Factor (B) bull Effect of Additional 13-Similar Observations

1

1 I

WEIGHTING APPROACH bull

bull ASSESSMENT OF PLATFORMS DATA

Select Platforms Most Mappable to the Analyzed Platforms

Unlikely SurvivalDamage Cases Identification

Based on Cut-Off Criteria Agemiddot Hs etc

bull INFORMATION REQUIRED FOR COMPARISON

bull CATEGORIZATION OF ADDITIONALmiddot PLATFORMS

bull UPDATING OF CALIBRATION RESULTS

--

OVERVIEW OF PLATFORMS SURVIVEDDAMAGED -~

Operator -

- Platforms

Survived CS)

Platforms

Damaged CD)

Platforms Used In

Phase IJIP -

JI~ ~Bill~IRBDll -

Amoco -

Chevron -

Exxon

Mob II

Phllllps

Shell

Unocal -

30

143

72 22

7

39 25shy -

1

12

bull

1 bull

1

1(S)+1 (D)

3 (S) +3(D)

2 (S)

bull shy

bull

bull

bull

Total Partlclnants Platforms 338 - 15 10C6S+4D) -

Other Operators

310

13

3 (D)

Total Platforms 648 28 13C6S+7D) -

Summary of Andrew JIP bull Phase I Platforms

Plairshy w v- N-ber Lea-le 1 shy car Hind_ o-w N-bull Deplh fl fl IJpe la bull n Yeadon Jadtet lbelledtf liefalit -- _ -OD

IL H - -shy _ICllshy IL IL - - rQClto ~ _

K STUIK 137 1963 8 365 608S4 Cellardeck

-middot STl-shy 170 1964 4 -~ K 39 62269 Cellardeck - --ST134W 137 1911 4 43 608S4 Smvlnlshy- _ - -

WD90A 184 1964 8 K 267Sbull34 83 SOS31 SUboftllar deck SmvMI

MC311 343 1978 8 x S7 62 - MC397 468 1991 4 x shy S33bull6SO 65911 -shySmeSmvlvol -- gt -

r-- T23ST52 63 1969 4 ton- shyKshy 52594 S03 -T25SSl39 62 1969 4 Kshy SOS94 S063 - middot ton-shy__ -

bullbull a ton--shyST161A 118 1964 8 K - 347S 51495 Cellarclecti

r~

ST1771 142 1965 8 K 39 60167 CeDordeck failme

STISIH 137 1964 8 K 395 608S4 CeDardeck Failme

K ST130A 140 1958 a SS 6095 I lblv Pai1uno - T21ST72 61 1969 4 Kshy S4S94 4974 I lhlv Pai1uno -

- -

JE

INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON

bull WATER DEPTH

bull bull YEAR OF INSTALLATIONDESIGN

bull PLATFORM ORIENTATION

bull PLATFORM LATITUDE AND LONGITUDE

bull KEY SKETCHES OF PLATFORM STRUCTURAL FRAMING

Number of Legs Plies Leg Batter Number of Conductors bull Deck Elevations (Including sub-cellar deck If any) Jacket Framing Elevations Brace type In Vertical Frames Sizes of Primary Members (leg pile vertical frames braces primary horizontals) Joint Details (can gap etc) LegPiie Annulus Grouting Status

bull middotPOST ANDREW INSPECTION REPORT Significant Wave Height Hs (ft)

bull INFORMATION ON DAMAGES PRIOR TO ANDREW

bull RESULTS FROM IN-HOUSE STUDIES IF AVAILABLE

bull ~RIGINAL DESIGN CRITERIA INFORMATION IF AVAILABLE

bull SOIL SHEAR STRENGTH PROFILE IF AVAILABLE

Platforms Identified for Evaluating Therr Likelihood of Falling Under Unexpected Survival Category

OPlRATOR

middotADA

BLOClt

llmucnJllE

NAME

YEAll

INSTAUJD

WA1ER

DBPTll

DISTANCE

JlllOM SHOllE

(MILU)

Bl

(M)

AMOCO 111113

STl11

WDI075

A

bull D

St

17

120

1n

44

20

10 I

CllllVllQfl STlllO c 61 110 21 11 STIUO D Q 1110 21 11

STl134 p Q 137 29 11

ST0151 G 61 137 11

ST0151 I 3 121 11

ST0151 J Q 140 11

STl151 It 137 11

ST0151 PJl001 Q UT 11

ST0151 PROD-2 Q UT 11

ST017 A 3 140 35 11

WDIUT c 65 214 34 10

llXXON STI055 II 51 6T 14 I

STl15 A t3 34

WDI073

WD0073

Amiddotshy Q

1111 1111

22

22

I

I

WDI073 c 1n 22 I

- WDOCl93-shy II

65 1110 2T A 3 194 31 MOBU 111121

SS IOl7

A B

501

50

50

31

T

I

T 6

MtlllPllY SS114 B T 50 T T

PlllllIPS SS0149 A 5 55 33 T

SAMllDAN STln A 107 T T

SBllLL WDl103 A 65 223 2T WDl103 B 65 221 2T WD0104 c 65 221 2T WD0104 D 6T 20 2T

UNOCAL SS020I

SS 020I SSl20I

SS 0209

SS0209

II p

PURll

A

B

3

n 61

103

97

97

95

100

34

34

34

34 34

bullbull bullbullbull

- SS0209 D u 95 34 I

- SS 1209 SSl253

SS0253

G A

c

6T

2

100

165

1T5

34

54

54

bullbull I

I I

I I

SOUTH MARCH

~LAND

N

SOUTH PaTOI ENE ISWltD AREA

MMS Defined Limits ror Offshore Structures Requiring Inspections Following Andrew

t---Jbulln3 T25bull I bullTtl

bull1$1Htl1Abull bull1SIK1779bull

~

middot-DaTA AREA~T

CRANO rsLE

AREA bull1300

bullll4W

=shy

bullPath of Hurricane Andrew with Platforms Used In Calibration

Likelihood Function vs b All Success Case STISI K STIJOQ ST134W WD90A MCJll MCJ97

I 0000

0 9000

0 8000

07000

-Bo6000

tiiii 0 5000 -

04000

0 3000

0 2000

01000

0 0000

0 05 25

ri1rt middot-middot-middot-middot-middot-

I 5 2

b

---11 middot151K(blsu~ccsI

- bull - bull - fk 130Q(bfU~lCSS)

lie bull134W(blsuccc]

lk-WD90A[blsultCS]

- - - lk middotMC397(bfUtlCS]

--o-- lk-MCJll(bbuctcsJ

--- ll (bl 6 out of 6 Mlttcs case]

_

( Likelihood Functions middotSuccess Cases - p - 1 c I

CATEGORIZATION OF ADDITIONAL PLATFORMS

I I I

bull CATEGORIES

Unexpected Survivals

Expected Survivals

bull APPROACH

Identify Corresponding Platform(s) Evaluated In the JIP

Determine Max Wave Height at Platform Location

Evaluate Whether Wave Hit the Deck

middot - Evaluate Likelihood of R and S to Differ from the Base Cases

Review Joint Information

Assess Special Features

bull IDENTIFY NUMBER OF PLATFORMS IN EACH CATEGORY

I

UPDATING OF CALIBRATION RESULTS

bull INCLUDE EFFECT OF ADDITIONAL PLATFORMS ON bullBbull

Consider Increased 1Uncertalntles In R and S

Develop bullNeW Ukellhood Functions for Different Categories

middotLet bullxbull Number of Ukely Unexpected Survival Cases and y Number of Ukely Expected Survival Cases

Determine Combined Likelihood Function lk (b Iadditional platforms) middot

bull REVISED LIKELIHOOD FUNCTION

lk(b I all cases) = lk(b I 9 platforms) middot x lk(b Ix platforms) x lk(b I y platforms)

bull ESTABLISH REVISED BIAS FACTOR B

f8 (b) u f8 (b) lk (b I all cases)

II

t t ~

shy

APPLICATIONS OF BIAS FACTORS

ii APPLICABLE TO PARTICULAR LOAD AND RESISTANCE RECIPE

bull USE BY THE API COMMITTEES AND MMS

Criteria Revisions

Recommended Design Guidelines Revisions bull

bullbull REASSESSMENT STUDIES AND DECISION MAKING

middot

bull RISK ASSESSMENT STUDIES BY THE OPERATORS

Individual Platforms

Fleet of Platforms

middot IMR Program Implementation

bullbull NOT TO BE USED IN DESIGN

I I fl J

EXAMPLE USE OF BIAS FACTORS

bull PHASE I EXAMPLE APPLICATION

bull EVALUATION OF ANNUAL PROBABILITY OF FAILURE

Establish Annual Maximum Seastate Data

Determine Coefficients to Define Base Shear

Determine Ultimate Capacity of Platform

Determine Probablllty of Failure Versus bullbbull

-bull DETERMINE UPDATED PROBABILITY OF FAILURE

Pf - P [S gt R]

Pf I P [~ gt Rb I b] f 8 (b) db

Pr - J P [S gt Rb I b] f bull 8 (b) db

bull TOTAL PROBABILITIES OF FAILURE

Total Pf - 1 bull Il [1 middotPf] - l Pf bull

Total Pf - 1 bull Il [1middot Pf] - l Pfmiddot

Total Pr = 1 - n [1middot PfbullJ = l Pfshy

I I I

EXAMPLE APPLICATION bull PHASE I

PLATFORM ST 151 K

bull GULF OF MEXICO

bull WATER DEPTH = 137 FT

bull ANNUAL MAXIMUM HS MEDIAN 175 ft COV 031

bull ASSUMED STORM DURATION 3 hoursmiddot

bull ASSUMED NUMBER OF WAVES 800

bull MEDIAN HS FOR STORM HOURS 1 2 AND 3

bull MEDIAN CURRENT VALUE bull

bull DETERMINE ULTIMATE CAPACITY FOR THREE DIRECTIONS

bull DETERMINE ANNUAL FAILURE PROBABILITIES

Broadsidemiddot Diagonal End-on

0037 0025 0007

0071 0044 0018

0022 0016 0004

DETERMINE TOTAL ANNUAL PROBABILITY OF FAILURE

Pf - 007

Pf - 013

Pr - 004

bull

EXAMPLE USE OF BIAS FACTORS - PHASE II middot

bull DETERMINE PREDICTIVE PROBABILITY OF FAILURE

Case - 1

Pf _ S S Pf (b b2) fa1a2 (b b2) db db2

g bull -1 (b1 R I S) - 1

92 - (b2 R2 I S) - 1

Case -2

Pf s ~ [bJ f11 (b) db

Pf [bJ p [g lt 0)

g - bull 1 (b1 R1 I S) - 1

1

bull1) l

I

BENEFITS OF ANDREW JIP bull PHASE II

lt

bull IMPROVED LOAD AND RESISTANCE RECIPE

bull IMPROVED CALIBRATION PROCEDURE

bull MULTIPLE BIAS FACTORS

bull PROCEDURE TO INCLUDE EFFECT OF OTHER PLATFORMS

bull PROCEDURE TO USE MULTIPLE BIAS FACTORS

  • Structure Bookmarks
    • ANDREW PHASE II JOINT1 INDUSTRY PROJECT
    • PROGRESS MEETING NUMBER 1
    • ANDREW PHASE II JOINT INDUSTRY PROJECT PROGRESS MEETING NUMBER 1 NOVEMBER 16 AGENDA
    • ANDREW PHASE II UST OF PARTICIPANTS
    • ANDREW PHASE I OBJECTIVES AND RESULTS
    • ANDREW PHASE II PROJECT OBJECTIVES
    • SUBSEQUENT PROJECTS
    • ANDREW PHASE II PROJECT TEAM
    • ANDREW PHASE II UPDATE MODELING middotAND ANALYSIS RECIPE
    • -ANDREW PHASE II CONDUCTOR RESISTANCE
    • ANDREW PHASE Il WA VE FORCE PROFILE
    • ANDREW PHASE II WA VE FORCE PROFILEmiddot
    • ANDREW PHASE II BRACE MODELING
    • ANDREW PHASE II FOUNDATIONS
    • ANDREW PHASE II middot middot JOINT MODELING
    • ANDREW PHASE Il JOINT MODELING OPTIONS DISCUSSION
    • ANDREW PHASE II JOINT MODEJJNG OPTIONS ILLUSTRATIONS
    • ANDREW PHASE II JOINT MODELING PHASE II METHOD
    • ANDREW PHASE II ANDREW STORM CONDmONS UPDATED
    • Platform Selection bullbullbull1111111111111
    • BasisCriteria for Selection bullbullbull1111111111111
    • CALIBRATION
    • WHERE TO INTRODUCE BIAS FACTOR (B
    • PHASE I -KEY CONCLUSIONS
    • bull CALIBRATION ISSUES PHASE II
    • PHASE II WORK SCOPE (CALIBRATION)
    • SOFTWARE
    • SENSITIVITY STUDIES POSTERIOR DISTRIBUTION OF B
    • WEIGHTING APPROACH bull
    • INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON
    • CATEGORIZATION OF ADDITIONAL PLATFORMS
    • UPDATING OF CALIBRATION RESULTS
    • APPLICATIONS OF BIAS FACTORS
    • EXAMPLE USE OF BIAS FACTORS
    • EXAMPLE APPLICATION bull PHASE I PLATFORM ST 151 K
    • EXAMPLE USE OF BIAS FACTORS -PHASE II middot
    • BENEFITS OF ANDREW JIP bull PHASE II
Page 2:  · . ' ANDREW PHASE II PROJECT OBJECTIVES . ' • ·UPDATE MODELING AND ANALYSIS RECIPE· JOINTS ' ' ·BRACES WAVE PROFILE FOUNDATIONS . ' ' CONDUCTORS· • ANALVZE PLATFORMS USING

bull

ANDREW PHASE II JOINT INDUSTRY PROJECT PROGRESS MEETING NUMBER 1

NOVEMBER 16

AGENDA

INTRODUCTION amp OVERVIEW OF PROJECT (R L SMITH) 900

UPDATE RECIPE (RW LITTON) 930

BREAK 1030

I

MSL PRESENTATION ON JOINTS amp CONTINUE UPDATE RECIPE 1045

ANDREW STORM CONDITIONS UPDATED (RW LITTON) 1145

LUNCH 1200

PLATFORM SELECTION (DK DOLAN) 100

BAYESIAN CALIBRATION (RAJIVAGGARWAL) 200

DISCUSSION(R L SMITH) 330

REVISIT WORK PLAN (R L SMITH) 400

middotADJOURN 430

I ~

1 It

I t I I

ANDREW PHASE II

UST OF PARTICIPANTS

bull

AMOCO PRODUCTION COMPANY

bull

CHEVRON PETROLEUM TECHNOLOGY COMPANY

bull

EXXON PRODUCTION RESEARCH COMPANY

bull MINERALS MANAGEMENT SERVICES

bullmiddot

MOBIL RESEARCH AND DEVELOPMENT CORPORATION

I bull PHILLIPS PETROLEUM COMPANY

bull bullSHELL OIL COMPANY

bull UNOCAL CORPORATION

i I I

j

ANDREW PHASE I

OBJECTIVES AND RESULTS

bull DEVELOPED A DATABASE OF PLATFORMS AFFECTED BY ANDREW

bull DEVELOPED AN ULTIMATE CAPACITY ANALYSIS RECIPE amp

PROCEDURE

bull IDENTIFIED DEFICIENCIES IN ULTIMATE CAPACITY ANALYSIS PROCESS

bull DEVELOPED A BAYESIAN UPDATING PROCEDURE

l

bull ESTABLISHED A GLOBAL BIAS FACTOR FOR PLATFORMS

bull PROVIDED EXAMPLES FOR THE USE OF THE BIAS FACTOR

ANDREW PHASE II

PROJECT OBJECTIVES

bull middotUPDATE MODELING AND ANALYSIS RECIPEmiddot

JOINTS

middotBRACES

WAVE PROFILE

FOUNDATIONS

CONDUCTORSmiddot

bull ANALVZE PLATFORMS USING UPDATED RECIPE

PLATFORMS SELECTED FROM FAMILY middotOF EXISTING STRUCTURES USED IN PHASE I

POSSIBLE middotREPLACEMENT IF MORE MEANINGFUL PLATFORM(S) middotFOUND

bull EXPAND BAYESIAN CALIBRATION

TWO BIAS FACTORSmiddot DEVELOPED JACKET AND FOUNDATION

UTILIZE RESULTS OF THE APlMMS FOUNDATION PROJECT

PRODUCE EXAMPLE USING EXPANDED PROCEDURE

-middot

bullbull

SUBSEQUENT PROJECTS

bull MMS INSPECTION PROJECT

bull APlMMS CAISSON PROJECT

bull APlMMS FOUNDATION PROJECT

bull TRIALSBENCHMARK JIP

I I It

ANDREW PHASE II

PROJECT TEAM

bull R L SMITH PROJECT MANAGER

bullbull F middot

J PUSKAR PROJECT ADVISOR

bull Rw LITTON TECHNICAL MANAGER

bull RAGGARWAL TECHNICAL DEVELOPER

bull D K DOLAN TECHNICAL ADVISOR middot

bull C A CORNELL RELIABILITY CONSULTANT

bull F MOSES RELIABILITY CONSULTANT

bull middotMSL ENGINEERING LTD TUBULAR JOINT CONSULTANT

q gtI

ANDREW PHASE II

UPDATE MODELING middotAND ANALYSIS RECIPE

bull CONDUCTOR RESISTANCE

bull WA VE FORCE PROFILE

bull BRACE MODELING

middotbull FOUNDATIONSmiddot

bull JOINT MODELING

~ I I

-ANDREW PHASE II

CONDUCTOR RESISTANCE

shybull RESISTANCE NOT INCLUDED IN PHASE I

bull CONTRIBUTION TO OVERALL RESISTANCE A PHASE II

STUDY ISSUE

middotbull middotCOMPARE FOR 4-LEG AND 8-LEG PLATFORM

bull CONSIDER CONDUCTOR POSmON IN GUIDES

- 13 LEFT 13 CENTER 13 RIGHT

Project ST151K Model Conductor version 1 Mon Nov 14 143140 1994

- ---- ---

--

117 s

111

7i

31

0 -5 -~ - ~()

- iltgt --lo

-(pO

-so

bull bullf OC)

-120

-ISo

-ISO

z

CAgt ~x Singl_e Conductor Model

bull

=I=

-_

-_

-gtshy

~ -

~

I

I~

- gt

~-----------____________________

Project STlSlK Model dir_2 version 7 Mon Nov 14 124200 1994

177 s

Ill

74

37

0- - e -zo - ~o

- Jo

-100

-10

-ISO

z

UV 1-+x

Project STlSlX Model Conductor Version 1 Mon Nov 14 140620 1994

bullmiddot

I

I

I I

I

I I

I I I I I

1 ==

-L

-L - L

-

- L

-

-

L

-

Uniform Displacement step 6 Inelastic Events Legend

Elastic

Strut Buckling Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield Beam Clmn FUlly Plastic Fracture

--

i~~~~~~~~~~~---~--~~-J___--JJ~_middot~~~=-

Project ST151K Model Conductor Version1 Mon Nov 14 142735 1994

---

~

I

I

middot

I

I

I

- shy- shy

- shy

- shy

4x z

CIV Uniform Displacement step 33 Inelastic Events Legend

bull Elasti~ Strut Buckling Strut Residual Strut Reloading Plastic StrutNLTruss Beam Clmn Initial Yield Beam Clmn-Fully Plastic Fracture

gt bull

Project STlSlK middot Model Conductor Version 2 Mon Nov 14 142025 1994

-----------------

bullf

I

I

I

II I I

I I I I

I ==

-

Variable Displacement step 9 Inelastic Events Legend

Elastic Strut Residual

bull

Strut Buckling Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield Beam Clmn Fully Plastic Fracture

--

Project ST151K Model Conductor Version 2 Mon Nov 14 142145 1994

I I I I

I I

I

I

I - - -

-Ishy-

-

Variable Displacement step 35 InelasticEvents Legend

Elastic Strut Buckling Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield Beam Clmn Fully Plastic Fracture

Project ST151K Model Conductormiddot

Version 1

Clgt - Cut Plane Force Fx Mon Nov 14 140051 1994

- -_ vii I fOIZM

0 _

~

N ID 0 bull 0

x 0

middotrt ~ () bull u 0 ~ 0 Ir

N bull

0

0 1 2 3 4 5 X Deck Disp [Ft)

I I 1 1 WbullTI- bullbull

lltcxi lr-C

011lt~~~~~~~---~~~~~~~~L~~~~~~~--~~~~~~~~i~~~~~~~--

Uniform Node Displacement

shy

Ji

r

Comparison of Pushover Curves for ST 151K Variable Load Pattern and Monotonic Load Pattern (Andrew)

Soll Improved 100

6000

5000 -i 4000

f

llS

3000 Ul

J 2000

10 c11 c1lgtolts 1000

0 v

bull ~

f

_ ----- --- ---shy

bull --Variable Lold

- - Andrew Load

I I I I I I

4 5 62 30 1

Deck Dlsplacement (ft)

bullbull r 1 II

~

ANDREW PHASE Il

WA VE FORCE PROFILE

bull DECK WA VE FORCE ~

PHASE I USED SECTION 17 (1056H degCd)

PHASE Il USE REVISED SECTION 17 (1000H Cd)

bull PHASE I USED SINGLE FORCE PROFILE FOR PUSHOVER

- USED WAVE HEIGHT CLOSE TO ULTIMATE CAPACITY

bull PHASE J

Il WILL STUDY THE

USE OF VARIABLE

PROFILE ~

- IF FORCE LEVEL INCREASES (OR DECREASES) BY A FACTOR EQUIVALENT TO A 2 FOOT CHANGE IN WA VE HEIGHT THE INCREMENTAL FO~CE PA1IERN IS CHANGED ACCORDINGLY (CAP FEATURE)

- INITIAL EXAMPLE

f~

a

middotn

a

I N -a a a amiddot

+

~ a

l

a w gt 0c 0 gt a I shyc -CJ wc w gtc(

== c( I shyZ w E W a 0 z- 8

ii sect

~ ~ bullT 0

I if

x ~

bullbullbull

II ~------------________

ANDREW PHASE II

WA VE FORCE PROFILEmiddot

CAP AUTOMATIC APPLICATION FEATURE

STAT

fR1 0 lt f lt1

STAT

STAT

f(R3-R2) middot f lt1

Project ST151K Model Wave_Loading Version 1 rue Nov 8 164107 1994 PT

E~ nss FT

SL 14amp FT

a 111 FT

EL 14 li

E-1 37 FT

El o FT

1middot1middot ______________ _ ________ __ _ __

0

200 400 600 800 1000 12000 Force at Elevatloii (kip)

200

a ll_

Wave Force at Major Elevations for Discrete Wave Heights

180

bullHmiddot40ft

Hmiddot495ft

H bull 51 ft

Hmiddot55ft_

H-59ft I

Andrew (H bull 60 86 ft) II

i I

_11---__

- middot- Hmiddot30ft

-

- bull -

middot bull bull bull middot

---Hmiddot53ft

bullbullbullbullHmiddot57ft

-----H-63ft

Elevation Above Mudllne 100

(ft)

180

140

120

80

60

40

20

Loading at Major Elevations

---H-49511

-deg-Hmiddot608611

0 200 400 600 800 1000 1200

Force It Elevatlon (kip)

--

Base Shear vs Wave Height ST151K

6000

5000 1---------Ult1mateCapacitybull 4890 kip (100 Improved Soll)-----------~--

4000a a lii ~ 3000

81 ampI

2000

1000

0 +-~~~--i~~~~~~~~~-+-~~~--lf--~~~~~~~~-+-~~~--l

30 35 40 45 50 55 60 65

Wave Height (ft)

bull middot

~

bull

II

2

g

~

It)

0

cCD E 0 E a c E euro~ CDshygtItgtmiddoto ~ti c -a Gic CD gt ~

-= c -Cll

I z

bullO 0i i ~ g 0

() CI -bull(q-ll) iuawow llu1wniiaAO

11

shy

116

114

112

g 110

bullc = 108

~ ~ 106 -sect 104

102

100

98

96

-

6555 60

Wave Height vs Center of Action ST151K

30 35 40 45 50

Wave Height (ft)

t

Comparison of Pushover Curves for ST 151K Variable Load Pattern and Monotonic Load Pattern (Andrew)

Soll Improved 100

--- --------- --- ----6000

5000

- 4000 8 - I

1 3000 Ul

81 m 2000

_J -shy

1000

---Variable Load

- - Andrew Load

0 f-~~~~~r--~~~~-t-~~~~~-t-~~~~~+-~~~~----t~~~~~-1

0 1 - 2 3 4 5 6

Deck Displacement (ft)

-middotshy

1 t I I ~

ANDREW PHASE II

BRACE MODELING

bull MARSHALL STRUTS USED FORPHASE I

bull PHASE II WILL ASSESS DIFFERENCE IN USING BUCKLING BEAMS AND MARSHALL STRUTS

middot bull COMPARE FOR 2D FRAME

bull COMPARE FOR PLATFORM 177 WHERE LOWER

BAY EXHIBITS LARGE BRACE MOMENTS

Project andrew Model beamframe Version 1 Mon Nov 14 145447 1994

~

middot Frame Test buckling beam-column model

-

Project andrew Model strutframe Version 1 Mon Nov 14 145754 1994

bull

Frame Test strut model

middot-middot _____________--_

_middotmiddotr_~ --ll--~-middot__

Buckling Beam bull

Strut (K=O 65)

Lateral Displacement

ANDREW PHASE II

FOUNDATIONS

bull ~LICIT NONLINEAR API FOUNDATION USED INiPHASE I

bull FOUNDATIONS

JIP SELECTED MINI-VANE DATA FOR

STRENGTH PROFILE

bull COMMENTS ON BENCHMARK COMPARISONS

- SIGNIFICANT VARIABILITY DUE TO FOUNDATION MODELING METHODS

II Ii

z

ctV 4x SPlO caissonfy-42s soil-api(lf-097step-114)

Inelastic Events Legend Elastic Strut Buckling

Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield

--------shyBeam Clmn Fully Plastic Fracture

~--~~~~~~~~~--~----~__i~~---~~~~__--~

Pro)ect apipile Model South Pelto 10 Version Wed May 18072235 1994 - -

1bullt

+

t

f i bullbullbull

bullbull bullbull

I I

5

~x SPlO Caissonfy=42s soil=api(lf=lOOstep=137)

Inelastic Events Legend

Elastic Strut Bucklilg

Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield

earn Clmn Full~ Plas1c Fracture

ANDREW PHASE II middot middot

JOINT MODELING

middotbull API RIGID-PLASTIC USED IN PHASE I

a ELASTIC-BRITTLE USED IN PHASE IB

amiddot MSL WILL PROVIDE NONLINEAR FORCE-DEFORMATION

AND MOMENT-DEFORMATION RELATIONS FOR PHASE II

a MODELING OPTIONS FOR PHASE II -

l

- UNCO~LEDmiddotONE DIMENSIONAL MEMBER END JOINT MODELING

J Ii

ANDREW PHASE Il

JOINT MODELING OPTIONS DISCUSSION

bull JD COUPLED INTERACTION SURFACE

- IDEAL

- DATA NOT AVAILABLE TO CALIBRATE

bull 2D COUPLED INTERACTION SURFACE

- SIGNIFICANT MODELING IMPROVEMENT

- DATAAVAILABLE

- IMPLEMENTATION PRACTICAL BUT NEEDS DEVELOPMENT AND CALIBRATION

bull 2D UNCOUPLED INTERACTION SURFACE

- SIGNIFICANT MODELING IMPROVEMENT

- DATA AVAILABLE

- IMPLEMENTATION PRACTICAL AND AVAILABLE

ANDREW PHASE II

JOINT MODEJJNG OPTIONS ILLUSTRATIONS

P

e

ANDREW PHASE II

JOINT MODELING PHASE II METHOD

e

I I I

ANDREW PHASE II

ANDREW STORM CONDmONS UPDATED

bull -PHASE I USED OCEANWEATHER HINDCAST-(Nov 1992)

bull middotmiddotPHASE II WILL USE THE REVISED HINDCAST (Nov 1994)

bull lt c lt I lt

2-2

l---4-~~~+-~~-+-~~~1--~~-+-~~--l--~T--+-~~--i24

f-tJVtMfiz~ 1Cfl (r STvb(

Plollol bullbull 5middotNOYmiddot9214bull1J Irbullbull Illbull [ANDREVl~NDREVCUSTbullMAXll 5-NOY-1992 10bull54 - 9208

HEIGHT lml

I~ MAXI

I L 1~middotamp SZ~- ~A I 13c

I J) ( r I X~0iWJ~0gtJ gt L gt gt I 1c I 120

-- N

~ deg

I I I I I I e C I ~=-41 122 -94 -92 -90 -88 -86 -84 -82 -80

-

Figure 7 contours of maximum hindcast significant wave height

_

~

~tl ~

NOltNttIJ 17lt ltNS S iJJ)y

Maximum Significant Wave Height - Hurricane Andrew 1992

30 ~

26 ~ I ~-~~~- == -shy -~ 35-_

24deg1--shy --------~- middot-middot-------middotmiddotmiddot----middot------middot-middot~middot- - - ~-----1---middot

22middot ~~~~~~~-~ -94 -92 -90 -88 -84deg-86 -82 -somiddot

c

-

Platform Selection bullbullbull1111111111111

0 BasisCriteria for Selection middot

0 Review of Phase I Platforms middot 0 Alternative Platforms

0 PMB Recommendations

0 Participants Comments

gt

L

l I

BasisCriteria for Selection bullbullbull1111111111111

0 Benefit for Calibration - Classification (UnexpectedExpected)

- Failure Mode (StructuresFoundations)

- Damaged StructuresKnown response

- Better Representation of Full ~opulation

0 Availability of Data - Existing Models

- Inspection Data

- Damage Data

PREDICTED BEHAVIOR

-

OBSERVED

BEHAVIOR

-SURVIVED DAMAGED FAILED

SURVIVED

ST134W WD90A

-MC311 -

MC397 -

--

-

-

ST151K ST130Q

-

-

-

DAMAGED - -

-

-

--

-

-

ST52 SS139 ST177B -

ST161A

-_

-

FAILED

-

-

-

ST151H ST130A ST72

Operator - Platfonns - Platfonns Survived Damagedamp

- Falled -

bull

-

-

Jle eBDIGIPllDbl -Amocoshy 30 1 Chevron -

143 12 Exxon 72 Mobll 22 1 Phllllps 7 bull

-Shell 39 Unocal 25 1

Total Partlclnts Platfonns 338 15

Other Operators bull- 310 13

-

Total Platfonns - -

648 28

fbml

Platforms Considered 6 7

tie of Total Platfonns 1 25

OVERVIEW OF PLATFORMS SURVIVEDDAMAGEDFAILED

-

shy

v

Platforms Evaluated in Phase I

-

-Platfonn Water Year Number Analysis Avallabillty of Availability Platform

- Name Depth rl of Model DetaDed rl ClassUlcatlon

Installation legs Inspection Soll Report Based on Information Phase I Calibration - -ft

- -- Suaill flalfnrm ~BSi -STISIK 137 1963 8 CAP Yes Yes Un eel Sarvlval

-STIJOO 170 1964 4 CAPmiddotshy Yes ST134data u ST134W 137 1981 4 CAP Yes Yesmiddot E

-

edSarvlval -

middotsurvival

WD90A 184 - 1964 8 CAP Yes E -

- Sarvlvill

MC3ll 343 1978 8 KARMA

MC397 468 1991 4 KARMA -

bull ~reglmmiddot- -

Yes Sure Survival

bull Yes - Sare Sanlval-

-T23STS2 63 1969 4 CAP

T2SSS139 62 1969 4 CAP -

Yes

Yes

Yes Ex

Yes Une

bull SurvivalLikely to DamalI

middot bull SanlvalLlkely to DamalI

ST161A 118 1964 8 USFOS

Eail11a pound1al(bullllm Clss ST177B 142 --1965 8 CAP

STISIH 137 1964 - 8 CAP C

STIJOA 140 19S8 8 CAP - -

Yes

Yes

ollamed

Yes

- Yes E

-

at about 1 mUe

Yes

ST134data

bull SafvlvalLlkelv to Dam- bull- -

E middotrahre

Emeried Failure -

Likely FaUare

T21ST72 61 1969 4 CAP Collamed No Likely FaUure

Platforms Identified for Evaluating Their Likelihood of Falling Under UnexpeCted Survival Category

CllEllATOa

AllB

llLOCll

Sl1lvcnJllE

NAME

YEAR

INSTAUJCD

WATER

IJBPJB

middotmiddotDISTANCE

ftOMSllOllE

(MIUIS)

Ill

(M)

AMOCO BllW

STllI

WDI075

A

bull D

4

n uo 17J

44

32

bull 20

10

bull CllEVllOH STtllO c I 1bull 21 11

STl130 D 0 161 21 11 STllJ4 I 0 137 D - 11

STIUI G I 137 32 11

STIUI I 0 121 32 11

STIUI

J 0 140 32 11

STIUI It 4 137 32 11

STllSI PROl)I 0 137 32 11

STllSI PROD-2 0 137 32 11

STll7 A 0

140 35 11

WDl117 c 65 214 34 10

DXON STshy E 51 14 bull STl165 A t3 bull 34 WD0073

WD0073

WD0073

A _ c

4

62

4

IA IA

17J

22

22

22

bullbullbull WD003-shy E 65 161 Z7

A 0 IN 31 MOBD Bl12I

ssn A bull

50

50

31

7

bull 7

MIJllPllY SS114 B 7 50 7 7

PlllUIPS SSl14t A 55 33 7

SAMEDAN ST17J A 4 117 7 bull 7

SJIELL WD0103 A 65 223 27 WDllOJ bull 65 221 27 WDll04 c 65 221 27 WDOI04 D 67 2 27

UNOCAL

SSl20I

SSl20I

bull SSl20I

SS0209

SS0209

SS0209

SS0209

SS 0253

SS 0253

E

I

PtJllE

A

bull D

G A

c

0

4

4

7J

61

0

0

69

103

97

97

9S

100

95

100

165

175

34

34

34

34

34

34

34

54

54

bullbullbullbullbullbullbullbullbull

Platforms Recommended for Phase II

llalr y_ w- Aot Av- _ I 1 Doplla Model nm bull bull _ bull I I H11 s18-1 Irht lbwl~ bullbullbull(-illtdlel OM lllMtD

II ---

-STISIIC 137 1963 I CAP Yeo Yeo I Yeo

STtshy -170 1964 4 CAP Yeo ST134dlU I Yeo

STl34W 137 Yeo Yeo -1981 4 CAP D Yeo

WD90A 114 1964 I CAP Yeo - D

MC311 30 1971 I KARMA Yeo -sre-

MC397 middot 461 1991 4 ICARMA Yeo Sireshy - - r- -

__ T13ST5l 63 1969 4 CAP Yeo Yeobull middot- ~n- D Yeo _ Tl5SSl39 62 1969 4 CAP Yeo Yeo I Yeo

_ STIT711 142 1965 I CAP Yeo I Yeo middot-1shySTl61A Ill 1964 I USFOS Yeo Yeo - D bulloshy

- r-

STISIH 137 1964 I CAP y D y -_ STl31A 140 1951 I CAP Yeo STl34dlU I Yeo

T21Sll 61 1969 4 CAP No rar-middot I Yeo

~-rmiddot-middot SS209A 95 1967 16 CAP Yeo D -SSU4R

11 UI A 41 -in D SS 2741AIM CAP bull

_ _ -shy - -shy- -shy

CALIBRATION

bull PHASEIAPPROACH

bull CALIBRATION ISSUES

bull PHASE II WORK SCOPE

bull SOFlWARE AVAILABILITY

bull EFFECT OF ADDITIONAL PLATFORMS

bull WEIGHTING APPROACH

bull EXAMPLE APPLICATION middot

(l=Ofce Mode] Wind

~

fcurrent l -middot

~SW ~leldl4

ultimate resistance force I first

design load Illgt

i _

deflection

-

bull

4

Historical Hurrlcane Dariiage and Survival of Platforms

Components Required for Calibration (Petrauskas 1992)

PRIOR Distribution

PLATFORM EXPERIENCE

BAYESIAN Survivals and Failures

UPDATING

Waves Winds and Currents bull

POSTERIOR Distribution

- A_middotJA ~ Survival Data Failure Data

Result Result

Load Resistance

middotfl

-

Bayesian Updaing middotGeneral Description (Petrauskas 1992)

L Imiddot

WHERE TO INTRODUCE BIAS FACTOR (B

On Resistance (Rmue - B (RPREDICreD I

middotOn Loads (Smue middot - B (SPRE01cTEo

On Safety Margin (RSmue - B (RSPREDICTED

CALIBRATION METHODOLOGY

gt

r

FIELD OBSERVATIONS

Observed Behavior

survivals Damages

Failures

r

bull

CAPACITY ANALYSIS

Predicted Load and Resistance Predicted Failure Modes

Survival Damage Failure cases

RELIABILITY ANALYSIS

- Probability of Failure ~

Ukellhood Failure

BAYESIAN UPDATING

Prior Distnbution of B Combined Likelihood Function

Posterior Dlstnbution of B

BIAS FACTOR

Mean Value cov bull

Represents Modeling Uncertainties

Prior Distribution of B Likelihood Functionmiddot Observing Success case STISIK and Posterior Distribution of B

2 00

uo

160

140

120

e 100 L $ uoIshy

0 60 ~ ~ 040

020

0 00

0 00 0 50

_ I 00

b

I 50 2 00

-shy lk [bl STISIK)

--shy Pnor-f(b)

--shy Pos1enor r(bl ISIK)

Prior or B Mean bull 10 COV bull 030

Posterlor of Bbull Meanbull 136COV 0 16

2 50

Likelihood Function and Posterior Distribution Example Platform STISIK

- -

~

- Prior and Posterior Distributions or B Failure cases (ST177B STlSlH ST130A T21) Damage cases (T23 T25 ST161A) Success cases (STlSlK ST130Q ST134W WD90A MC311 MC397)

400

350

300

2 so 0-

I

bull bull

bull bull gt

- bullbull ~ bull bull I bull

_ bull bull bullbull I bull bull bullII V -

bull bull

I bull ~ bull

bullbull bullI bull bullI bull bull bull

middot -~

bull__ ~ middotI middotmiddot

L 2 00

e ~ I SO

-I 00

- 0 so

000 gt

0 00 020 0 40 060 0 80 I 00 120 140 160 110 200

b -

-

shy

Prior-r

- - - - Posterior fLlll failure cases

bull bull bull bull bull bull bull Posterior flall damage cases

- bull bull - bull middot Postenor flall success cases

Posterior flall success fulure damage cases

-

Posterior of B Mean= 119 COV 0101

Posterior Distribution or Blas Factor (8) bull Different Categories and All Cases Combined

shy

l

PHASE I - KEY CONCLUSIONS

bull GLOBAL CORRECTION FACTOR

bull 15 TO 20 CONSERVATISM MEAN)

bull UNEXPECTED SURVIVALS HAVE GREATEST IMPACT

bull DAMAGED CASES MORE INFORMATIVE

bull CALIBRATION ISSUES PHASE II

bull

EFFECT OF IMPROVED CAPACITY ANALYSIS RECIPE

bull APPLICABILITY TO MULTIPLE FAILURE MODES

bull NO FOUNDATION DAMAGE CASE INCLUDED

bull IMPROVED REPRESENTATION OF FULL ANDREW DISTRIBUTION

-bull NUMBER OF PLATFORMS CONSIDERED

I 1 I

I t ) bull ~

PHASE II WORK SCOPE (CALIBRATION)

bull

bull ESTABLISH MULTIPLE BIAS FACTORS

Use Nine Base Case Platforms

middot - Capacity Analysis for Revised recipe

Foundation Blas Factor from Foundation JIP

bull DEVELOP WEIGHTING APPROACH

bull UPDATE MULTIPLE BIAS FACTORS

bullEffect of Platforms Not In the Nine Base Cases

bull EXAMPLE APPLICATION OF BIAS FACTORS I

bullbull

PHASE I PHASE II

CAPACITYbull ANALYSISRu bull

---------------------1----middot

Jc1 - iJbullraquoJ dbull

FttlS frt[f~l ratdhJJ

-

LIKELIHOOD FUNCTION

DEVELOPMENT

lk Jb Jfall1N) bull Jb)

111 Ill Isurvtvoq bull 1 - bl

-middot -------------------shy ~---------------------middot-middot

r111ir8 Cb1 111 Cbl BAYESIAN

UPDATING

~cove

R1R2 bull middot---------------------~----

Jblbull P(g cOJ

1 bull JbRI (MBSI I bull 1

~middot

llcJllJfalluro) bull P(gcOJ Ill 111 b21fallun) bull P ((g1 CO) U 1112 c OD

middot--------------------shy ----------------------shy bull

r111111r8 Cbl 111 Ill i-1nro1

deg ~1b2)laquottfrr (b1b21-lnfo) re Ill bull 1-11middot b2) bull lodegamp

middot

i1 covamp i2 cov~i

-

shy

--

Likelihood Function vs b All Success Case STISI K STl30Q STl34W WD90A MC311 MC397

I 0000

0 9000

OIOOO

0 7000

~ 0amp000

ta osoOO

~ 0 4000

0 3000

0 2000

0 1000

bull 0 0000

bull

I

middot

middot bull I

Ibull

-sibull ~ -middot-middot-middot-middot -middot-

----7----+-----~ I

Ibull bull bull I

rbull

--- Jk ISIK[blsuccess)

- bull - bull - lk 130Q[blsuccess) bull

lk 134W[blsuccess)

Jk-WD90A(blsuccc~I

- - - lk MC397[blsucccss)

--ltgt-- lk middotMC31 l[blsucccss)

--- Jk [bl 6 out of6succcsHascs)

05 I 5 2 250

b

Likelihood Functionsmiddot Success Cases - Phbullsc L

shy

bullbull bull bull bull bull bull -------------------~

bull bull bullbull bullbull bullbull bullbull bullbull bullbull bull

bull bullbull

bullbullbullbull

bullbull

bullbullbullbullbullbullbull

bullbullbullbullbullbull

bullbullbullbullbullbullbull

z 0

t z ) LL

c 0 0c i ~ i

J

~ sect ~--------------~ e l5

I

I

SOFTWARE

bull PF PROGRAM

bull CALREL

bull PF +CALREL

bull RELACS

bull

gt i

l

SENSITIVITY STUDIES

POSTERIOR DISTRIBUTION OF B

bull EFFECT OF CHANGE IN DISTRIBUTION OF PRIOR OF B

bull EFFECT OF REMOVAL OF ONE PLATFORM FROM EACH CATEGORY

bull EFFECT OF REMOVAL OF THREE DAMAGE CASES

bull EFFECT OF ADDITIONAL 13 SIMILAR PLATFORMS

-I

Prior and Posterior Distributions of B Variation In COV of PriorofB fromOl to04 -- -

4500

-4000

-

- -

3500 --middot - bull bull - bull bull Prior-r ICOV =02

3000

- - - - Prior-r ICOV =03 -

bull -~ e 2soo -

bull bull bull bull bull bull bull Pnor-r ICOV =o 4t frJ ~ 2000 Posterior rlCOV =0 2

bull bull Postenor rlCOV =0 3 bull bull 1500 bull middot bull - shy -middot- Posterior rlCOV =04

middot~ I 000 - bull

-middotmiddot ~ ~middotmiddot M -~

- middot 0 500

wbullbullbull _ I middot~ middotmiddotmiddotmiddotmiddot middotmiddotmiddotmiddotmiddot - -~ ~ __ middot_ _ bull -shy0 000

0 000 0200 0 400 0 600 0 800 IOOO 1200 1400 1600 1800 2000 -

- - b

-

=shy

Posterior Distribution of Bias Factor (B) bull Effect of Variation In COV of Prior of B

bull

bullPrior and Posterior Distributions of B Evaluating Effect of Removal of One Platform from Each Category from the Base Case of 13 Platforms

400

~ ~ 350

j IA i

3 00 bull I bull I ---Prior-r I bull l

250 1 I 1

I bull bull ---Posterior r lall 13 casesshy- ~E 2 00 bull bull bull bull bull bull bull Posterior r without STISIK

- ~ 1 bull - bull I - bull bull - bull bull Posterior r Iwithout T2I I 50 I

J-tt~i - bull - bull - Posterior r Iwithout T2S

tOO [-

I I ~ ~ I 0 50

0 00

___ r J i

l~T

~--

0 00 0 20 0 40 0 60 080 I 00 120 140 160 110 2 00

b

zshy

Posterior Distribution of Bias Factor (8) bull Effect of Removal of middotPlatform from Any Category

bullPrior and Posterior Distributions or B Evaluating Effect or Removal or 3 Damage Cases from the Base Case or 13 Platrorms

4

35

3

25

e L 2 ~ bulle shy

I 15

bull I

05

0

0 02 04 06 I 2 I 4 16 IS 2

I I

I

I

I Ibull

J -tmiddot L--

bullI

~

I

fl

II ~

I I v I

middot bull bull

bull L

~ ll- I-~lt

OS I

b

I

bull

1middot bullbullbullbull

~ __

bull

- - - - Prior-r

- - - - Postenor flall failure cases

Postenor flall success cases

--- Posterior flall 13 cases

---Posterior fI success failure cases only

IPosterior or B Mean =t11 COV =0121

-

_I

Posterir Distribution or Bias Factor (B) bull Effect or Removal or 3-Damage Cases

-

bullPrior and Posterior Distributions or B Evaluating Effect or Additional 13 Similar Observations as

Used in Andrew JIP The POsterior or B from Andrew-JIP Considered as New Prior or B

- 500

450

400

350

bull bull

bull bull bull bull J bull

bull bull f ~ I

I ~ Ibull bull

bullbull bull

l bull I -bullbull bull bull

I

I bull bull

bull I I~~bull bull bull

bull ~

I bull bull

~ ~_ ~~~

- - New Prior r bull

Posterior 114 new failure cases ~ 300- bull bull bull bull bull bull bull Posterior f 13 new damage casesl 150 ~

- bull bull - bull bull Posterior 116 new success cases i 2 00

Postenor 1113 new success failure -150 damage cases

I 00

0 50 -

0 00 0 00 0 20 0 40 0 60 0 80 I 00 120 I 40 160 110 2 00

Posterior ol B Mean= 120 COV= 0071 b

shy

shy

_rshy

Posterior Distribution of Bias Factor (B) bull Effect of Additional 13-Similar Observations

1

1 I

WEIGHTING APPROACH bull

bull ASSESSMENT OF PLATFORMS DATA

Select Platforms Most Mappable to the Analyzed Platforms

Unlikely SurvivalDamage Cases Identification

Based on Cut-Off Criteria Agemiddot Hs etc

bull INFORMATION REQUIRED FOR COMPARISON

bull CATEGORIZATION OF ADDITIONALmiddot PLATFORMS

bull UPDATING OF CALIBRATION RESULTS

--

OVERVIEW OF PLATFORMS SURVIVEDDAMAGED -~

Operator -

- Platforms

Survived CS)

Platforms

Damaged CD)

Platforms Used In

Phase IJIP -

JI~ ~Bill~IRBDll -

Amoco -

Chevron -

Exxon

Mob II

Phllllps

Shell

Unocal -

30

143

72 22

7

39 25shy -

1

12

bull

1 bull

1

1(S)+1 (D)

3 (S) +3(D)

2 (S)

bull shy

bull

bull

bull

Total Partlclnants Platforms 338 - 15 10C6S+4D) -

Other Operators

310

13

3 (D)

Total Platforms 648 28 13C6S+7D) -

Summary of Andrew JIP bull Phase I Platforms

Plairshy w v- N-ber Lea-le 1 shy car Hind_ o-w N-bull Deplh fl fl IJpe la bull n Yeadon Jadtet lbelledtf liefalit -- _ -OD

IL H - -shy _ICllshy IL IL - - rQClto ~ _

K STUIK 137 1963 8 365 608S4 Cellardeck

-middot STl-shy 170 1964 4 -~ K 39 62269 Cellardeck - --ST134W 137 1911 4 43 608S4 Smvlnlshy- _ - -

WD90A 184 1964 8 K 267Sbull34 83 SOS31 SUboftllar deck SmvMI

MC311 343 1978 8 x S7 62 - MC397 468 1991 4 x shy S33bull6SO 65911 -shySmeSmvlvol -- gt -

r-- T23ST52 63 1969 4 ton- shyKshy 52594 S03 -T25SSl39 62 1969 4 Kshy SOS94 S063 - middot ton-shy__ -

bullbull a ton--shyST161A 118 1964 8 K - 347S 51495 Cellarclecti

r~

ST1771 142 1965 8 K 39 60167 CeDordeck failme

STISIH 137 1964 8 K 395 608S4 CeDardeck Failme

K ST130A 140 1958 a SS 6095 I lblv Pai1uno - T21ST72 61 1969 4 Kshy S4S94 4974 I lhlv Pai1uno -

- -

JE

INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON

bull WATER DEPTH

bull bull YEAR OF INSTALLATIONDESIGN

bull PLATFORM ORIENTATION

bull PLATFORM LATITUDE AND LONGITUDE

bull KEY SKETCHES OF PLATFORM STRUCTURAL FRAMING

Number of Legs Plies Leg Batter Number of Conductors bull Deck Elevations (Including sub-cellar deck If any) Jacket Framing Elevations Brace type In Vertical Frames Sizes of Primary Members (leg pile vertical frames braces primary horizontals) Joint Details (can gap etc) LegPiie Annulus Grouting Status

bull middotPOST ANDREW INSPECTION REPORT Significant Wave Height Hs (ft)

bull INFORMATION ON DAMAGES PRIOR TO ANDREW

bull RESULTS FROM IN-HOUSE STUDIES IF AVAILABLE

bull ~RIGINAL DESIGN CRITERIA INFORMATION IF AVAILABLE

bull SOIL SHEAR STRENGTH PROFILE IF AVAILABLE

Platforms Identified for Evaluating Therr Likelihood of Falling Under Unexpected Survival Category

OPlRATOR

middotADA

BLOClt

llmucnJllE

NAME

YEAll

INSTAUJD

WA1ER

DBPTll

DISTANCE

JlllOM SHOllE

(MILU)

Bl

(M)

AMOCO 111113

STl11

WDI075

A

bull D

St

17

120

1n

44

20

10 I

CllllVllQfl STlllO c 61 110 21 11 STIUO D Q 1110 21 11

STl134 p Q 137 29 11

ST0151 G 61 137 11

ST0151 I 3 121 11

ST0151 J Q 140 11

STl151 It 137 11

ST0151 PJl001 Q UT 11

ST0151 PROD-2 Q UT 11

ST017 A 3 140 35 11

WDIUT c 65 214 34 10

llXXON STI055 II 51 6T 14 I

STl15 A t3 34

WDI073

WD0073

Amiddotshy Q

1111 1111

22

22

I

I

WDI073 c 1n 22 I

- WDOCl93-shy II

65 1110 2T A 3 194 31 MOBU 111121

SS IOl7

A B

501

50

50

31

T

I

T 6

MtlllPllY SS114 B T 50 T T

PlllllIPS SS0149 A 5 55 33 T

SAMllDAN STln A 107 T T

SBllLL WDl103 A 65 223 2T WDl103 B 65 221 2T WD0104 c 65 221 2T WD0104 D 6T 20 2T

UNOCAL SS020I

SS 020I SSl20I

SS 0209

SS0209

II p

PURll

A

B

3

n 61

103

97

97

95

100

34

34

34

34 34

bullbull bullbullbull

- SS0209 D u 95 34 I

- SS 1209 SSl253

SS0253

G A

c

6T

2

100

165

1T5

34

54

54

bullbull I

I I

I I

SOUTH MARCH

~LAND

N

SOUTH PaTOI ENE ISWltD AREA

MMS Defined Limits ror Offshore Structures Requiring Inspections Following Andrew

t---Jbulln3 T25bull I bullTtl

bull1$1Htl1Abull bull1SIK1779bull

~

middot-DaTA AREA~T

CRANO rsLE

AREA bull1300

bullll4W

=shy

bullPath of Hurricane Andrew with Platforms Used In Calibration

Likelihood Function vs b All Success Case STISI K STIJOQ ST134W WD90A MCJll MCJ97

I 0000

0 9000

0 8000

07000

-Bo6000

tiiii 0 5000 -

04000

0 3000

0 2000

01000

0 0000

0 05 25

ri1rt middot-middot-middot-middot-middot-

I 5 2

b

---11 middot151K(blsu~ccsI

- bull - bull - fk 130Q(bfU~lCSS)

lie bull134W(blsuccc]

lk-WD90A[blsultCS]

- - - lk middotMC397(bfUtlCS]

--o-- lk-MCJll(bbuctcsJ

--- ll (bl 6 out of 6 Mlttcs case]

_

( Likelihood Functions middotSuccess Cases - p - 1 c I

CATEGORIZATION OF ADDITIONAL PLATFORMS

I I I

bull CATEGORIES

Unexpected Survivals

Expected Survivals

bull APPROACH

Identify Corresponding Platform(s) Evaluated In the JIP

Determine Max Wave Height at Platform Location

Evaluate Whether Wave Hit the Deck

middot - Evaluate Likelihood of R and S to Differ from the Base Cases

Review Joint Information

Assess Special Features

bull IDENTIFY NUMBER OF PLATFORMS IN EACH CATEGORY

I

UPDATING OF CALIBRATION RESULTS

bull INCLUDE EFFECT OF ADDITIONAL PLATFORMS ON bullBbull

Consider Increased 1Uncertalntles In R and S

Develop bullNeW Ukellhood Functions for Different Categories

middotLet bullxbull Number of Ukely Unexpected Survival Cases and y Number of Ukely Expected Survival Cases

Determine Combined Likelihood Function lk (b Iadditional platforms) middot

bull REVISED LIKELIHOOD FUNCTION

lk(b I all cases) = lk(b I 9 platforms) middot x lk(b Ix platforms) x lk(b I y platforms)

bull ESTABLISH REVISED BIAS FACTOR B

f8 (b) u f8 (b) lk (b I all cases)

II

t t ~

shy

APPLICATIONS OF BIAS FACTORS

ii APPLICABLE TO PARTICULAR LOAD AND RESISTANCE RECIPE

bull USE BY THE API COMMITTEES AND MMS

Criteria Revisions

Recommended Design Guidelines Revisions bull

bullbull REASSESSMENT STUDIES AND DECISION MAKING

middot

bull RISK ASSESSMENT STUDIES BY THE OPERATORS

Individual Platforms

Fleet of Platforms

middot IMR Program Implementation

bullbull NOT TO BE USED IN DESIGN

I I fl J

EXAMPLE USE OF BIAS FACTORS

bull PHASE I EXAMPLE APPLICATION

bull EVALUATION OF ANNUAL PROBABILITY OF FAILURE

Establish Annual Maximum Seastate Data

Determine Coefficients to Define Base Shear

Determine Ultimate Capacity of Platform

Determine Probablllty of Failure Versus bullbbull

-bull DETERMINE UPDATED PROBABILITY OF FAILURE

Pf - P [S gt R]

Pf I P [~ gt Rb I b] f 8 (b) db

Pr - J P [S gt Rb I b] f bull 8 (b) db

bull TOTAL PROBABILITIES OF FAILURE

Total Pf - 1 bull Il [1 middotPf] - l Pf bull

Total Pf - 1 bull Il [1middot Pf] - l Pfmiddot

Total Pr = 1 - n [1middot PfbullJ = l Pfshy

I I I

EXAMPLE APPLICATION bull PHASE I

PLATFORM ST 151 K

bull GULF OF MEXICO

bull WATER DEPTH = 137 FT

bull ANNUAL MAXIMUM HS MEDIAN 175 ft COV 031

bull ASSUMED STORM DURATION 3 hoursmiddot

bull ASSUMED NUMBER OF WAVES 800

bull MEDIAN HS FOR STORM HOURS 1 2 AND 3

bull MEDIAN CURRENT VALUE bull

bull DETERMINE ULTIMATE CAPACITY FOR THREE DIRECTIONS

bull DETERMINE ANNUAL FAILURE PROBABILITIES

Broadsidemiddot Diagonal End-on

0037 0025 0007

0071 0044 0018

0022 0016 0004

DETERMINE TOTAL ANNUAL PROBABILITY OF FAILURE

Pf - 007

Pf - 013

Pr - 004

bull

EXAMPLE USE OF BIAS FACTORS - PHASE II middot

bull DETERMINE PREDICTIVE PROBABILITY OF FAILURE

Case - 1

Pf _ S S Pf (b b2) fa1a2 (b b2) db db2

g bull -1 (b1 R I S) - 1

92 - (b2 R2 I S) - 1

Case -2

Pf s ~ [bJ f11 (b) db

Pf [bJ p [g lt 0)

g - bull 1 (b1 R1 I S) - 1

1

bull1) l

I

BENEFITS OF ANDREW JIP bull PHASE II

lt

bull IMPROVED LOAD AND RESISTANCE RECIPE

bull IMPROVED CALIBRATION PROCEDURE

bull MULTIPLE BIAS FACTORS

bull PROCEDURE TO INCLUDE EFFECT OF OTHER PLATFORMS

bull PROCEDURE TO USE MULTIPLE BIAS FACTORS

  • Structure Bookmarks
    • ANDREW PHASE II JOINT1 INDUSTRY PROJECT
    • PROGRESS MEETING NUMBER 1
    • ANDREW PHASE II JOINT INDUSTRY PROJECT PROGRESS MEETING NUMBER 1 NOVEMBER 16 AGENDA
    • ANDREW PHASE II UST OF PARTICIPANTS
    • ANDREW PHASE I OBJECTIVES AND RESULTS
    • ANDREW PHASE II PROJECT OBJECTIVES
    • SUBSEQUENT PROJECTS
    • ANDREW PHASE II PROJECT TEAM
    • ANDREW PHASE II UPDATE MODELING middotAND ANALYSIS RECIPE
    • -ANDREW PHASE II CONDUCTOR RESISTANCE
    • ANDREW PHASE Il WA VE FORCE PROFILE
    • ANDREW PHASE II WA VE FORCE PROFILEmiddot
    • ANDREW PHASE II BRACE MODELING
    • ANDREW PHASE II FOUNDATIONS
    • ANDREW PHASE II middot middot JOINT MODELING
    • ANDREW PHASE Il JOINT MODELING OPTIONS DISCUSSION
    • ANDREW PHASE II JOINT MODEJJNG OPTIONS ILLUSTRATIONS
    • ANDREW PHASE II JOINT MODELING PHASE II METHOD
    • ANDREW PHASE II ANDREW STORM CONDmONS UPDATED
    • Platform Selection bullbullbull1111111111111
    • BasisCriteria for Selection bullbullbull1111111111111
    • CALIBRATION
    • WHERE TO INTRODUCE BIAS FACTOR (B
    • PHASE I -KEY CONCLUSIONS
    • bull CALIBRATION ISSUES PHASE II
    • PHASE II WORK SCOPE (CALIBRATION)
    • SOFTWARE
    • SENSITIVITY STUDIES POSTERIOR DISTRIBUTION OF B
    • WEIGHTING APPROACH bull
    • INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON
    • CATEGORIZATION OF ADDITIONAL PLATFORMS
    • UPDATING OF CALIBRATION RESULTS
    • APPLICATIONS OF BIAS FACTORS
    • EXAMPLE USE OF BIAS FACTORS
    • EXAMPLE APPLICATION bull PHASE I PLATFORM ST 151 K
    • EXAMPLE USE OF BIAS FACTORS -PHASE II middot
    • BENEFITS OF ANDREW JIP bull PHASE II
Page 3:  · . ' ANDREW PHASE II PROJECT OBJECTIVES . ' • ·UPDATE MODELING AND ANALYSIS RECIPE· JOINTS ' ' ·BRACES WAVE PROFILE FOUNDATIONS . ' ' CONDUCTORS· • ANALVZE PLATFORMS USING

I ~

1 It

I t I I

ANDREW PHASE II

UST OF PARTICIPANTS

bull

AMOCO PRODUCTION COMPANY

bull

CHEVRON PETROLEUM TECHNOLOGY COMPANY

bull

EXXON PRODUCTION RESEARCH COMPANY

bull MINERALS MANAGEMENT SERVICES

bullmiddot

MOBIL RESEARCH AND DEVELOPMENT CORPORATION

I bull PHILLIPS PETROLEUM COMPANY

bull bullSHELL OIL COMPANY

bull UNOCAL CORPORATION

i I I

j

ANDREW PHASE I

OBJECTIVES AND RESULTS

bull DEVELOPED A DATABASE OF PLATFORMS AFFECTED BY ANDREW

bull DEVELOPED AN ULTIMATE CAPACITY ANALYSIS RECIPE amp

PROCEDURE

bull IDENTIFIED DEFICIENCIES IN ULTIMATE CAPACITY ANALYSIS PROCESS

bull DEVELOPED A BAYESIAN UPDATING PROCEDURE

l

bull ESTABLISHED A GLOBAL BIAS FACTOR FOR PLATFORMS

bull PROVIDED EXAMPLES FOR THE USE OF THE BIAS FACTOR

ANDREW PHASE II

PROJECT OBJECTIVES

bull middotUPDATE MODELING AND ANALYSIS RECIPEmiddot

JOINTS

middotBRACES

WAVE PROFILE

FOUNDATIONS

CONDUCTORSmiddot

bull ANALVZE PLATFORMS USING UPDATED RECIPE

PLATFORMS SELECTED FROM FAMILY middotOF EXISTING STRUCTURES USED IN PHASE I

POSSIBLE middotREPLACEMENT IF MORE MEANINGFUL PLATFORM(S) middotFOUND

bull EXPAND BAYESIAN CALIBRATION

TWO BIAS FACTORSmiddot DEVELOPED JACKET AND FOUNDATION

UTILIZE RESULTS OF THE APlMMS FOUNDATION PROJECT

PRODUCE EXAMPLE USING EXPANDED PROCEDURE

-middot

bullbull

SUBSEQUENT PROJECTS

bull MMS INSPECTION PROJECT

bull APlMMS CAISSON PROJECT

bull APlMMS FOUNDATION PROJECT

bull TRIALSBENCHMARK JIP

I I It

ANDREW PHASE II

PROJECT TEAM

bull R L SMITH PROJECT MANAGER

bullbull F middot

J PUSKAR PROJECT ADVISOR

bull Rw LITTON TECHNICAL MANAGER

bull RAGGARWAL TECHNICAL DEVELOPER

bull D K DOLAN TECHNICAL ADVISOR middot

bull C A CORNELL RELIABILITY CONSULTANT

bull F MOSES RELIABILITY CONSULTANT

bull middotMSL ENGINEERING LTD TUBULAR JOINT CONSULTANT

q gtI

ANDREW PHASE II

UPDATE MODELING middotAND ANALYSIS RECIPE

bull CONDUCTOR RESISTANCE

bull WA VE FORCE PROFILE

bull BRACE MODELING

middotbull FOUNDATIONSmiddot

bull JOINT MODELING

~ I I

-ANDREW PHASE II

CONDUCTOR RESISTANCE

shybull RESISTANCE NOT INCLUDED IN PHASE I

bull CONTRIBUTION TO OVERALL RESISTANCE A PHASE II

STUDY ISSUE

middotbull middotCOMPARE FOR 4-LEG AND 8-LEG PLATFORM

bull CONSIDER CONDUCTOR POSmON IN GUIDES

- 13 LEFT 13 CENTER 13 RIGHT

Project ST151K Model Conductor version 1 Mon Nov 14 143140 1994

- ---- ---

--

117 s

111

7i

31

0 -5 -~ - ~()

- iltgt --lo

-(pO

-so

bull bullf OC)

-120

-ISo

-ISO

z

CAgt ~x Singl_e Conductor Model

bull

=I=

-_

-_

-gtshy

~ -

~

I

I~

- gt

~-----------____________________

Project STlSlK Model dir_2 version 7 Mon Nov 14 124200 1994

177 s

Ill

74

37

0- - e -zo - ~o

- Jo

-100

-10

-ISO

z

UV 1-+x

Project STlSlX Model Conductor Version 1 Mon Nov 14 140620 1994

bullmiddot

I

I

I I

I

I I

I I I I I

1 ==

-L

-L - L

-

- L

-

-

L

-

Uniform Displacement step 6 Inelastic Events Legend

Elastic

Strut Buckling Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield Beam Clmn FUlly Plastic Fracture

--

i~~~~~~~~~~~---~--~~-J___--JJ~_middot~~~=-

Project ST151K Model Conductor Version1 Mon Nov 14 142735 1994

---

~

I

I

middot

I

I

I

- shy- shy

- shy

- shy

4x z

CIV Uniform Displacement step 33 Inelastic Events Legend

bull Elasti~ Strut Buckling Strut Residual Strut Reloading Plastic StrutNLTruss Beam Clmn Initial Yield Beam Clmn-Fully Plastic Fracture

gt bull

Project STlSlK middot Model Conductor Version 2 Mon Nov 14 142025 1994

-----------------

bullf

I

I

I

II I I

I I I I

I ==

-

Variable Displacement step 9 Inelastic Events Legend

Elastic Strut Residual

bull

Strut Buckling Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield Beam Clmn Fully Plastic Fracture

--

Project ST151K Model Conductor Version 2 Mon Nov 14 142145 1994

I I I I

I I

I

I

I - - -

-Ishy-

-

Variable Displacement step 35 InelasticEvents Legend

Elastic Strut Buckling Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield Beam Clmn Fully Plastic Fracture

Project ST151K Model Conductormiddot

Version 1

Clgt - Cut Plane Force Fx Mon Nov 14 140051 1994

- -_ vii I fOIZM

0 _

~

N ID 0 bull 0

x 0

middotrt ~ () bull u 0 ~ 0 Ir

N bull

0

0 1 2 3 4 5 X Deck Disp [Ft)

I I 1 1 WbullTI- bullbull

lltcxi lr-C

011lt~~~~~~~---~~~~~~~~L~~~~~~~--~~~~~~~~i~~~~~~~--

Uniform Node Displacement

shy

Ji

r

Comparison of Pushover Curves for ST 151K Variable Load Pattern and Monotonic Load Pattern (Andrew)

Soll Improved 100

6000

5000 -i 4000

f

llS

3000 Ul

J 2000

10 c11 c1lgtolts 1000

0 v

bull ~

f

_ ----- --- ---shy

bull --Variable Lold

- - Andrew Load

I I I I I I

4 5 62 30 1

Deck Dlsplacement (ft)

bullbull r 1 II

~

ANDREW PHASE Il

WA VE FORCE PROFILE

bull DECK WA VE FORCE ~

PHASE I USED SECTION 17 (1056H degCd)

PHASE Il USE REVISED SECTION 17 (1000H Cd)

bull PHASE I USED SINGLE FORCE PROFILE FOR PUSHOVER

- USED WAVE HEIGHT CLOSE TO ULTIMATE CAPACITY

bull PHASE J

Il WILL STUDY THE

USE OF VARIABLE

PROFILE ~

- IF FORCE LEVEL INCREASES (OR DECREASES) BY A FACTOR EQUIVALENT TO A 2 FOOT CHANGE IN WA VE HEIGHT THE INCREMENTAL FO~CE PA1IERN IS CHANGED ACCORDINGLY (CAP FEATURE)

- INITIAL EXAMPLE

f~

a

middotn

a

I N -a a a amiddot

+

~ a

l

a w gt 0c 0 gt a I shyc -CJ wc w gtc(

== c( I shyZ w E W a 0 z- 8

ii sect

~ ~ bullT 0

I if

x ~

bullbullbull

II ~------------________

ANDREW PHASE II

WA VE FORCE PROFILEmiddot

CAP AUTOMATIC APPLICATION FEATURE

STAT

fR1 0 lt f lt1

STAT

STAT

f(R3-R2) middot f lt1

Project ST151K Model Wave_Loading Version 1 rue Nov 8 164107 1994 PT

E~ nss FT

SL 14amp FT

a 111 FT

EL 14 li

E-1 37 FT

El o FT

1middot1middot ______________ _ ________ __ _ __

0

200 400 600 800 1000 12000 Force at Elevatloii (kip)

200

a ll_

Wave Force at Major Elevations for Discrete Wave Heights

180

bullHmiddot40ft

Hmiddot495ft

H bull 51 ft

Hmiddot55ft_

H-59ft I

Andrew (H bull 60 86 ft) II

i I

_11---__

- middot- Hmiddot30ft

-

- bull -

middot bull bull bull middot

---Hmiddot53ft

bullbullbullbullHmiddot57ft

-----H-63ft

Elevation Above Mudllne 100

(ft)

180

140

120

80

60

40

20

Loading at Major Elevations

---H-49511

-deg-Hmiddot608611

0 200 400 600 800 1000 1200

Force It Elevatlon (kip)

--

Base Shear vs Wave Height ST151K

6000

5000 1---------Ult1mateCapacitybull 4890 kip (100 Improved Soll)-----------~--

4000a a lii ~ 3000

81 ampI

2000

1000

0 +-~~~--i~~~~~~~~~-+-~~~--lf--~~~~~~~~-+-~~~--l

30 35 40 45 50 55 60 65

Wave Height (ft)

bull middot

~

bull

II

2

g

~

It)

0

cCD E 0 E a c E euro~ CDshygtItgtmiddoto ~ti c -a Gic CD gt ~

-= c -Cll

I z

bullO 0i i ~ g 0

() CI -bull(q-ll) iuawow llu1wniiaAO

11

shy

116

114

112

g 110

bullc = 108

~ ~ 106 -sect 104

102

100

98

96

-

6555 60

Wave Height vs Center of Action ST151K

30 35 40 45 50

Wave Height (ft)

t

Comparison of Pushover Curves for ST 151K Variable Load Pattern and Monotonic Load Pattern (Andrew)

Soll Improved 100

--- --------- --- ----6000

5000

- 4000 8 - I

1 3000 Ul

81 m 2000

_J -shy

1000

---Variable Load

- - Andrew Load

0 f-~~~~~r--~~~~-t-~~~~~-t-~~~~~+-~~~~----t~~~~~-1

0 1 - 2 3 4 5 6

Deck Displacement (ft)

-middotshy

1 t I I ~

ANDREW PHASE II

BRACE MODELING

bull MARSHALL STRUTS USED FORPHASE I

bull PHASE II WILL ASSESS DIFFERENCE IN USING BUCKLING BEAMS AND MARSHALL STRUTS

middot bull COMPARE FOR 2D FRAME

bull COMPARE FOR PLATFORM 177 WHERE LOWER

BAY EXHIBITS LARGE BRACE MOMENTS

Project andrew Model beamframe Version 1 Mon Nov 14 145447 1994

~

middot Frame Test buckling beam-column model

-

Project andrew Model strutframe Version 1 Mon Nov 14 145754 1994

bull

Frame Test strut model

middot-middot _____________--_

_middotmiddotr_~ --ll--~-middot__

Buckling Beam bull

Strut (K=O 65)

Lateral Displacement

ANDREW PHASE II

FOUNDATIONS

bull ~LICIT NONLINEAR API FOUNDATION USED INiPHASE I

bull FOUNDATIONS

JIP SELECTED MINI-VANE DATA FOR

STRENGTH PROFILE

bull COMMENTS ON BENCHMARK COMPARISONS

- SIGNIFICANT VARIABILITY DUE TO FOUNDATION MODELING METHODS

II Ii

z

ctV 4x SPlO caissonfy-42s soil-api(lf-097step-114)

Inelastic Events Legend Elastic Strut Buckling

Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield

--------shyBeam Clmn Fully Plastic Fracture

~--~~~~~~~~~--~----~__i~~---~~~~__--~

Pro)ect apipile Model South Pelto 10 Version Wed May 18072235 1994 - -

1bullt

+

t

f i bullbullbull

bullbull bullbull

I I

5

~x SPlO Caissonfy=42s soil=api(lf=lOOstep=137)

Inelastic Events Legend

Elastic Strut Bucklilg

Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield

earn Clmn Full~ Plas1c Fracture

ANDREW PHASE II middot middot

JOINT MODELING

middotbull API RIGID-PLASTIC USED IN PHASE I

a ELASTIC-BRITTLE USED IN PHASE IB

amiddot MSL WILL PROVIDE NONLINEAR FORCE-DEFORMATION

AND MOMENT-DEFORMATION RELATIONS FOR PHASE II

a MODELING OPTIONS FOR PHASE II -

l

- UNCO~LEDmiddotONE DIMENSIONAL MEMBER END JOINT MODELING

J Ii

ANDREW PHASE Il

JOINT MODELING OPTIONS DISCUSSION

bull JD COUPLED INTERACTION SURFACE

- IDEAL

- DATA NOT AVAILABLE TO CALIBRATE

bull 2D COUPLED INTERACTION SURFACE

- SIGNIFICANT MODELING IMPROVEMENT

- DATAAVAILABLE

- IMPLEMENTATION PRACTICAL BUT NEEDS DEVELOPMENT AND CALIBRATION

bull 2D UNCOUPLED INTERACTION SURFACE

- SIGNIFICANT MODELING IMPROVEMENT

- DATA AVAILABLE

- IMPLEMENTATION PRACTICAL AND AVAILABLE

ANDREW PHASE II

JOINT MODEJJNG OPTIONS ILLUSTRATIONS

P

e

ANDREW PHASE II

JOINT MODELING PHASE II METHOD

e

I I I

ANDREW PHASE II

ANDREW STORM CONDmONS UPDATED

bull -PHASE I USED OCEANWEATHER HINDCAST-(Nov 1992)

bull middotmiddotPHASE II WILL USE THE REVISED HINDCAST (Nov 1994)

bull lt c lt I lt

2-2

l---4-~~~+-~~-+-~~~1--~~-+-~~--l--~T--+-~~--i24

f-tJVtMfiz~ 1Cfl (r STvb(

Plollol bullbull 5middotNOYmiddot9214bull1J Irbullbull Illbull [ANDREVl~NDREVCUSTbullMAXll 5-NOY-1992 10bull54 - 9208

HEIGHT lml

I~ MAXI

I L 1~middotamp SZ~- ~A I 13c

I J) ( r I X~0iWJ~0gtJ gt L gt gt I 1c I 120

-- N

~ deg

I I I I I I e C I ~=-41 122 -94 -92 -90 -88 -86 -84 -82 -80

-

Figure 7 contours of maximum hindcast significant wave height

_

~

~tl ~

NOltNttIJ 17lt ltNS S iJJ)y

Maximum Significant Wave Height - Hurricane Andrew 1992

30 ~

26 ~ I ~-~~~- == -shy -~ 35-_

24deg1--shy --------~- middot-middot-------middotmiddotmiddot----middot------middot-middot~middot- - - ~-----1---middot

22middot ~~~~~~~-~ -94 -92 -90 -88 -84deg-86 -82 -somiddot

c

-

Platform Selection bullbullbull1111111111111

0 BasisCriteria for Selection middot

0 Review of Phase I Platforms middot 0 Alternative Platforms

0 PMB Recommendations

0 Participants Comments

gt

L

l I

BasisCriteria for Selection bullbullbull1111111111111

0 Benefit for Calibration - Classification (UnexpectedExpected)

- Failure Mode (StructuresFoundations)

- Damaged StructuresKnown response

- Better Representation of Full ~opulation

0 Availability of Data - Existing Models

- Inspection Data

- Damage Data

PREDICTED BEHAVIOR

-

OBSERVED

BEHAVIOR

-SURVIVED DAMAGED FAILED

SURVIVED

ST134W WD90A

-MC311 -

MC397 -

--

-

-

ST151K ST130Q

-

-

-

DAMAGED - -

-

-

--

-

-

ST52 SS139 ST177B -

ST161A

-_

-

FAILED

-

-

-

ST151H ST130A ST72

Operator - Platfonns - Platfonns Survived Damagedamp

- Falled -

bull

-

-

Jle eBDIGIPllDbl -Amocoshy 30 1 Chevron -

143 12 Exxon 72 Mobll 22 1 Phllllps 7 bull

-Shell 39 Unocal 25 1

Total Partlclnts Platfonns 338 15

Other Operators bull- 310 13

-

Total Platfonns - -

648 28

fbml

Platforms Considered 6 7

tie of Total Platfonns 1 25

OVERVIEW OF PLATFORMS SURVIVEDDAMAGEDFAILED

-

shy

v

Platforms Evaluated in Phase I

-

-Platfonn Water Year Number Analysis Avallabillty of Availability Platform

- Name Depth rl of Model DetaDed rl ClassUlcatlon

Installation legs Inspection Soll Report Based on Information Phase I Calibration - -ft

- -- Suaill flalfnrm ~BSi -STISIK 137 1963 8 CAP Yes Yes Un eel Sarvlval

-STIJOO 170 1964 4 CAPmiddotshy Yes ST134data u ST134W 137 1981 4 CAP Yes Yesmiddot E

-

edSarvlval -

middotsurvival

WD90A 184 - 1964 8 CAP Yes E -

- Sarvlvill

MC3ll 343 1978 8 KARMA

MC397 468 1991 4 KARMA -

bull ~reglmmiddot- -

Yes Sure Survival

bull Yes - Sare Sanlval-

-T23STS2 63 1969 4 CAP

T2SSS139 62 1969 4 CAP -

Yes

Yes

Yes Ex

Yes Une

bull SurvivalLikely to DamalI

middot bull SanlvalLlkely to DamalI

ST161A 118 1964 8 USFOS

Eail11a pound1al(bullllm Clss ST177B 142 --1965 8 CAP

STISIH 137 1964 - 8 CAP C

STIJOA 140 19S8 8 CAP - -

Yes

Yes

ollamed

Yes

- Yes E

-

at about 1 mUe

Yes

ST134data

bull SafvlvalLlkelv to Dam- bull- -

E middotrahre

Emeried Failure -

Likely FaUare

T21ST72 61 1969 4 CAP Collamed No Likely FaUure

Platforms Identified for Evaluating Their Likelihood of Falling Under UnexpeCted Survival Category

CllEllATOa

AllB

llLOCll

Sl1lvcnJllE

NAME

YEAR

INSTAUJCD

WATER

IJBPJB

middotmiddotDISTANCE

ftOMSllOllE

(MIUIS)

Ill

(M)

AMOCO BllW

STllI

WDI075

A

bull D

4

n uo 17J

44

32

bull 20

10

bull CllEVllOH STtllO c I 1bull 21 11

STl130 D 0 161 21 11 STllJ4 I 0 137 D - 11

STIUI G I 137 32 11

STIUI I 0 121 32 11

STIUI

J 0 140 32 11

STIUI It 4 137 32 11

STllSI PROl)I 0 137 32 11

STllSI PROD-2 0 137 32 11

STll7 A 0

140 35 11

WDl117 c 65 214 34 10

DXON STshy E 51 14 bull STl165 A t3 bull 34 WD0073

WD0073

WD0073

A _ c

4

62

4

IA IA

17J

22

22

22

bullbullbull WD003-shy E 65 161 Z7

A 0 IN 31 MOBD Bl12I

ssn A bull

50

50

31

7

bull 7

MIJllPllY SS114 B 7 50 7 7

PlllUIPS SSl14t A 55 33 7

SAMEDAN ST17J A 4 117 7 bull 7

SJIELL WD0103 A 65 223 27 WDllOJ bull 65 221 27 WDll04 c 65 221 27 WDOI04 D 67 2 27

UNOCAL

SSl20I

SSl20I

bull SSl20I

SS0209

SS0209

SS0209

SS0209

SS 0253

SS 0253

E

I

PtJllE

A

bull D

G A

c

0

4

4

7J

61

0

0

69

103

97

97

9S

100

95

100

165

175

34

34

34

34

34

34

34

54

54

bullbullbullbullbullbullbullbullbull

Platforms Recommended for Phase II

llalr y_ w- Aot Av- _ I 1 Doplla Model nm bull bull _ bull I I H11 s18-1 Irht lbwl~ bullbullbull(-illtdlel OM lllMtD

II ---

-STISIIC 137 1963 I CAP Yeo Yeo I Yeo

STtshy -170 1964 4 CAP Yeo ST134dlU I Yeo

STl34W 137 Yeo Yeo -1981 4 CAP D Yeo

WD90A 114 1964 I CAP Yeo - D

MC311 30 1971 I KARMA Yeo -sre-

MC397 middot 461 1991 4 ICARMA Yeo Sireshy - - r- -

__ T13ST5l 63 1969 4 CAP Yeo Yeobull middot- ~n- D Yeo _ Tl5SSl39 62 1969 4 CAP Yeo Yeo I Yeo

_ STIT711 142 1965 I CAP Yeo I Yeo middot-1shySTl61A Ill 1964 I USFOS Yeo Yeo - D bulloshy

- r-

STISIH 137 1964 I CAP y D y -_ STl31A 140 1951 I CAP Yeo STl34dlU I Yeo

T21Sll 61 1969 4 CAP No rar-middot I Yeo

~-rmiddot-middot SS209A 95 1967 16 CAP Yeo D -SSU4R

11 UI A 41 -in D SS 2741AIM CAP bull

_ _ -shy - -shy- -shy

CALIBRATION

bull PHASEIAPPROACH

bull CALIBRATION ISSUES

bull PHASE II WORK SCOPE

bull SOFlWARE AVAILABILITY

bull EFFECT OF ADDITIONAL PLATFORMS

bull WEIGHTING APPROACH

bull EXAMPLE APPLICATION middot

(l=Ofce Mode] Wind

~

fcurrent l -middot

~SW ~leldl4

ultimate resistance force I first

design load Illgt

i _

deflection

-

bull

4

Historical Hurrlcane Dariiage and Survival of Platforms

Components Required for Calibration (Petrauskas 1992)

PRIOR Distribution

PLATFORM EXPERIENCE

BAYESIAN Survivals and Failures

UPDATING

Waves Winds and Currents bull

POSTERIOR Distribution

- A_middotJA ~ Survival Data Failure Data

Result Result

Load Resistance

middotfl

-

Bayesian Updaing middotGeneral Description (Petrauskas 1992)

L Imiddot

WHERE TO INTRODUCE BIAS FACTOR (B

On Resistance (Rmue - B (RPREDICreD I

middotOn Loads (Smue middot - B (SPRE01cTEo

On Safety Margin (RSmue - B (RSPREDICTED

CALIBRATION METHODOLOGY

gt

r

FIELD OBSERVATIONS

Observed Behavior

survivals Damages

Failures

r

bull

CAPACITY ANALYSIS

Predicted Load and Resistance Predicted Failure Modes

Survival Damage Failure cases

RELIABILITY ANALYSIS

- Probability of Failure ~

Ukellhood Failure

BAYESIAN UPDATING

Prior Distnbution of B Combined Likelihood Function

Posterior Dlstnbution of B

BIAS FACTOR

Mean Value cov bull

Represents Modeling Uncertainties

Prior Distribution of B Likelihood Functionmiddot Observing Success case STISIK and Posterior Distribution of B

2 00

uo

160

140

120

e 100 L $ uoIshy

0 60 ~ ~ 040

020

0 00

0 00 0 50

_ I 00

b

I 50 2 00

-shy lk [bl STISIK)

--shy Pnor-f(b)

--shy Pos1enor r(bl ISIK)

Prior or B Mean bull 10 COV bull 030

Posterlor of Bbull Meanbull 136COV 0 16

2 50

Likelihood Function and Posterior Distribution Example Platform STISIK

- -

~

- Prior and Posterior Distributions or B Failure cases (ST177B STlSlH ST130A T21) Damage cases (T23 T25 ST161A) Success cases (STlSlK ST130Q ST134W WD90A MC311 MC397)

400

350

300

2 so 0-

I

bull bull

bull bull gt

- bullbull ~ bull bull I bull

_ bull bull bullbull I bull bull bullII V -

bull bull

I bull ~ bull

bullbull bullI bull bullI bull bull bull

middot -~

bull__ ~ middotI middotmiddot

L 2 00

e ~ I SO

-I 00

- 0 so

000 gt

0 00 020 0 40 060 0 80 I 00 120 140 160 110 200

b -

-

shy

Prior-r

- - - - Posterior fLlll failure cases

bull bull bull bull bull bull bull Posterior flall damage cases

- bull bull - bull middot Postenor flall success cases

Posterior flall success fulure damage cases

-

Posterior of B Mean= 119 COV 0101

Posterior Distribution or Blas Factor (8) bull Different Categories and All Cases Combined

shy

l

PHASE I - KEY CONCLUSIONS

bull GLOBAL CORRECTION FACTOR

bull 15 TO 20 CONSERVATISM MEAN)

bull UNEXPECTED SURVIVALS HAVE GREATEST IMPACT

bull DAMAGED CASES MORE INFORMATIVE

bull CALIBRATION ISSUES PHASE II

bull

EFFECT OF IMPROVED CAPACITY ANALYSIS RECIPE

bull APPLICABILITY TO MULTIPLE FAILURE MODES

bull NO FOUNDATION DAMAGE CASE INCLUDED

bull IMPROVED REPRESENTATION OF FULL ANDREW DISTRIBUTION

-bull NUMBER OF PLATFORMS CONSIDERED

I 1 I

I t ) bull ~

PHASE II WORK SCOPE (CALIBRATION)

bull

bull ESTABLISH MULTIPLE BIAS FACTORS

Use Nine Base Case Platforms

middot - Capacity Analysis for Revised recipe

Foundation Blas Factor from Foundation JIP

bull DEVELOP WEIGHTING APPROACH

bull UPDATE MULTIPLE BIAS FACTORS

bullEffect of Platforms Not In the Nine Base Cases

bull EXAMPLE APPLICATION OF BIAS FACTORS I

bullbull

PHASE I PHASE II

CAPACITYbull ANALYSISRu bull

---------------------1----middot

Jc1 - iJbullraquoJ dbull

FttlS frt[f~l ratdhJJ

-

LIKELIHOOD FUNCTION

DEVELOPMENT

lk Jb Jfall1N) bull Jb)

111 Ill Isurvtvoq bull 1 - bl

-middot -------------------shy ~---------------------middot-middot

r111ir8 Cb1 111 Cbl BAYESIAN

UPDATING

~cove

R1R2 bull middot---------------------~----

Jblbull P(g cOJ

1 bull JbRI (MBSI I bull 1

~middot

llcJllJfalluro) bull P(gcOJ Ill 111 b21fallun) bull P ((g1 CO) U 1112 c OD

middot--------------------shy ----------------------shy bull

r111111r8 Cbl 111 Ill i-1nro1

deg ~1b2)laquottfrr (b1b21-lnfo) re Ill bull 1-11middot b2) bull lodegamp

middot

i1 covamp i2 cov~i

-

shy

--

Likelihood Function vs b All Success Case STISI K STl30Q STl34W WD90A MC311 MC397

I 0000

0 9000

OIOOO

0 7000

~ 0amp000

ta osoOO

~ 0 4000

0 3000

0 2000

0 1000

bull 0 0000

bull

I

middot

middot bull I

Ibull

-sibull ~ -middot-middot-middot-middot -middot-

----7----+-----~ I

Ibull bull bull I

rbull

--- Jk ISIK[blsuccess)

- bull - bull - lk 130Q[blsuccess) bull

lk 134W[blsuccess)

Jk-WD90A(blsuccc~I

- - - lk MC397[blsucccss)

--ltgt-- lk middotMC31 l[blsucccss)

--- Jk [bl 6 out of6succcsHascs)

05 I 5 2 250

b

Likelihood Functionsmiddot Success Cases - Phbullsc L

shy

bullbull bull bull bull bull bull -------------------~

bull bull bullbull bullbull bullbull bullbull bullbull bullbull bull

bull bullbull

bullbullbullbull

bullbull

bullbullbullbullbullbullbull

bullbullbullbullbullbull

bullbullbullbullbullbullbull

z 0

t z ) LL

c 0 0c i ~ i

J

~ sect ~--------------~ e l5

I

I

SOFTWARE

bull PF PROGRAM

bull CALREL

bull PF +CALREL

bull RELACS

bull

gt i

l

SENSITIVITY STUDIES

POSTERIOR DISTRIBUTION OF B

bull EFFECT OF CHANGE IN DISTRIBUTION OF PRIOR OF B

bull EFFECT OF REMOVAL OF ONE PLATFORM FROM EACH CATEGORY

bull EFFECT OF REMOVAL OF THREE DAMAGE CASES

bull EFFECT OF ADDITIONAL 13 SIMILAR PLATFORMS

-I

Prior and Posterior Distributions of B Variation In COV of PriorofB fromOl to04 -- -

4500

-4000

-

- -

3500 --middot - bull bull - bull bull Prior-r ICOV =02

3000

- - - - Prior-r ICOV =03 -

bull -~ e 2soo -

bull bull bull bull bull bull bull Pnor-r ICOV =o 4t frJ ~ 2000 Posterior rlCOV =0 2

bull bull Postenor rlCOV =0 3 bull bull 1500 bull middot bull - shy -middot- Posterior rlCOV =04

middot~ I 000 - bull

-middotmiddot ~ ~middotmiddot M -~

- middot 0 500

wbullbullbull _ I middot~ middotmiddotmiddotmiddotmiddot middotmiddotmiddotmiddotmiddot - -~ ~ __ middot_ _ bull -shy0 000

0 000 0200 0 400 0 600 0 800 IOOO 1200 1400 1600 1800 2000 -

- - b

-

=shy

Posterior Distribution of Bias Factor (B) bull Effect of Variation In COV of Prior of B

bull

bullPrior and Posterior Distributions of B Evaluating Effect of Removal of One Platform from Each Category from the Base Case of 13 Platforms

400

~ ~ 350

j IA i

3 00 bull I bull I ---Prior-r I bull l

250 1 I 1

I bull bull ---Posterior r lall 13 casesshy- ~E 2 00 bull bull bull bull bull bull bull Posterior r without STISIK

- ~ 1 bull - bull I - bull bull - bull bull Posterior r Iwithout T2I I 50 I

J-tt~i - bull - bull - Posterior r Iwithout T2S

tOO [-

I I ~ ~ I 0 50

0 00

___ r J i

l~T

~--

0 00 0 20 0 40 0 60 080 I 00 120 140 160 110 2 00

b

zshy

Posterior Distribution of Bias Factor (8) bull Effect of Removal of middotPlatform from Any Category

bullPrior and Posterior Distributions or B Evaluating Effect or Removal or 3 Damage Cases from the Base Case or 13 Platrorms

4

35

3

25

e L 2 ~ bulle shy

I 15

bull I

05

0

0 02 04 06 I 2 I 4 16 IS 2

I I

I

I

I Ibull

J -tmiddot L--

bullI

~

I

fl

II ~

I I v I

middot bull bull

bull L

~ ll- I-~lt

OS I

b

I

bull

1middot bullbullbullbull

~ __

bull

- - - - Prior-r

- - - - Postenor flall failure cases

Postenor flall success cases

--- Posterior flall 13 cases

---Posterior fI success failure cases only

IPosterior or B Mean =t11 COV =0121

-

_I

Posterir Distribution or Bias Factor (B) bull Effect or Removal or 3-Damage Cases

-

bullPrior and Posterior Distributions or B Evaluating Effect or Additional 13 Similar Observations as

Used in Andrew JIP The POsterior or B from Andrew-JIP Considered as New Prior or B

- 500

450

400

350

bull bull

bull bull bull bull J bull

bull bull f ~ I

I ~ Ibull bull

bullbull bull

l bull I -bullbull bull bull

I

I bull bull

bull I I~~bull bull bull

bull ~

I bull bull

~ ~_ ~~~

- - New Prior r bull

Posterior 114 new failure cases ~ 300- bull bull bull bull bull bull bull Posterior f 13 new damage casesl 150 ~

- bull bull - bull bull Posterior 116 new success cases i 2 00

Postenor 1113 new success failure -150 damage cases

I 00

0 50 -

0 00 0 00 0 20 0 40 0 60 0 80 I 00 120 I 40 160 110 2 00

Posterior ol B Mean= 120 COV= 0071 b

shy

shy

_rshy

Posterior Distribution of Bias Factor (B) bull Effect of Additional 13-Similar Observations

1

1 I

WEIGHTING APPROACH bull

bull ASSESSMENT OF PLATFORMS DATA

Select Platforms Most Mappable to the Analyzed Platforms

Unlikely SurvivalDamage Cases Identification

Based on Cut-Off Criteria Agemiddot Hs etc

bull INFORMATION REQUIRED FOR COMPARISON

bull CATEGORIZATION OF ADDITIONALmiddot PLATFORMS

bull UPDATING OF CALIBRATION RESULTS

--

OVERVIEW OF PLATFORMS SURVIVEDDAMAGED -~

Operator -

- Platforms

Survived CS)

Platforms

Damaged CD)

Platforms Used In

Phase IJIP -

JI~ ~Bill~IRBDll -

Amoco -

Chevron -

Exxon

Mob II

Phllllps

Shell

Unocal -

30

143

72 22

7

39 25shy -

1

12

bull

1 bull

1

1(S)+1 (D)

3 (S) +3(D)

2 (S)

bull shy

bull

bull

bull

Total Partlclnants Platforms 338 - 15 10C6S+4D) -

Other Operators

310

13

3 (D)

Total Platforms 648 28 13C6S+7D) -

Summary of Andrew JIP bull Phase I Platforms

Plairshy w v- N-ber Lea-le 1 shy car Hind_ o-w N-bull Deplh fl fl IJpe la bull n Yeadon Jadtet lbelledtf liefalit -- _ -OD

IL H - -shy _ICllshy IL IL - - rQClto ~ _

K STUIK 137 1963 8 365 608S4 Cellardeck

-middot STl-shy 170 1964 4 -~ K 39 62269 Cellardeck - --ST134W 137 1911 4 43 608S4 Smvlnlshy- _ - -

WD90A 184 1964 8 K 267Sbull34 83 SOS31 SUboftllar deck SmvMI

MC311 343 1978 8 x S7 62 - MC397 468 1991 4 x shy S33bull6SO 65911 -shySmeSmvlvol -- gt -

r-- T23ST52 63 1969 4 ton- shyKshy 52594 S03 -T25SSl39 62 1969 4 Kshy SOS94 S063 - middot ton-shy__ -

bullbull a ton--shyST161A 118 1964 8 K - 347S 51495 Cellarclecti

r~

ST1771 142 1965 8 K 39 60167 CeDordeck failme

STISIH 137 1964 8 K 395 608S4 CeDardeck Failme

K ST130A 140 1958 a SS 6095 I lblv Pai1uno - T21ST72 61 1969 4 Kshy S4S94 4974 I lhlv Pai1uno -

- -

JE

INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON

bull WATER DEPTH

bull bull YEAR OF INSTALLATIONDESIGN

bull PLATFORM ORIENTATION

bull PLATFORM LATITUDE AND LONGITUDE

bull KEY SKETCHES OF PLATFORM STRUCTURAL FRAMING

Number of Legs Plies Leg Batter Number of Conductors bull Deck Elevations (Including sub-cellar deck If any) Jacket Framing Elevations Brace type In Vertical Frames Sizes of Primary Members (leg pile vertical frames braces primary horizontals) Joint Details (can gap etc) LegPiie Annulus Grouting Status

bull middotPOST ANDREW INSPECTION REPORT Significant Wave Height Hs (ft)

bull INFORMATION ON DAMAGES PRIOR TO ANDREW

bull RESULTS FROM IN-HOUSE STUDIES IF AVAILABLE

bull ~RIGINAL DESIGN CRITERIA INFORMATION IF AVAILABLE

bull SOIL SHEAR STRENGTH PROFILE IF AVAILABLE

Platforms Identified for Evaluating Therr Likelihood of Falling Under Unexpected Survival Category

OPlRATOR

middotADA

BLOClt

llmucnJllE

NAME

YEAll

INSTAUJD

WA1ER

DBPTll

DISTANCE

JlllOM SHOllE

(MILU)

Bl

(M)

AMOCO 111113

STl11

WDI075

A

bull D

St

17

120

1n

44

20

10 I

CllllVllQfl STlllO c 61 110 21 11 STIUO D Q 1110 21 11

STl134 p Q 137 29 11

ST0151 G 61 137 11

ST0151 I 3 121 11

ST0151 J Q 140 11

STl151 It 137 11

ST0151 PJl001 Q UT 11

ST0151 PROD-2 Q UT 11

ST017 A 3 140 35 11

WDIUT c 65 214 34 10

llXXON STI055 II 51 6T 14 I

STl15 A t3 34

WDI073

WD0073

Amiddotshy Q

1111 1111

22

22

I

I

WDI073 c 1n 22 I

- WDOCl93-shy II

65 1110 2T A 3 194 31 MOBU 111121

SS IOl7

A B

501

50

50

31

T

I

T 6

MtlllPllY SS114 B T 50 T T

PlllllIPS SS0149 A 5 55 33 T

SAMllDAN STln A 107 T T

SBllLL WDl103 A 65 223 2T WDl103 B 65 221 2T WD0104 c 65 221 2T WD0104 D 6T 20 2T

UNOCAL SS020I

SS 020I SSl20I

SS 0209

SS0209

II p

PURll

A

B

3

n 61

103

97

97

95

100

34

34

34

34 34

bullbull bullbullbull

- SS0209 D u 95 34 I

- SS 1209 SSl253

SS0253

G A

c

6T

2

100

165

1T5

34

54

54

bullbull I

I I

I I

SOUTH MARCH

~LAND

N

SOUTH PaTOI ENE ISWltD AREA

MMS Defined Limits ror Offshore Structures Requiring Inspections Following Andrew

t---Jbulln3 T25bull I bullTtl

bull1$1Htl1Abull bull1SIK1779bull

~

middot-DaTA AREA~T

CRANO rsLE

AREA bull1300

bullll4W

=shy

bullPath of Hurricane Andrew with Platforms Used In Calibration

Likelihood Function vs b All Success Case STISI K STIJOQ ST134W WD90A MCJll MCJ97

I 0000

0 9000

0 8000

07000

-Bo6000

tiiii 0 5000 -

04000

0 3000

0 2000

01000

0 0000

0 05 25

ri1rt middot-middot-middot-middot-middot-

I 5 2

b

---11 middot151K(blsu~ccsI

- bull - bull - fk 130Q(bfU~lCSS)

lie bull134W(blsuccc]

lk-WD90A[blsultCS]

- - - lk middotMC397(bfUtlCS]

--o-- lk-MCJll(bbuctcsJ

--- ll (bl 6 out of 6 Mlttcs case]

_

( Likelihood Functions middotSuccess Cases - p - 1 c I

CATEGORIZATION OF ADDITIONAL PLATFORMS

I I I

bull CATEGORIES

Unexpected Survivals

Expected Survivals

bull APPROACH

Identify Corresponding Platform(s) Evaluated In the JIP

Determine Max Wave Height at Platform Location

Evaluate Whether Wave Hit the Deck

middot - Evaluate Likelihood of R and S to Differ from the Base Cases

Review Joint Information

Assess Special Features

bull IDENTIFY NUMBER OF PLATFORMS IN EACH CATEGORY

I

UPDATING OF CALIBRATION RESULTS

bull INCLUDE EFFECT OF ADDITIONAL PLATFORMS ON bullBbull

Consider Increased 1Uncertalntles In R and S

Develop bullNeW Ukellhood Functions for Different Categories

middotLet bullxbull Number of Ukely Unexpected Survival Cases and y Number of Ukely Expected Survival Cases

Determine Combined Likelihood Function lk (b Iadditional platforms) middot

bull REVISED LIKELIHOOD FUNCTION

lk(b I all cases) = lk(b I 9 platforms) middot x lk(b Ix platforms) x lk(b I y platforms)

bull ESTABLISH REVISED BIAS FACTOR B

f8 (b) u f8 (b) lk (b I all cases)

II

t t ~

shy

APPLICATIONS OF BIAS FACTORS

ii APPLICABLE TO PARTICULAR LOAD AND RESISTANCE RECIPE

bull USE BY THE API COMMITTEES AND MMS

Criteria Revisions

Recommended Design Guidelines Revisions bull

bullbull REASSESSMENT STUDIES AND DECISION MAKING

middot

bull RISK ASSESSMENT STUDIES BY THE OPERATORS

Individual Platforms

Fleet of Platforms

middot IMR Program Implementation

bullbull NOT TO BE USED IN DESIGN

I I fl J

EXAMPLE USE OF BIAS FACTORS

bull PHASE I EXAMPLE APPLICATION

bull EVALUATION OF ANNUAL PROBABILITY OF FAILURE

Establish Annual Maximum Seastate Data

Determine Coefficients to Define Base Shear

Determine Ultimate Capacity of Platform

Determine Probablllty of Failure Versus bullbbull

-bull DETERMINE UPDATED PROBABILITY OF FAILURE

Pf - P [S gt R]

Pf I P [~ gt Rb I b] f 8 (b) db

Pr - J P [S gt Rb I b] f bull 8 (b) db

bull TOTAL PROBABILITIES OF FAILURE

Total Pf - 1 bull Il [1 middotPf] - l Pf bull

Total Pf - 1 bull Il [1middot Pf] - l Pfmiddot

Total Pr = 1 - n [1middot PfbullJ = l Pfshy

I I I

EXAMPLE APPLICATION bull PHASE I

PLATFORM ST 151 K

bull GULF OF MEXICO

bull WATER DEPTH = 137 FT

bull ANNUAL MAXIMUM HS MEDIAN 175 ft COV 031

bull ASSUMED STORM DURATION 3 hoursmiddot

bull ASSUMED NUMBER OF WAVES 800

bull MEDIAN HS FOR STORM HOURS 1 2 AND 3

bull MEDIAN CURRENT VALUE bull

bull DETERMINE ULTIMATE CAPACITY FOR THREE DIRECTIONS

bull DETERMINE ANNUAL FAILURE PROBABILITIES

Broadsidemiddot Diagonal End-on

0037 0025 0007

0071 0044 0018

0022 0016 0004

DETERMINE TOTAL ANNUAL PROBABILITY OF FAILURE

Pf - 007

Pf - 013

Pr - 004

bull

EXAMPLE USE OF BIAS FACTORS - PHASE II middot

bull DETERMINE PREDICTIVE PROBABILITY OF FAILURE

Case - 1

Pf _ S S Pf (b b2) fa1a2 (b b2) db db2

g bull -1 (b1 R I S) - 1

92 - (b2 R2 I S) - 1

Case -2

Pf s ~ [bJ f11 (b) db

Pf [bJ p [g lt 0)

g - bull 1 (b1 R1 I S) - 1

1

bull1) l

I

BENEFITS OF ANDREW JIP bull PHASE II

lt

bull IMPROVED LOAD AND RESISTANCE RECIPE

bull IMPROVED CALIBRATION PROCEDURE

bull MULTIPLE BIAS FACTORS

bull PROCEDURE TO INCLUDE EFFECT OF OTHER PLATFORMS

bull PROCEDURE TO USE MULTIPLE BIAS FACTORS

  • Structure Bookmarks
    • ANDREW PHASE II JOINT1 INDUSTRY PROJECT
    • PROGRESS MEETING NUMBER 1
    • ANDREW PHASE II JOINT INDUSTRY PROJECT PROGRESS MEETING NUMBER 1 NOVEMBER 16 AGENDA
    • ANDREW PHASE II UST OF PARTICIPANTS
    • ANDREW PHASE I OBJECTIVES AND RESULTS
    • ANDREW PHASE II PROJECT OBJECTIVES
    • SUBSEQUENT PROJECTS
    • ANDREW PHASE II PROJECT TEAM
    • ANDREW PHASE II UPDATE MODELING middotAND ANALYSIS RECIPE
    • -ANDREW PHASE II CONDUCTOR RESISTANCE
    • ANDREW PHASE Il WA VE FORCE PROFILE
    • ANDREW PHASE II WA VE FORCE PROFILEmiddot
    • ANDREW PHASE II BRACE MODELING
    • ANDREW PHASE II FOUNDATIONS
    • ANDREW PHASE II middot middot JOINT MODELING
    • ANDREW PHASE Il JOINT MODELING OPTIONS DISCUSSION
    • ANDREW PHASE II JOINT MODEJJNG OPTIONS ILLUSTRATIONS
    • ANDREW PHASE II JOINT MODELING PHASE II METHOD
    • ANDREW PHASE II ANDREW STORM CONDmONS UPDATED
    • Platform Selection bullbullbull1111111111111
    • BasisCriteria for Selection bullbullbull1111111111111
    • CALIBRATION
    • WHERE TO INTRODUCE BIAS FACTOR (B
    • PHASE I -KEY CONCLUSIONS
    • bull CALIBRATION ISSUES PHASE II
    • PHASE II WORK SCOPE (CALIBRATION)
    • SOFTWARE
    • SENSITIVITY STUDIES POSTERIOR DISTRIBUTION OF B
    • WEIGHTING APPROACH bull
    • INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON
    • CATEGORIZATION OF ADDITIONAL PLATFORMS
    • UPDATING OF CALIBRATION RESULTS
    • APPLICATIONS OF BIAS FACTORS
    • EXAMPLE USE OF BIAS FACTORS
    • EXAMPLE APPLICATION bull PHASE I PLATFORM ST 151 K
    • EXAMPLE USE OF BIAS FACTORS -PHASE II middot
    • BENEFITS OF ANDREW JIP bull PHASE II
Page 4:  · . ' ANDREW PHASE II PROJECT OBJECTIVES . ' • ·UPDATE MODELING AND ANALYSIS RECIPE· JOINTS ' ' ·BRACES WAVE PROFILE FOUNDATIONS . ' ' CONDUCTORS· • ANALVZE PLATFORMS USING

i I I

j

ANDREW PHASE I

OBJECTIVES AND RESULTS

bull DEVELOPED A DATABASE OF PLATFORMS AFFECTED BY ANDREW

bull DEVELOPED AN ULTIMATE CAPACITY ANALYSIS RECIPE amp

PROCEDURE

bull IDENTIFIED DEFICIENCIES IN ULTIMATE CAPACITY ANALYSIS PROCESS

bull DEVELOPED A BAYESIAN UPDATING PROCEDURE

l

bull ESTABLISHED A GLOBAL BIAS FACTOR FOR PLATFORMS

bull PROVIDED EXAMPLES FOR THE USE OF THE BIAS FACTOR

ANDREW PHASE II

PROJECT OBJECTIVES

bull middotUPDATE MODELING AND ANALYSIS RECIPEmiddot

JOINTS

middotBRACES

WAVE PROFILE

FOUNDATIONS

CONDUCTORSmiddot

bull ANALVZE PLATFORMS USING UPDATED RECIPE

PLATFORMS SELECTED FROM FAMILY middotOF EXISTING STRUCTURES USED IN PHASE I

POSSIBLE middotREPLACEMENT IF MORE MEANINGFUL PLATFORM(S) middotFOUND

bull EXPAND BAYESIAN CALIBRATION

TWO BIAS FACTORSmiddot DEVELOPED JACKET AND FOUNDATION

UTILIZE RESULTS OF THE APlMMS FOUNDATION PROJECT

PRODUCE EXAMPLE USING EXPANDED PROCEDURE

-middot

bullbull

SUBSEQUENT PROJECTS

bull MMS INSPECTION PROJECT

bull APlMMS CAISSON PROJECT

bull APlMMS FOUNDATION PROJECT

bull TRIALSBENCHMARK JIP

I I It

ANDREW PHASE II

PROJECT TEAM

bull R L SMITH PROJECT MANAGER

bullbull F middot

J PUSKAR PROJECT ADVISOR

bull Rw LITTON TECHNICAL MANAGER

bull RAGGARWAL TECHNICAL DEVELOPER

bull D K DOLAN TECHNICAL ADVISOR middot

bull C A CORNELL RELIABILITY CONSULTANT

bull F MOSES RELIABILITY CONSULTANT

bull middotMSL ENGINEERING LTD TUBULAR JOINT CONSULTANT

q gtI

ANDREW PHASE II

UPDATE MODELING middotAND ANALYSIS RECIPE

bull CONDUCTOR RESISTANCE

bull WA VE FORCE PROFILE

bull BRACE MODELING

middotbull FOUNDATIONSmiddot

bull JOINT MODELING

~ I I

-ANDREW PHASE II

CONDUCTOR RESISTANCE

shybull RESISTANCE NOT INCLUDED IN PHASE I

bull CONTRIBUTION TO OVERALL RESISTANCE A PHASE II

STUDY ISSUE

middotbull middotCOMPARE FOR 4-LEG AND 8-LEG PLATFORM

bull CONSIDER CONDUCTOR POSmON IN GUIDES

- 13 LEFT 13 CENTER 13 RIGHT

Project ST151K Model Conductor version 1 Mon Nov 14 143140 1994

- ---- ---

--

117 s

111

7i

31

0 -5 -~ - ~()

- iltgt --lo

-(pO

-so

bull bullf OC)

-120

-ISo

-ISO

z

CAgt ~x Singl_e Conductor Model

bull

=I=

-_

-_

-gtshy

~ -

~

I

I~

- gt

~-----------____________________

Project STlSlK Model dir_2 version 7 Mon Nov 14 124200 1994

177 s

Ill

74

37

0- - e -zo - ~o

- Jo

-100

-10

-ISO

z

UV 1-+x

Project STlSlX Model Conductor Version 1 Mon Nov 14 140620 1994

bullmiddot

I

I

I I

I

I I

I I I I I

1 ==

-L

-L - L

-

- L

-

-

L

-

Uniform Displacement step 6 Inelastic Events Legend

Elastic

Strut Buckling Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield Beam Clmn FUlly Plastic Fracture

--

i~~~~~~~~~~~---~--~~-J___--JJ~_middot~~~=-

Project ST151K Model Conductor Version1 Mon Nov 14 142735 1994

---

~

I

I

middot

I

I

I

- shy- shy

- shy

- shy

4x z

CIV Uniform Displacement step 33 Inelastic Events Legend

bull Elasti~ Strut Buckling Strut Residual Strut Reloading Plastic StrutNLTruss Beam Clmn Initial Yield Beam Clmn-Fully Plastic Fracture

gt bull

Project STlSlK middot Model Conductor Version 2 Mon Nov 14 142025 1994

-----------------

bullf

I

I

I

II I I

I I I I

I ==

-

Variable Displacement step 9 Inelastic Events Legend

Elastic Strut Residual

bull

Strut Buckling Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield Beam Clmn Fully Plastic Fracture

--

Project ST151K Model Conductor Version 2 Mon Nov 14 142145 1994

I I I I

I I

I

I

I - - -

-Ishy-

-

Variable Displacement step 35 InelasticEvents Legend

Elastic Strut Buckling Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield Beam Clmn Fully Plastic Fracture

Project ST151K Model Conductormiddot

Version 1

Clgt - Cut Plane Force Fx Mon Nov 14 140051 1994

- -_ vii I fOIZM

0 _

~

N ID 0 bull 0

x 0

middotrt ~ () bull u 0 ~ 0 Ir

N bull

0

0 1 2 3 4 5 X Deck Disp [Ft)

I I 1 1 WbullTI- bullbull

lltcxi lr-C

011lt~~~~~~~---~~~~~~~~L~~~~~~~--~~~~~~~~i~~~~~~~--

Uniform Node Displacement

shy

Ji

r

Comparison of Pushover Curves for ST 151K Variable Load Pattern and Monotonic Load Pattern (Andrew)

Soll Improved 100

6000

5000 -i 4000

f

llS

3000 Ul

J 2000

10 c11 c1lgtolts 1000

0 v

bull ~

f

_ ----- --- ---shy

bull --Variable Lold

- - Andrew Load

I I I I I I

4 5 62 30 1

Deck Dlsplacement (ft)

bullbull r 1 II

~

ANDREW PHASE Il

WA VE FORCE PROFILE

bull DECK WA VE FORCE ~

PHASE I USED SECTION 17 (1056H degCd)

PHASE Il USE REVISED SECTION 17 (1000H Cd)

bull PHASE I USED SINGLE FORCE PROFILE FOR PUSHOVER

- USED WAVE HEIGHT CLOSE TO ULTIMATE CAPACITY

bull PHASE J

Il WILL STUDY THE

USE OF VARIABLE

PROFILE ~

- IF FORCE LEVEL INCREASES (OR DECREASES) BY A FACTOR EQUIVALENT TO A 2 FOOT CHANGE IN WA VE HEIGHT THE INCREMENTAL FO~CE PA1IERN IS CHANGED ACCORDINGLY (CAP FEATURE)

- INITIAL EXAMPLE

f~

a

middotn

a

I N -a a a amiddot

+

~ a

l

a w gt 0c 0 gt a I shyc -CJ wc w gtc(

== c( I shyZ w E W a 0 z- 8

ii sect

~ ~ bullT 0

I if

x ~

bullbullbull

II ~------------________

ANDREW PHASE II

WA VE FORCE PROFILEmiddot

CAP AUTOMATIC APPLICATION FEATURE

STAT

fR1 0 lt f lt1

STAT

STAT

f(R3-R2) middot f lt1

Project ST151K Model Wave_Loading Version 1 rue Nov 8 164107 1994 PT

E~ nss FT

SL 14amp FT

a 111 FT

EL 14 li

E-1 37 FT

El o FT

1middot1middot ______________ _ ________ __ _ __

0

200 400 600 800 1000 12000 Force at Elevatloii (kip)

200

a ll_

Wave Force at Major Elevations for Discrete Wave Heights

180

bullHmiddot40ft

Hmiddot495ft

H bull 51 ft

Hmiddot55ft_

H-59ft I

Andrew (H bull 60 86 ft) II

i I

_11---__

- middot- Hmiddot30ft

-

- bull -

middot bull bull bull middot

---Hmiddot53ft

bullbullbullbullHmiddot57ft

-----H-63ft

Elevation Above Mudllne 100

(ft)

180

140

120

80

60

40

20

Loading at Major Elevations

---H-49511

-deg-Hmiddot608611

0 200 400 600 800 1000 1200

Force It Elevatlon (kip)

--

Base Shear vs Wave Height ST151K

6000

5000 1---------Ult1mateCapacitybull 4890 kip (100 Improved Soll)-----------~--

4000a a lii ~ 3000

81 ampI

2000

1000

0 +-~~~--i~~~~~~~~~-+-~~~--lf--~~~~~~~~-+-~~~--l

30 35 40 45 50 55 60 65

Wave Height (ft)

bull middot

~

bull

II

2

g

~

It)

0

cCD E 0 E a c E euro~ CDshygtItgtmiddoto ~ti c -a Gic CD gt ~

-= c -Cll

I z

bullO 0i i ~ g 0

() CI -bull(q-ll) iuawow llu1wniiaAO

11

shy

116

114

112

g 110

bullc = 108

~ ~ 106 -sect 104

102

100

98

96

-

6555 60

Wave Height vs Center of Action ST151K

30 35 40 45 50

Wave Height (ft)

t

Comparison of Pushover Curves for ST 151K Variable Load Pattern and Monotonic Load Pattern (Andrew)

Soll Improved 100

--- --------- --- ----6000

5000

- 4000 8 - I

1 3000 Ul

81 m 2000

_J -shy

1000

---Variable Load

- - Andrew Load

0 f-~~~~~r--~~~~-t-~~~~~-t-~~~~~+-~~~~----t~~~~~-1

0 1 - 2 3 4 5 6

Deck Displacement (ft)

-middotshy

1 t I I ~

ANDREW PHASE II

BRACE MODELING

bull MARSHALL STRUTS USED FORPHASE I

bull PHASE II WILL ASSESS DIFFERENCE IN USING BUCKLING BEAMS AND MARSHALL STRUTS

middot bull COMPARE FOR 2D FRAME

bull COMPARE FOR PLATFORM 177 WHERE LOWER

BAY EXHIBITS LARGE BRACE MOMENTS

Project andrew Model beamframe Version 1 Mon Nov 14 145447 1994

~

middot Frame Test buckling beam-column model

-

Project andrew Model strutframe Version 1 Mon Nov 14 145754 1994

bull

Frame Test strut model

middot-middot _____________--_

_middotmiddotr_~ --ll--~-middot__

Buckling Beam bull

Strut (K=O 65)

Lateral Displacement

ANDREW PHASE II

FOUNDATIONS

bull ~LICIT NONLINEAR API FOUNDATION USED INiPHASE I

bull FOUNDATIONS

JIP SELECTED MINI-VANE DATA FOR

STRENGTH PROFILE

bull COMMENTS ON BENCHMARK COMPARISONS

- SIGNIFICANT VARIABILITY DUE TO FOUNDATION MODELING METHODS

II Ii

z

ctV 4x SPlO caissonfy-42s soil-api(lf-097step-114)

Inelastic Events Legend Elastic Strut Buckling

Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield

--------shyBeam Clmn Fully Plastic Fracture

~--~~~~~~~~~--~----~__i~~---~~~~__--~

Pro)ect apipile Model South Pelto 10 Version Wed May 18072235 1994 - -

1bullt

+

t

f i bullbullbull

bullbull bullbull

I I

5

~x SPlO Caissonfy=42s soil=api(lf=lOOstep=137)

Inelastic Events Legend

Elastic Strut Bucklilg

Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield

earn Clmn Full~ Plas1c Fracture

ANDREW PHASE II middot middot

JOINT MODELING

middotbull API RIGID-PLASTIC USED IN PHASE I

a ELASTIC-BRITTLE USED IN PHASE IB

amiddot MSL WILL PROVIDE NONLINEAR FORCE-DEFORMATION

AND MOMENT-DEFORMATION RELATIONS FOR PHASE II

a MODELING OPTIONS FOR PHASE II -

l

- UNCO~LEDmiddotONE DIMENSIONAL MEMBER END JOINT MODELING

J Ii

ANDREW PHASE Il

JOINT MODELING OPTIONS DISCUSSION

bull JD COUPLED INTERACTION SURFACE

- IDEAL

- DATA NOT AVAILABLE TO CALIBRATE

bull 2D COUPLED INTERACTION SURFACE

- SIGNIFICANT MODELING IMPROVEMENT

- DATAAVAILABLE

- IMPLEMENTATION PRACTICAL BUT NEEDS DEVELOPMENT AND CALIBRATION

bull 2D UNCOUPLED INTERACTION SURFACE

- SIGNIFICANT MODELING IMPROVEMENT

- DATA AVAILABLE

- IMPLEMENTATION PRACTICAL AND AVAILABLE

ANDREW PHASE II

JOINT MODEJJNG OPTIONS ILLUSTRATIONS

P

e

ANDREW PHASE II

JOINT MODELING PHASE II METHOD

e

I I I

ANDREW PHASE II

ANDREW STORM CONDmONS UPDATED

bull -PHASE I USED OCEANWEATHER HINDCAST-(Nov 1992)

bull middotmiddotPHASE II WILL USE THE REVISED HINDCAST (Nov 1994)

bull lt c lt I lt

2-2

l---4-~~~+-~~-+-~~~1--~~-+-~~--l--~T--+-~~--i24

f-tJVtMfiz~ 1Cfl (r STvb(

Plollol bullbull 5middotNOYmiddot9214bull1J Irbullbull Illbull [ANDREVl~NDREVCUSTbullMAXll 5-NOY-1992 10bull54 - 9208

HEIGHT lml

I~ MAXI

I L 1~middotamp SZ~- ~A I 13c

I J) ( r I X~0iWJ~0gtJ gt L gt gt I 1c I 120

-- N

~ deg

I I I I I I e C I ~=-41 122 -94 -92 -90 -88 -86 -84 -82 -80

-

Figure 7 contours of maximum hindcast significant wave height

_

~

~tl ~

NOltNttIJ 17lt ltNS S iJJ)y

Maximum Significant Wave Height - Hurricane Andrew 1992

30 ~

26 ~ I ~-~~~- == -shy -~ 35-_

24deg1--shy --------~- middot-middot-------middotmiddotmiddot----middot------middot-middot~middot- - - ~-----1---middot

22middot ~~~~~~~-~ -94 -92 -90 -88 -84deg-86 -82 -somiddot

c

-

Platform Selection bullbullbull1111111111111

0 BasisCriteria for Selection middot

0 Review of Phase I Platforms middot 0 Alternative Platforms

0 PMB Recommendations

0 Participants Comments

gt

L

l I

BasisCriteria for Selection bullbullbull1111111111111

0 Benefit for Calibration - Classification (UnexpectedExpected)

- Failure Mode (StructuresFoundations)

- Damaged StructuresKnown response

- Better Representation of Full ~opulation

0 Availability of Data - Existing Models

- Inspection Data

- Damage Data

PREDICTED BEHAVIOR

-

OBSERVED

BEHAVIOR

-SURVIVED DAMAGED FAILED

SURVIVED

ST134W WD90A

-MC311 -

MC397 -

--

-

-

ST151K ST130Q

-

-

-

DAMAGED - -

-

-

--

-

-

ST52 SS139 ST177B -

ST161A

-_

-

FAILED

-

-

-

ST151H ST130A ST72

Operator - Platfonns - Platfonns Survived Damagedamp

- Falled -

bull

-

-

Jle eBDIGIPllDbl -Amocoshy 30 1 Chevron -

143 12 Exxon 72 Mobll 22 1 Phllllps 7 bull

-Shell 39 Unocal 25 1

Total Partlclnts Platfonns 338 15

Other Operators bull- 310 13

-

Total Platfonns - -

648 28

fbml

Platforms Considered 6 7

tie of Total Platfonns 1 25

OVERVIEW OF PLATFORMS SURVIVEDDAMAGEDFAILED

-

shy

v

Platforms Evaluated in Phase I

-

-Platfonn Water Year Number Analysis Avallabillty of Availability Platform

- Name Depth rl of Model DetaDed rl ClassUlcatlon

Installation legs Inspection Soll Report Based on Information Phase I Calibration - -ft

- -- Suaill flalfnrm ~BSi -STISIK 137 1963 8 CAP Yes Yes Un eel Sarvlval

-STIJOO 170 1964 4 CAPmiddotshy Yes ST134data u ST134W 137 1981 4 CAP Yes Yesmiddot E

-

edSarvlval -

middotsurvival

WD90A 184 - 1964 8 CAP Yes E -

- Sarvlvill

MC3ll 343 1978 8 KARMA

MC397 468 1991 4 KARMA -

bull ~reglmmiddot- -

Yes Sure Survival

bull Yes - Sare Sanlval-

-T23STS2 63 1969 4 CAP

T2SSS139 62 1969 4 CAP -

Yes

Yes

Yes Ex

Yes Une

bull SurvivalLikely to DamalI

middot bull SanlvalLlkely to DamalI

ST161A 118 1964 8 USFOS

Eail11a pound1al(bullllm Clss ST177B 142 --1965 8 CAP

STISIH 137 1964 - 8 CAP C

STIJOA 140 19S8 8 CAP - -

Yes

Yes

ollamed

Yes

- Yes E

-

at about 1 mUe

Yes

ST134data

bull SafvlvalLlkelv to Dam- bull- -

E middotrahre

Emeried Failure -

Likely FaUare

T21ST72 61 1969 4 CAP Collamed No Likely FaUure

Platforms Identified for Evaluating Their Likelihood of Falling Under UnexpeCted Survival Category

CllEllATOa

AllB

llLOCll

Sl1lvcnJllE

NAME

YEAR

INSTAUJCD

WATER

IJBPJB

middotmiddotDISTANCE

ftOMSllOllE

(MIUIS)

Ill

(M)

AMOCO BllW

STllI

WDI075

A

bull D

4

n uo 17J

44

32

bull 20

10

bull CllEVllOH STtllO c I 1bull 21 11

STl130 D 0 161 21 11 STllJ4 I 0 137 D - 11

STIUI G I 137 32 11

STIUI I 0 121 32 11

STIUI

J 0 140 32 11

STIUI It 4 137 32 11

STllSI PROl)I 0 137 32 11

STllSI PROD-2 0 137 32 11

STll7 A 0

140 35 11

WDl117 c 65 214 34 10

DXON STshy E 51 14 bull STl165 A t3 bull 34 WD0073

WD0073

WD0073

A _ c

4

62

4

IA IA

17J

22

22

22

bullbullbull WD003-shy E 65 161 Z7

A 0 IN 31 MOBD Bl12I

ssn A bull

50

50

31

7

bull 7

MIJllPllY SS114 B 7 50 7 7

PlllUIPS SSl14t A 55 33 7

SAMEDAN ST17J A 4 117 7 bull 7

SJIELL WD0103 A 65 223 27 WDllOJ bull 65 221 27 WDll04 c 65 221 27 WDOI04 D 67 2 27

UNOCAL

SSl20I

SSl20I

bull SSl20I

SS0209

SS0209

SS0209

SS0209

SS 0253

SS 0253

E

I

PtJllE

A

bull D

G A

c

0

4

4

7J

61

0

0

69

103

97

97

9S

100

95

100

165

175

34

34

34

34

34

34

34

54

54

bullbullbullbullbullbullbullbullbull

Platforms Recommended for Phase II

llalr y_ w- Aot Av- _ I 1 Doplla Model nm bull bull _ bull I I H11 s18-1 Irht lbwl~ bullbullbull(-illtdlel OM lllMtD

II ---

-STISIIC 137 1963 I CAP Yeo Yeo I Yeo

STtshy -170 1964 4 CAP Yeo ST134dlU I Yeo

STl34W 137 Yeo Yeo -1981 4 CAP D Yeo

WD90A 114 1964 I CAP Yeo - D

MC311 30 1971 I KARMA Yeo -sre-

MC397 middot 461 1991 4 ICARMA Yeo Sireshy - - r- -

__ T13ST5l 63 1969 4 CAP Yeo Yeobull middot- ~n- D Yeo _ Tl5SSl39 62 1969 4 CAP Yeo Yeo I Yeo

_ STIT711 142 1965 I CAP Yeo I Yeo middot-1shySTl61A Ill 1964 I USFOS Yeo Yeo - D bulloshy

- r-

STISIH 137 1964 I CAP y D y -_ STl31A 140 1951 I CAP Yeo STl34dlU I Yeo

T21Sll 61 1969 4 CAP No rar-middot I Yeo

~-rmiddot-middot SS209A 95 1967 16 CAP Yeo D -SSU4R

11 UI A 41 -in D SS 2741AIM CAP bull

_ _ -shy - -shy- -shy

CALIBRATION

bull PHASEIAPPROACH

bull CALIBRATION ISSUES

bull PHASE II WORK SCOPE

bull SOFlWARE AVAILABILITY

bull EFFECT OF ADDITIONAL PLATFORMS

bull WEIGHTING APPROACH

bull EXAMPLE APPLICATION middot

(l=Ofce Mode] Wind

~

fcurrent l -middot

~SW ~leldl4

ultimate resistance force I first

design load Illgt

i _

deflection

-

bull

4

Historical Hurrlcane Dariiage and Survival of Platforms

Components Required for Calibration (Petrauskas 1992)

PRIOR Distribution

PLATFORM EXPERIENCE

BAYESIAN Survivals and Failures

UPDATING

Waves Winds and Currents bull

POSTERIOR Distribution

- A_middotJA ~ Survival Data Failure Data

Result Result

Load Resistance

middotfl

-

Bayesian Updaing middotGeneral Description (Petrauskas 1992)

L Imiddot

WHERE TO INTRODUCE BIAS FACTOR (B

On Resistance (Rmue - B (RPREDICreD I

middotOn Loads (Smue middot - B (SPRE01cTEo

On Safety Margin (RSmue - B (RSPREDICTED

CALIBRATION METHODOLOGY

gt

r

FIELD OBSERVATIONS

Observed Behavior

survivals Damages

Failures

r

bull

CAPACITY ANALYSIS

Predicted Load and Resistance Predicted Failure Modes

Survival Damage Failure cases

RELIABILITY ANALYSIS

- Probability of Failure ~

Ukellhood Failure

BAYESIAN UPDATING

Prior Distnbution of B Combined Likelihood Function

Posterior Dlstnbution of B

BIAS FACTOR

Mean Value cov bull

Represents Modeling Uncertainties

Prior Distribution of B Likelihood Functionmiddot Observing Success case STISIK and Posterior Distribution of B

2 00

uo

160

140

120

e 100 L $ uoIshy

0 60 ~ ~ 040

020

0 00

0 00 0 50

_ I 00

b

I 50 2 00

-shy lk [bl STISIK)

--shy Pnor-f(b)

--shy Pos1enor r(bl ISIK)

Prior or B Mean bull 10 COV bull 030

Posterlor of Bbull Meanbull 136COV 0 16

2 50

Likelihood Function and Posterior Distribution Example Platform STISIK

- -

~

- Prior and Posterior Distributions or B Failure cases (ST177B STlSlH ST130A T21) Damage cases (T23 T25 ST161A) Success cases (STlSlK ST130Q ST134W WD90A MC311 MC397)

400

350

300

2 so 0-

I

bull bull

bull bull gt

- bullbull ~ bull bull I bull

_ bull bull bullbull I bull bull bullII V -

bull bull

I bull ~ bull

bullbull bullI bull bullI bull bull bull

middot -~

bull__ ~ middotI middotmiddot

L 2 00

e ~ I SO

-I 00

- 0 so

000 gt

0 00 020 0 40 060 0 80 I 00 120 140 160 110 200

b -

-

shy

Prior-r

- - - - Posterior fLlll failure cases

bull bull bull bull bull bull bull Posterior flall damage cases

- bull bull - bull middot Postenor flall success cases

Posterior flall success fulure damage cases

-

Posterior of B Mean= 119 COV 0101

Posterior Distribution or Blas Factor (8) bull Different Categories and All Cases Combined

shy

l

PHASE I - KEY CONCLUSIONS

bull GLOBAL CORRECTION FACTOR

bull 15 TO 20 CONSERVATISM MEAN)

bull UNEXPECTED SURVIVALS HAVE GREATEST IMPACT

bull DAMAGED CASES MORE INFORMATIVE

bull CALIBRATION ISSUES PHASE II

bull

EFFECT OF IMPROVED CAPACITY ANALYSIS RECIPE

bull APPLICABILITY TO MULTIPLE FAILURE MODES

bull NO FOUNDATION DAMAGE CASE INCLUDED

bull IMPROVED REPRESENTATION OF FULL ANDREW DISTRIBUTION

-bull NUMBER OF PLATFORMS CONSIDERED

I 1 I

I t ) bull ~

PHASE II WORK SCOPE (CALIBRATION)

bull

bull ESTABLISH MULTIPLE BIAS FACTORS

Use Nine Base Case Platforms

middot - Capacity Analysis for Revised recipe

Foundation Blas Factor from Foundation JIP

bull DEVELOP WEIGHTING APPROACH

bull UPDATE MULTIPLE BIAS FACTORS

bullEffect of Platforms Not In the Nine Base Cases

bull EXAMPLE APPLICATION OF BIAS FACTORS I

bullbull

PHASE I PHASE II

CAPACITYbull ANALYSISRu bull

---------------------1----middot

Jc1 - iJbullraquoJ dbull

FttlS frt[f~l ratdhJJ

-

LIKELIHOOD FUNCTION

DEVELOPMENT

lk Jb Jfall1N) bull Jb)

111 Ill Isurvtvoq bull 1 - bl

-middot -------------------shy ~---------------------middot-middot

r111ir8 Cb1 111 Cbl BAYESIAN

UPDATING

~cove

R1R2 bull middot---------------------~----

Jblbull P(g cOJ

1 bull JbRI (MBSI I bull 1

~middot

llcJllJfalluro) bull P(gcOJ Ill 111 b21fallun) bull P ((g1 CO) U 1112 c OD

middot--------------------shy ----------------------shy bull

r111111r8 Cbl 111 Ill i-1nro1

deg ~1b2)laquottfrr (b1b21-lnfo) re Ill bull 1-11middot b2) bull lodegamp

middot

i1 covamp i2 cov~i

-

shy

--

Likelihood Function vs b All Success Case STISI K STl30Q STl34W WD90A MC311 MC397

I 0000

0 9000

OIOOO

0 7000

~ 0amp000

ta osoOO

~ 0 4000

0 3000

0 2000

0 1000

bull 0 0000

bull

I

middot

middot bull I

Ibull

-sibull ~ -middot-middot-middot-middot -middot-

----7----+-----~ I

Ibull bull bull I

rbull

--- Jk ISIK[blsuccess)

- bull - bull - lk 130Q[blsuccess) bull

lk 134W[blsuccess)

Jk-WD90A(blsuccc~I

- - - lk MC397[blsucccss)

--ltgt-- lk middotMC31 l[blsucccss)

--- Jk [bl 6 out of6succcsHascs)

05 I 5 2 250

b

Likelihood Functionsmiddot Success Cases - Phbullsc L

shy

bullbull bull bull bull bull bull -------------------~

bull bull bullbull bullbull bullbull bullbull bullbull bullbull bull

bull bullbull

bullbullbullbull

bullbull

bullbullbullbullbullbullbull

bullbullbullbullbullbull

bullbullbullbullbullbullbull

z 0

t z ) LL

c 0 0c i ~ i

J

~ sect ~--------------~ e l5

I

I

SOFTWARE

bull PF PROGRAM

bull CALREL

bull PF +CALREL

bull RELACS

bull

gt i

l

SENSITIVITY STUDIES

POSTERIOR DISTRIBUTION OF B

bull EFFECT OF CHANGE IN DISTRIBUTION OF PRIOR OF B

bull EFFECT OF REMOVAL OF ONE PLATFORM FROM EACH CATEGORY

bull EFFECT OF REMOVAL OF THREE DAMAGE CASES

bull EFFECT OF ADDITIONAL 13 SIMILAR PLATFORMS

-I

Prior and Posterior Distributions of B Variation In COV of PriorofB fromOl to04 -- -

4500

-4000

-

- -

3500 --middot - bull bull - bull bull Prior-r ICOV =02

3000

- - - - Prior-r ICOV =03 -

bull -~ e 2soo -

bull bull bull bull bull bull bull Pnor-r ICOV =o 4t frJ ~ 2000 Posterior rlCOV =0 2

bull bull Postenor rlCOV =0 3 bull bull 1500 bull middot bull - shy -middot- Posterior rlCOV =04

middot~ I 000 - bull

-middotmiddot ~ ~middotmiddot M -~

- middot 0 500

wbullbullbull _ I middot~ middotmiddotmiddotmiddotmiddot middotmiddotmiddotmiddotmiddot - -~ ~ __ middot_ _ bull -shy0 000

0 000 0200 0 400 0 600 0 800 IOOO 1200 1400 1600 1800 2000 -

- - b

-

=shy

Posterior Distribution of Bias Factor (B) bull Effect of Variation In COV of Prior of B

bull

bullPrior and Posterior Distributions of B Evaluating Effect of Removal of One Platform from Each Category from the Base Case of 13 Platforms

400

~ ~ 350

j IA i

3 00 bull I bull I ---Prior-r I bull l

250 1 I 1

I bull bull ---Posterior r lall 13 casesshy- ~E 2 00 bull bull bull bull bull bull bull Posterior r without STISIK

- ~ 1 bull - bull I - bull bull - bull bull Posterior r Iwithout T2I I 50 I

J-tt~i - bull - bull - Posterior r Iwithout T2S

tOO [-

I I ~ ~ I 0 50

0 00

___ r J i

l~T

~--

0 00 0 20 0 40 0 60 080 I 00 120 140 160 110 2 00

b

zshy

Posterior Distribution of Bias Factor (8) bull Effect of Removal of middotPlatform from Any Category

bullPrior and Posterior Distributions or B Evaluating Effect or Removal or 3 Damage Cases from the Base Case or 13 Platrorms

4

35

3

25

e L 2 ~ bulle shy

I 15

bull I

05

0

0 02 04 06 I 2 I 4 16 IS 2

I I

I

I

I Ibull

J -tmiddot L--

bullI

~

I

fl

II ~

I I v I

middot bull bull

bull L

~ ll- I-~lt

OS I

b

I

bull

1middot bullbullbullbull

~ __

bull

- - - - Prior-r

- - - - Postenor flall failure cases

Postenor flall success cases

--- Posterior flall 13 cases

---Posterior fI success failure cases only

IPosterior or B Mean =t11 COV =0121

-

_I

Posterir Distribution or Bias Factor (B) bull Effect or Removal or 3-Damage Cases

-

bullPrior and Posterior Distributions or B Evaluating Effect or Additional 13 Similar Observations as

Used in Andrew JIP The POsterior or B from Andrew-JIP Considered as New Prior or B

- 500

450

400

350

bull bull

bull bull bull bull J bull

bull bull f ~ I

I ~ Ibull bull

bullbull bull

l bull I -bullbull bull bull

I

I bull bull

bull I I~~bull bull bull

bull ~

I bull bull

~ ~_ ~~~

- - New Prior r bull

Posterior 114 new failure cases ~ 300- bull bull bull bull bull bull bull Posterior f 13 new damage casesl 150 ~

- bull bull - bull bull Posterior 116 new success cases i 2 00

Postenor 1113 new success failure -150 damage cases

I 00

0 50 -

0 00 0 00 0 20 0 40 0 60 0 80 I 00 120 I 40 160 110 2 00

Posterior ol B Mean= 120 COV= 0071 b

shy

shy

_rshy

Posterior Distribution of Bias Factor (B) bull Effect of Additional 13-Similar Observations

1

1 I

WEIGHTING APPROACH bull

bull ASSESSMENT OF PLATFORMS DATA

Select Platforms Most Mappable to the Analyzed Platforms

Unlikely SurvivalDamage Cases Identification

Based on Cut-Off Criteria Agemiddot Hs etc

bull INFORMATION REQUIRED FOR COMPARISON

bull CATEGORIZATION OF ADDITIONALmiddot PLATFORMS

bull UPDATING OF CALIBRATION RESULTS

--

OVERVIEW OF PLATFORMS SURVIVEDDAMAGED -~

Operator -

- Platforms

Survived CS)

Platforms

Damaged CD)

Platforms Used In

Phase IJIP -

JI~ ~Bill~IRBDll -

Amoco -

Chevron -

Exxon

Mob II

Phllllps

Shell

Unocal -

30

143

72 22

7

39 25shy -

1

12

bull

1 bull

1

1(S)+1 (D)

3 (S) +3(D)

2 (S)

bull shy

bull

bull

bull

Total Partlclnants Platforms 338 - 15 10C6S+4D) -

Other Operators

310

13

3 (D)

Total Platforms 648 28 13C6S+7D) -

Summary of Andrew JIP bull Phase I Platforms

Plairshy w v- N-ber Lea-le 1 shy car Hind_ o-w N-bull Deplh fl fl IJpe la bull n Yeadon Jadtet lbelledtf liefalit -- _ -OD

IL H - -shy _ICllshy IL IL - - rQClto ~ _

K STUIK 137 1963 8 365 608S4 Cellardeck

-middot STl-shy 170 1964 4 -~ K 39 62269 Cellardeck - --ST134W 137 1911 4 43 608S4 Smvlnlshy- _ - -

WD90A 184 1964 8 K 267Sbull34 83 SOS31 SUboftllar deck SmvMI

MC311 343 1978 8 x S7 62 - MC397 468 1991 4 x shy S33bull6SO 65911 -shySmeSmvlvol -- gt -

r-- T23ST52 63 1969 4 ton- shyKshy 52594 S03 -T25SSl39 62 1969 4 Kshy SOS94 S063 - middot ton-shy__ -

bullbull a ton--shyST161A 118 1964 8 K - 347S 51495 Cellarclecti

r~

ST1771 142 1965 8 K 39 60167 CeDordeck failme

STISIH 137 1964 8 K 395 608S4 CeDardeck Failme

K ST130A 140 1958 a SS 6095 I lblv Pai1uno - T21ST72 61 1969 4 Kshy S4S94 4974 I lhlv Pai1uno -

- -

JE

INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON

bull WATER DEPTH

bull bull YEAR OF INSTALLATIONDESIGN

bull PLATFORM ORIENTATION

bull PLATFORM LATITUDE AND LONGITUDE

bull KEY SKETCHES OF PLATFORM STRUCTURAL FRAMING

Number of Legs Plies Leg Batter Number of Conductors bull Deck Elevations (Including sub-cellar deck If any) Jacket Framing Elevations Brace type In Vertical Frames Sizes of Primary Members (leg pile vertical frames braces primary horizontals) Joint Details (can gap etc) LegPiie Annulus Grouting Status

bull middotPOST ANDREW INSPECTION REPORT Significant Wave Height Hs (ft)

bull INFORMATION ON DAMAGES PRIOR TO ANDREW

bull RESULTS FROM IN-HOUSE STUDIES IF AVAILABLE

bull ~RIGINAL DESIGN CRITERIA INFORMATION IF AVAILABLE

bull SOIL SHEAR STRENGTH PROFILE IF AVAILABLE

Platforms Identified for Evaluating Therr Likelihood of Falling Under Unexpected Survival Category

OPlRATOR

middotADA

BLOClt

llmucnJllE

NAME

YEAll

INSTAUJD

WA1ER

DBPTll

DISTANCE

JlllOM SHOllE

(MILU)

Bl

(M)

AMOCO 111113

STl11

WDI075

A

bull D

St

17

120

1n

44

20

10 I

CllllVllQfl STlllO c 61 110 21 11 STIUO D Q 1110 21 11

STl134 p Q 137 29 11

ST0151 G 61 137 11

ST0151 I 3 121 11

ST0151 J Q 140 11

STl151 It 137 11

ST0151 PJl001 Q UT 11

ST0151 PROD-2 Q UT 11

ST017 A 3 140 35 11

WDIUT c 65 214 34 10

llXXON STI055 II 51 6T 14 I

STl15 A t3 34

WDI073

WD0073

Amiddotshy Q

1111 1111

22

22

I

I

WDI073 c 1n 22 I

- WDOCl93-shy II

65 1110 2T A 3 194 31 MOBU 111121

SS IOl7

A B

501

50

50

31

T

I

T 6

MtlllPllY SS114 B T 50 T T

PlllllIPS SS0149 A 5 55 33 T

SAMllDAN STln A 107 T T

SBllLL WDl103 A 65 223 2T WDl103 B 65 221 2T WD0104 c 65 221 2T WD0104 D 6T 20 2T

UNOCAL SS020I

SS 020I SSl20I

SS 0209

SS0209

II p

PURll

A

B

3

n 61

103

97

97

95

100

34

34

34

34 34

bullbull bullbullbull

- SS0209 D u 95 34 I

- SS 1209 SSl253

SS0253

G A

c

6T

2

100

165

1T5

34

54

54

bullbull I

I I

I I

SOUTH MARCH

~LAND

N

SOUTH PaTOI ENE ISWltD AREA

MMS Defined Limits ror Offshore Structures Requiring Inspections Following Andrew

t---Jbulln3 T25bull I bullTtl

bull1$1Htl1Abull bull1SIK1779bull

~

middot-DaTA AREA~T

CRANO rsLE

AREA bull1300

bullll4W

=shy

bullPath of Hurricane Andrew with Platforms Used In Calibration

Likelihood Function vs b All Success Case STISI K STIJOQ ST134W WD90A MCJll MCJ97

I 0000

0 9000

0 8000

07000

-Bo6000

tiiii 0 5000 -

04000

0 3000

0 2000

01000

0 0000

0 05 25

ri1rt middot-middot-middot-middot-middot-

I 5 2

b

---11 middot151K(blsu~ccsI

- bull - bull - fk 130Q(bfU~lCSS)

lie bull134W(blsuccc]

lk-WD90A[blsultCS]

- - - lk middotMC397(bfUtlCS]

--o-- lk-MCJll(bbuctcsJ

--- ll (bl 6 out of 6 Mlttcs case]

_

( Likelihood Functions middotSuccess Cases - p - 1 c I

CATEGORIZATION OF ADDITIONAL PLATFORMS

I I I

bull CATEGORIES

Unexpected Survivals

Expected Survivals

bull APPROACH

Identify Corresponding Platform(s) Evaluated In the JIP

Determine Max Wave Height at Platform Location

Evaluate Whether Wave Hit the Deck

middot - Evaluate Likelihood of R and S to Differ from the Base Cases

Review Joint Information

Assess Special Features

bull IDENTIFY NUMBER OF PLATFORMS IN EACH CATEGORY

I

UPDATING OF CALIBRATION RESULTS

bull INCLUDE EFFECT OF ADDITIONAL PLATFORMS ON bullBbull

Consider Increased 1Uncertalntles In R and S

Develop bullNeW Ukellhood Functions for Different Categories

middotLet bullxbull Number of Ukely Unexpected Survival Cases and y Number of Ukely Expected Survival Cases

Determine Combined Likelihood Function lk (b Iadditional platforms) middot

bull REVISED LIKELIHOOD FUNCTION

lk(b I all cases) = lk(b I 9 platforms) middot x lk(b Ix platforms) x lk(b I y platforms)

bull ESTABLISH REVISED BIAS FACTOR B

f8 (b) u f8 (b) lk (b I all cases)

II

t t ~

shy

APPLICATIONS OF BIAS FACTORS

ii APPLICABLE TO PARTICULAR LOAD AND RESISTANCE RECIPE

bull USE BY THE API COMMITTEES AND MMS

Criteria Revisions

Recommended Design Guidelines Revisions bull

bullbull REASSESSMENT STUDIES AND DECISION MAKING

middot

bull RISK ASSESSMENT STUDIES BY THE OPERATORS

Individual Platforms

Fleet of Platforms

middot IMR Program Implementation

bullbull NOT TO BE USED IN DESIGN

I I fl J

EXAMPLE USE OF BIAS FACTORS

bull PHASE I EXAMPLE APPLICATION

bull EVALUATION OF ANNUAL PROBABILITY OF FAILURE

Establish Annual Maximum Seastate Data

Determine Coefficients to Define Base Shear

Determine Ultimate Capacity of Platform

Determine Probablllty of Failure Versus bullbbull

-bull DETERMINE UPDATED PROBABILITY OF FAILURE

Pf - P [S gt R]

Pf I P [~ gt Rb I b] f 8 (b) db

Pr - J P [S gt Rb I b] f bull 8 (b) db

bull TOTAL PROBABILITIES OF FAILURE

Total Pf - 1 bull Il [1 middotPf] - l Pf bull

Total Pf - 1 bull Il [1middot Pf] - l Pfmiddot

Total Pr = 1 - n [1middot PfbullJ = l Pfshy

I I I

EXAMPLE APPLICATION bull PHASE I

PLATFORM ST 151 K

bull GULF OF MEXICO

bull WATER DEPTH = 137 FT

bull ANNUAL MAXIMUM HS MEDIAN 175 ft COV 031

bull ASSUMED STORM DURATION 3 hoursmiddot

bull ASSUMED NUMBER OF WAVES 800

bull MEDIAN HS FOR STORM HOURS 1 2 AND 3

bull MEDIAN CURRENT VALUE bull

bull DETERMINE ULTIMATE CAPACITY FOR THREE DIRECTIONS

bull DETERMINE ANNUAL FAILURE PROBABILITIES

Broadsidemiddot Diagonal End-on

0037 0025 0007

0071 0044 0018

0022 0016 0004

DETERMINE TOTAL ANNUAL PROBABILITY OF FAILURE

Pf - 007

Pf - 013

Pr - 004

bull

EXAMPLE USE OF BIAS FACTORS - PHASE II middot

bull DETERMINE PREDICTIVE PROBABILITY OF FAILURE

Case - 1

Pf _ S S Pf (b b2) fa1a2 (b b2) db db2

g bull -1 (b1 R I S) - 1

92 - (b2 R2 I S) - 1

Case -2

Pf s ~ [bJ f11 (b) db

Pf [bJ p [g lt 0)

g - bull 1 (b1 R1 I S) - 1

1

bull1) l

I

BENEFITS OF ANDREW JIP bull PHASE II

lt

bull IMPROVED LOAD AND RESISTANCE RECIPE

bull IMPROVED CALIBRATION PROCEDURE

bull MULTIPLE BIAS FACTORS

bull PROCEDURE TO INCLUDE EFFECT OF OTHER PLATFORMS

bull PROCEDURE TO USE MULTIPLE BIAS FACTORS

  • Structure Bookmarks
    • ANDREW PHASE II JOINT1 INDUSTRY PROJECT
    • PROGRESS MEETING NUMBER 1
    • ANDREW PHASE II JOINT INDUSTRY PROJECT PROGRESS MEETING NUMBER 1 NOVEMBER 16 AGENDA
    • ANDREW PHASE II UST OF PARTICIPANTS
    • ANDREW PHASE I OBJECTIVES AND RESULTS
    • ANDREW PHASE II PROJECT OBJECTIVES
    • SUBSEQUENT PROJECTS
    • ANDREW PHASE II PROJECT TEAM
    • ANDREW PHASE II UPDATE MODELING middotAND ANALYSIS RECIPE
    • -ANDREW PHASE II CONDUCTOR RESISTANCE
    • ANDREW PHASE Il WA VE FORCE PROFILE
    • ANDREW PHASE II WA VE FORCE PROFILEmiddot
    • ANDREW PHASE II BRACE MODELING
    • ANDREW PHASE II FOUNDATIONS
    • ANDREW PHASE II middot middot JOINT MODELING
    • ANDREW PHASE Il JOINT MODELING OPTIONS DISCUSSION
    • ANDREW PHASE II JOINT MODEJJNG OPTIONS ILLUSTRATIONS
    • ANDREW PHASE II JOINT MODELING PHASE II METHOD
    • ANDREW PHASE II ANDREW STORM CONDmONS UPDATED
    • Platform Selection bullbullbull1111111111111
    • BasisCriteria for Selection bullbullbull1111111111111
    • CALIBRATION
    • WHERE TO INTRODUCE BIAS FACTOR (B
    • PHASE I -KEY CONCLUSIONS
    • bull CALIBRATION ISSUES PHASE II
    • PHASE II WORK SCOPE (CALIBRATION)
    • SOFTWARE
    • SENSITIVITY STUDIES POSTERIOR DISTRIBUTION OF B
    • WEIGHTING APPROACH bull
    • INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON
    • CATEGORIZATION OF ADDITIONAL PLATFORMS
    • UPDATING OF CALIBRATION RESULTS
    • APPLICATIONS OF BIAS FACTORS
    • EXAMPLE USE OF BIAS FACTORS
    • EXAMPLE APPLICATION bull PHASE I PLATFORM ST 151 K
    • EXAMPLE USE OF BIAS FACTORS -PHASE II middot
    • BENEFITS OF ANDREW JIP bull PHASE II
Page 5:  · . ' ANDREW PHASE II PROJECT OBJECTIVES . ' • ·UPDATE MODELING AND ANALYSIS RECIPE· JOINTS ' ' ·BRACES WAVE PROFILE FOUNDATIONS . ' ' CONDUCTORS· • ANALVZE PLATFORMS USING

ANDREW PHASE II

PROJECT OBJECTIVES

bull middotUPDATE MODELING AND ANALYSIS RECIPEmiddot

JOINTS

middotBRACES

WAVE PROFILE

FOUNDATIONS

CONDUCTORSmiddot

bull ANALVZE PLATFORMS USING UPDATED RECIPE

PLATFORMS SELECTED FROM FAMILY middotOF EXISTING STRUCTURES USED IN PHASE I

POSSIBLE middotREPLACEMENT IF MORE MEANINGFUL PLATFORM(S) middotFOUND

bull EXPAND BAYESIAN CALIBRATION

TWO BIAS FACTORSmiddot DEVELOPED JACKET AND FOUNDATION

UTILIZE RESULTS OF THE APlMMS FOUNDATION PROJECT

PRODUCE EXAMPLE USING EXPANDED PROCEDURE

-middot

bullbull

SUBSEQUENT PROJECTS

bull MMS INSPECTION PROJECT

bull APlMMS CAISSON PROJECT

bull APlMMS FOUNDATION PROJECT

bull TRIALSBENCHMARK JIP

I I It

ANDREW PHASE II

PROJECT TEAM

bull R L SMITH PROJECT MANAGER

bullbull F middot

J PUSKAR PROJECT ADVISOR

bull Rw LITTON TECHNICAL MANAGER

bull RAGGARWAL TECHNICAL DEVELOPER

bull D K DOLAN TECHNICAL ADVISOR middot

bull C A CORNELL RELIABILITY CONSULTANT

bull F MOSES RELIABILITY CONSULTANT

bull middotMSL ENGINEERING LTD TUBULAR JOINT CONSULTANT

q gtI

ANDREW PHASE II

UPDATE MODELING middotAND ANALYSIS RECIPE

bull CONDUCTOR RESISTANCE

bull WA VE FORCE PROFILE

bull BRACE MODELING

middotbull FOUNDATIONSmiddot

bull JOINT MODELING

~ I I

-ANDREW PHASE II

CONDUCTOR RESISTANCE

shybull RESISTANCE NOT INCLUDED IN PHASE I

bull CONTRIBUTION TO OVERALL RESISTANCE A PHASE II

STUDY ISSUE

middotbull middotCOMPARE FOR 4-LEG AND 8-LEG PLATFORM

bull CONSIDER CONDUCTOR POSmON IN GUIDES

- 13 LEFT 13 CENTER 13 RIGHT

Project ST151K Model Conductor version 1 Mon Nov 14 143140 1994

- ---- ---

--

117 s

111

7i

31

0 -5 -~ - ~()

- iltgt --lo

-(pO

-so

bull bullf OC)

-120

-ISo

-ISO

z

CAgt ~x Singl_e Conductor Model

bull

=I=

-_

-_

-gtshy

~ -

~

I

I~

- gt

~-----------____________________

Project STlSlK Model dir_2 version 7 Mon Nov 14 124200 1994

177 s

Ill

74

37

0- - e -zo - ~o

- Jo

-100

-10

-ISO

z

UV 1-+x

Project STlSlX Model Conductor Version 1 Mon Nov 14 140620 1994

bullmiddot

I

I

I I

I

I I

I I I I I

1 ==

-L

-L - L

-

- L

-

-

L

-

Uniform Displacement step 6 Inelastic Events Legend

Elastic

Strut Buckling Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield Beam Clmn FUlly Plastic Fracture

--

i~~~~~~~~~~~---~--~~-J___--JJ~_middot~~~=-

Project ST151K Model Conductor Version1 Mon Nov 14 142735 1994

---

~

I

I

middot

I

I

I

- shy- shy

- shy

- shy

4x z

CIV Uniform Displacement step 33 Inelastic Events Legend

bull Elasti~ Strut Buckling Strut Residual Strut Reloading Plastic StrutNLTruss Beam Clmn Initial Yield Beam Clmn-Fully Plastic Fracture

gt bull

Project STlSlK middot Model Conductor Version 2 Mon Nov 14 142025 1994

-----------------

bullf

I

I

I

II I I

I I I I

I ==

-

Variable Displacement step 9 Inelastic Events Legend

Elastic Strut Residual

bull

Strut Buckling Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield Beam Clmn Fully Plastic Fracture

--

Project ST151K Model Conductor Version 2 Mon Nov 14 142145 1994

I I I I

I I

I

I

I - - -

-Ishy-

-

Variable Displacement step 35 InelasticEvents Legend

Elastic Strut Buckling Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield Beam Clmn Fully Plastic Fracture

Project ST151K Model Conductormiddot

Version 1

Clgt - Cut Plane Force Fx Mon Nov 14 140051 1994

- -_ vii I fOIZM

0 _

~

N ID 0 bull 0

x 0

middotrt ~ () bull u 0 ~ 0 Ir

N bull

0

0 1 2 3 4 5 X Deck Disp [Ft)

I I 1 1 WbullTI- bullbull

lltcxi lr-C

011lt~~~~~~~---~~~~~~~~L~~~~~~~--~~~~~~~~i~~~~~~~--

Uniform Node Displacement

shy

Ji

r

Comparison of Pushover Curves for ST 151K Variable Load Pattern and Monotonic Load Pattern (Andrew)

Soll Improved 100

6000

5000 -i 4000

f

llS

3000 Ul

J 2000

10 c11 c1lgtolts 1000

0 v

bull ~

f

_ ----- --- ---shy

bull --Variable Lold

- - Andrew Load

I I I I I I

4 5 62 30 1

Deck Dlsplacement (ft)

bullbull r 1 II

~

ANDREW PHASE Il

WA VE FORCE PROFILE

bull DECK WA VE FORCE ~

PHASE I USED SECTION 17 (1056H degCd)

PHASE Il USE REVISED SECTION 17 (1000H Cd)

bull PHASE I USED SINGLE FORCE PROFILE FOR PUSHOVER

- USED WAVE HEIGHT CLOSE TO ULTIMATE CAPACITY

bull PHASE J

Il WILL STUDY THE

USE OF VARIABLE

PROFILE ~

- IF FORCE LEVEL INCREASES (OR DECREASES) BY A FACTOR EQUIVALENT TO A 2 FOOT CHANGE IN WA VE HEIGHT THE INCREMENTAL FO~CE PA1IERN IS CHANGED ACCORDINGLY (CAP FEATURE)

- INITIAL EXAMPLE

f~

a

middotn

a

I N -a a a amiddot

+

~ a

l

a w gt 0c 0 gt a I shyc -CJ wc w gtc(

== c( I shyZ w E W a 0 z- 8

ii sect

~ ~ bullT 0

I if

x ~

bullbullbull

II ~------------________

ANDREW PHASE II

WA VE FORCE PROFILEmiddot

CAP AUTOMATIC APPLICATION FEATURE

STAT

fR1 0 lt f lt1

STAT

STAT

f(R3-R2) middot f lt1

Project ST151K Model Wave_Loading Version 1 rue Nov 8 164107 1994 PT

E~ nss FT

SL 14amp FT

a 111 FT

EL 14 li

E-1 37 FT

El o FT

1middot1middot ______________ _ ________ __ _ __

0

200 400 600 800 1000 12000 Force at Elevatloii (kip)

200

a ll_

Wave Force at Major Elevations for Discrete Wave Heights

180

bullHmiddot40ft

Hmiddot495ft

H bull 51 ft

Hmiddot55ft_

H-59ft I

Andrew (H bull 60 86 ft) II

i I

_11---__

- middot- Hmiddot30ft

-

- bull -

middot bull bull bull middot

---Hmiddot53ft

bullbullbullbullHmiddot57ft

-----H-63ft

Elevation Above Mudllne 100

(ft)

180

140

120

80

60

40

20

Loading at Major Elevations

---H-49511

-deg-Hmiddot608611

0 200 400 600 800 1000 1200

Force It Elevatlon (kip)

--

Base Shear vs Wave Height ST151K

6000

5000 1---------Ult1mateCapacitybull 4890 kip (100 Improved Soll)-----------~--

4000a a lii ~ 3000

81 ampI

2000

1000

0 +-~~~--i~~~~~~~~~-+-~~~--lf--~~~~~~~~-+-~~~--l

30 35 40 45 50 55 60 65

Wave Height (ft)

bull middot

~

bull

II

2

g

~

It)

0

cCD E 0 E a c E euro~ CDshygtItgtmiddoto ~ti c -a Gic CD gt ~

-= c -Cll

I z

bullO 0i i ~ g 0

() CI -bull(q-ll) iuawow llu1wniiaAO

11

shy

116

114

112

g 110

bullc = 108

~ ~ 106 -sect 104

102

100

98

96

-

6555 60

Wave Height vs Center of Action ST151K

30 35 40 45 50

Wave Height (ft)

t

Comparison of Pushover Curves for ST 151K Variable Load Pattern and Monotonic Load Pattern (Andrew)

Soll Improved 100

--- --------- --- ----6000

5000

- 4000 8 - I

1 3000 Ul

81 m 2000

_J -shy

1000

---Variable Load

- - Andrew Load

0 f-~~~~~r--~~~~-t-~~~~~-t-~~~~~+-~~~~----t~~~~~-1

0 1 - 2 3 4 5 6

Deck Displacement (ft)

-middotshy

1 t I I ~

ANDREW PHASE II

BRACE MODELING

bull MARSHALL STRUTS USED FORPHASE I

bull PHASE II WILL ASSESS DIFFERENCE IN USING BUCKLING BEAMS AND MARSHALL STRUTS

middot bull COMPARE FOR 2D FRAME

bull COMPARE FOR PLATFORM 177 WHERE LOWER

BAY EXHIBITS LARGE BRACE MOMENTS

Project andrew Model beamframe Version 1 Mon Nov 14 145447 1994

~

middot Frame Test buckling beam-column model

-

Project andrew Model strutframe Version 1 Mon Nov 14 145754 1994

bull

Frame Test strut model

middot-middot _____________--_

_middotmiddotr_~ --ll--~-middot__

Buckling Beam bull

Strut (K=O 65)

Lateral Displacement

ANDREW PHASE II

FOUNDATIONS

bull ~LICIT NONLINEAR API FOUNDATION USED INiPHASE I

bull FOUNDATIONS

JIP SELECTED MINI-VANE DATA FOR

STRENGTH PROFILE

bull COMMENTS ON BENCHMARK COMPARISONS

- SIGNIFICANT VARIABILITY DUE TO FOUNDATION MODELING METHODS

II Ii

z

ctV 4x SPlO caissonfy-42s soil-api(lf-097step-114)

Inelastic Events Legend Elastic Strut Buckling

Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield

--------shyBeam Clmn Fully Plastic Fracture

~--~~~~~~~~~--~----~__i~~---~~~~__--~

Pro)ect apipile Model South Pelto 10 Version Wed May 18072235 1994 - -

1bullt

+

t

f i bullbullbull

bullbull bullbull

I I

5

~x SPlO Caissonfy=42s soil=api(lf=lOOstep=137)

Inelastic Events Legend

Elastic Strut Bucklilg

Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield

earn Clmn Full~ Plas1c Fracture

ANDREW PHASE II middot middot

JOINT MODELING

middotbull API RIGID-PLASTIC USED IN PHASE I

a ELASTIC-BRITTLE USED IN PHASE IB

amiddot MSL WILL PROVIDE NONLINEAR FORCE-DEFORMATION

AND MOMENT-DEFORMATION RELATIONS FOR PHASE II

a MODELING OPTIONS FOR PHASE II -

l

- UNCO~LEDmiddotONE DIMENSIONAL MEMBER END JOINT MODELING

J Ii

ANDREW PHASE Il

JOINT MODELING OPTIONS DISCUSSION

bull JD COUPLED INTERACTION SURFACE

- IDEAL

- DATA NOT AVAILABLE TO CALIBRATE

bull 2D COUPLED INTERACTION SURFACE

- SIGNIFICANT MODELING IMPROVEMENT

- DATAAVAILABLE

- IMPLEMENTATION PRACTICAL BUT NEEDS DEVELOPMENT AND CALIBRATION

bull 2D UNCOUPLED INTERACTION SURFACE

- SIGNIFICANT MODELING IMPROVEMENT

- DATA AVAILABLE

- IMPLEMENTATION PRACTICAL AND AVAILABLE

ANDREW PHASE II

JOINT MODEJJNG OPTIONS ILLUSTRATIONS

P

e

ANDREW PHASE II

JOINT MODELING PHASE II METHOD

e

I I I

ANDREW PHASE II

ANDREW STORM CONDmONS UPDATED

bull -PHASE I USED OCEANWEATHER HINDCAST-(Nov 1992)

bull middotmiddotPHASE II WILL USE THE REVISED HINDCAST (Nov 1994)

bull lt c lt I lt

2-2

l---4-~~~+-~~-+-~~~1--~~-+-~~--l--~T--+-~~--i24

f-tJVtMfiz~ 1Cfl (r STvb(

Plollol bullbull 5middotNOYmiddot9214bull1J Irbullbull Illbull [ANDREVl~NDREVCUSTbullMAXll 5-NOY-1992 10bull54 - 9208

HEIGHT lml

I~ MAXI

I L 1~middotamp SZ~- ~A I 13c

I J) ( r I X~0iWJ~0gtJ gt L gt gt I 1c I 120

-- N

~ deg

I I I I I I e C I ~=-41 122 -94 -92 -90 -88 -86 -84 -82 -80

-

Figure 7 contours of maximum hindcast significant wave height

_

~

~tl ~

NOltNttIJ 17lt ltNS S iJJ)y

Maximum Significant Wave Height - Hurricane Andrew 1992

30 ~

26 ~ I ~-~~~- == -shy -~ 35-_

24deg1--shy --------~- middot-middot-------middotmiddotmiddot----middot------middot-middot~middot- - - ~-----1---middot

22middot ~~~~~~~-~ -94 -92 -90 -88 -84deg-86 -82 -somiddot

c

-

Platform Selection bullbullbull1111111111111

0 BasisCriteria for Selection middot

0 Review of Phase I Platforms middot 0 Alternative Platforms

0 PMB Recommendations

0 Participants Comments

gt

L

l I

BasisCriteria for Selection bullbullbull1111111111111

0 Benefit for Calibration - Classification (UnexpectedExpected)

- Failure Mode (StructuresFoundations)

- Damaged StructuresKnown response

- Better Representation of Full ~opulation

0 Availability of Data - Existing Models

- Inspection Data

- Damage Data

PREDICTED BEHAVIOR

-

OBSERVED

BEHAVIOR

-SURVIVED DAMAGED FAILED

SURVIVED

ST134W WD90A

-MC311 -

MC397 -

--

-

-

ST151K ST130Q

-

-

-

DAMAGED - -

-

-

--

-

-

ST52 SS139 ST177B -

ST161A

-_

-

FAILED

-

-

-

ST151H ST130A ST72

Operator - Platfonns - Platfonns Survived Damagedamp

- Falled -

bull

-

-

Jle eBDIGIPllDbl -Amocoshy 30 1 Chevron -

143 12 Exxon 72 Mobll 22 1 Phllllps 7 bull

-Shell 39 Unocal 25 1

Total Partlclnts Platfonns 338 15

Other Operators bull- 310 13

-

Total Platfonns - -

648 28

fbml

Platforms Considered 6 7

tie of Total Platfonns 1 25

OVERVIEW OF PLATFORMS SURVIVEDDAMAGEDFAILED

-

shy

v

Platforms Evaluated in Phase I

-

-Platfonn Water Year Number Analysis Avallabillty of Availability Platform

- Name Depth rl of Model DetaDed rl ClassUlcatlon

Installation legs Inspection Soll Report Based on Information Phase I Calibration - -ft

- -- Suaill flalfnrm ~BSi -STISIK 137 1963 8 CAP Yes Yes Un eel Sarvlval

-STIJOO 170 1964 4 CAPmiddotshy Yes ST134data u ST134W 137 1981 4 CAP Yes Yesmiddot E

-

edSarvlval -

middotsurvival

WD90A 184 - 1964 8 CAP Yes E -

- Sarvlvill

MC3ll 343 1978 8 KARMA

MC397 468 1991 4 KARMA -

bull ~reglmmiddot- -

Yes Sure Survival

bull Yes - Sare Sanlval-

-T23STS2 63 1969 4 CAP

T2SSS139 62 1969 4 CAP -

Yes

Yes

Yes Ex

Yes Une

bull SurvivalLikely to DamalI

middot bull SanlvalLlkely to DamalI

ST161A 118 1964 8 USFOS

Eail11a pound1al(bullllm Clss ST177B 142 --1965 8 CAP

STISIH 137 1964 - 8 CAP C

STIJOA 140 19S8 8 CAP - -

Yes

Yes

ollamed

Yes

- Yes E

-

at about 1 mUe

Yes

ST134data

bull SafvlvalLlkelv to Dam- bull- -

E middotrahre

Emeried Failure -

Likely FaUare

T21ST72 61 1969 4 CAP Collamed No Likely FaUure

Platforms Identified for Evaluating Their Likelihood of Falling Under UnexpeCted Survival Category

CllEllATOa

AllB

llLOCll

Sl1lvcnJllE

NAME

YEAR

INSTAUJCD

WATER

IJBPJB

middotmiddotDISTANCE

ftOMSllOllE

(MIUIS)

Ill

(M)

AMOCO BllW

STllI

WDI075

A

bull D

4

n uo 17J

44

32

bull 20

10

bull CllEVllOH STtllO c I 1bull 21 11

STl130 D 0 161 21 11 STllJ4 I 0 137 D - 11

STIUI G I 137 32 11

STIUI I 0 121 32 11

STIUI

J 0 140 32 11

STIUI It 4 137 32 11

STllSI PROl)I 0 137 32 11

STllSI PROD-2 0 137 32 11

STll7 A 0

140 35 11

WDl117 c 65 214 34 10

DXON STshy E 51 14 bull STl165 A t3 bull 34 WD0073

WD0073

WD0073

A _ c

4

62

4

IA IA

17J

22

22

22

bullbullbull WD003-shy E 65 161 Z7

A 0 IN 31 MOBD Bl12I

ssn A bull

50

50

31

7

bull 7

MIJllPllY SS114 B 7 50 7 7

PlllUIPS SSl14t A 55 33 7

SAMEDAN ST17J A 4 117 7 bull 7

SJIELL WD0103 A 65 223 27 WDllOJ bull 65 221 27 WDll04 c 65 221 27 WDOI04 D 67 2 27

UNOCAL

SSl20I

SSl20I

bull SSl20I

SS0209

SS0209

SS0209

SS0209

SS 0253

SS 0253

E

I

PtJllE

A

bull D

G A

c

0

4

4

7J

61

0

0

69

103

97

97

9S

100

95

100

165

175

34

34

34

34

34

34

34

54

54

bullbullbullbullbullbullbullbullbull

Platforms Recommended for Phase II

llalr y_ w- Aot Av- _ I 1 Doplla Model nm bull bull _ bull I I H11 s18-1 Irht lbwl~ bullbullbull(-illtdlel OM lllMtD

II ---

-STISIIC 137 1963 I CAP Yeo Yeo I Yeo

STtshy -170 1964 4 CAP Yeo ST134dlU I Yeo

STl34W 137 Yeo Yeo -1981 4 CAP D Yeo

WD90A 114 1964 I CAP Yeo - D

MC311 30 1971 I KARMA Yeo -sre-

MC397 middot 461 1991 4 ICARMA Yeo Sireshy - - r- -

__ T13ST5l 63 1969 4 CAP Yeo Yeobull middot- ~n- D Yeo _ Tl5SSl39 62 1969 4 CAP Yeo Yeo I Yeo

_ STIT711 142 1965 I CAP Yeo I Yeo middot-1shySTl61A Ill 1964 I USFOS Yeo Yeo - D bulloshy

- r-

STISIH 137 1964 I CAP y D y -_ STl31A 140 1951 I CAP Yeo STl34dlU I Yeo

T21Sll 61 1969 4 CAP No rar-middot I Yeo

~-rmiddot-middot SS209A 95 1967 16 CAP Yeo D -SSU4R

11 UI A 41 -in D SS 2741AIM CAP bull

_ _ -shy - -shy- -shy

CALIBRATION

bull PHASEIAPPROACH

bull CALIBRATION ISSUES

bull PHASE II WORK SCOPE

bull SOFlWARE AVAILABILITY

bull EFFECT OF ADDITIONAL PLATFORMS

bull WEIGHTING APPROACH

bull EXAMPLE APPLICATION middot

(l=Ofce Mode] Wind

~

fcurrent l -middot

~SW ~leldl4

ultimate resistance force I first

design load Illgt

i _

deflection

-

bull

4

Historical Hurrlcane Dariiage and Survival of Platforms

Components Required for Calibration (Petrauskas 1992)

PRIOR Distribution

PLATFORM EXPERIENCE

BAYESIAN Survivals and Failures

UPDATING

Waves Winds and Currents bull

POSTERIOR Distribution

- A_middotJA ~ Survival Data Failure Data

Result Result

Load Resistance

middotfl

-

Bayesian Updaing middotGeneral Description (Petrauskas 1992)

L Imiddot

WHERE TO INTRODUCE BIAS FACTOR (B

On Resistance (Rmue - B (RPREDICreD I

middotOn Loads (Smue middot - B (SPRE01cTEo

On Safety Margin (RSmue - B (RSPREDICTED

CALIBRATION METHODOLOGY

gt

r

FIELD OBSERVATIONS

Observed Behavior

survivals Damages

Failures

r

bull

CAPACITY ANALYSIS

Predicted Load and Resistance Predicted Failure Modes

Survival Damage Failure cases

RELIABILITY ANALYSIS

- Probability of Failure ~

Ukellhood Failure

BAYESIAN UPDATING

Prior Distnbution of B Combined Likelihood Function

Posterior Dlstnbution of B

BIAS FACTOR

Mean Value cov bull

Represents Modeling Uncertainties

Prior Distribution of B Likelihood Functionmiddot Observing Success case STISIK and Posterior Distribution of B

2 00

uo

160

140

120

e 100 L $ uoIshy

0 60 ~ ~ 040

020

0 00

0 00 0 50

_ I 00

b

I 50 2 00

-shy lk [bl STISIK)

--shy Pnor-f(b)

--shy Pos1enor r(bl ISIK)

Prior or B Mean bull 10 COV bull 030

Posterlor of Bbull Meanbull 136COV 0 16

2 50

Likelihood Function and Posterior Distribution Example Platform STISIK

- -

~

- Prior and Posterior Distributions or B Failure cases (ST177B STlSlH ST130A T21) Damage cases (T23 T25 ST161A) Success cases (STlSlK ST130Q ST134W WD90A MC311 MC397)

400

350

300

2 so 0-

I

bull bull

bull bull gt

- bullbull ~ bull bull I bull

_ bull bull bullbull I bull bull bullII V -

bull bull

I bull ~ bull

bullbull bullI bull bullI bull bull bull

middot -~

bull__ ~ middotI middotmiddot

L 2 00

e ~ I SO

-I 00

- 0 so

000 gt

0 00 020 0 40 060 0 80 I 00 120 140 160 110 200

b -

-

shy

Prior-r

- - - - Posterior fLlll failure cases

bull bull bull bull bull bull bull Posterior flall damage cases

- bull bull - bull middot Postenor flall success cases

Posterior flall success fulure damage cases

-

Posterior of B Mean= 119 COV 0101

Posterior Distribution or Blas Factor (8) bull Different Categories and All Cases Combined

shy

l

PHASE I - KEY CONCLUSIONS

bull GLOBAL CORRECTION FACTOR

bull 15 TO 20 CONSERVATISM MEAN)

bull UNEXPECTED SURVIVALS HAVE GREATEST IMPACT

bull DAMAGED CASES MORE INFORMATIVE

bull CALIBRATION ISSUES PHASE II

bull

EFFECT OF IMPROVED CAPACITY ANALYSIS RECIPE

bull APPLICABILITY TO MULTIPLE FAILURE MODES

bull NO FOUNDATION DAMAGE CASE INCLUDED

bull IMPROVED REPRESENTATION OF FULL ANDREW DISTRIBUTION

-bull NUMBER OF PLATFORMS CONSIDERED

I 1 I

I t ) bull ~

PHASE II WORK SCOPE (CALIBRATION)

bull

bull ESTABLISH MULTIPLE BIAS FACTORS

Use Nine Base Case Platforms

middot - Capacity Analysis for Revised recipe

Foundation Blas Factor from Foundation JIP

bull DEVELOP WEIGHTING APPROACH

bull UPDATE MULTIPLE BIAS FACTORS

bullEffect of Platforms Not In the Nine Base Cases

bull EXAMPLE APPLICATION OF BIAS FACTORS I

bullbull

PHASE I PHASE II

CAPACITYbull ANALYSISRu bull

---------------------1----middot

Jc1 - iJbullraquoJ dbull

FttlS frt[f~l ratdhJJ

-

LIKELIHOOD FUNCTION

DEVELOPMENT

lk Jb Jfall1N) bull Jb)

111 Ill Isurvtvoq bull 1 - bl

-middot -------------------shy ~---------------------middot-middot

r111ir8 Cb1 111 Cbl BAYESIAN

UPDATING

~cove

R1R2 bull middot---------------------~----

Jblbull P(g cOJ

1 bull JbRI (MBSI I bull 1

~middot

llcJllJfalluro) bull P(gcOJ Ill 111 b21fallun) bull P ((g1 CO) U 1112 c OD

middot--------------------shy ----------------------shy bull

r111111r8 Cbl 111 Ill i-1nro1

deg ~1b2)laquottfrr (b1b21-lnfo) re Ill bull 1-11middot b2) bull lodegamp

middot

i1 covamp i2 cov~i

-

shy

--

Likelihood Function vs b All Success Case STISI K STl30Q STl34W WD90A MC311 MC397

I 0000

0 9000

OIOOO

0 7000

~ 0amp000

ta osoOO

~ 0 4000

0 3000

0 2000

0 1000

bull 0 0000

bull

I

middot

middot bull I

Ibull

-sibull ~ -middot-middot-middot-middot -middot-

----7----+-----~ I

Ibull bull bull I

rbull

--- Jk ISIK[blsuccess)

- bull - bull - lk 130Q[blsuccess) bull

lk 134W[blsuccess)

Jk-WD90A(blsuccc~I

- - - lk MC397[blsucccss)

--ltgt-- lk middotMC31 l[blsucccss)

--- Jk [bl 6 out of6succcsHascs)

05 I 5 2 250

b

Likelihood Functionsmiddot Success Cases - Phbullsc L

shy

bullbull bull bull bull bull bull -------------------~

bull bull bullbull bullbull bullbull bullbull bullbull bullbull bull

bull bullbull

bullbullbullbull

bullbull

bullbullbullbullbullbullbull

bullbullbullbullbullbull

bullbullbullbullbullbullbull

z 0

t z ) LL

c 0 0c i ~ i

J

~ sect ~--------------~ e l5

I

I

SOFTWARE

bull PF PROGRAM

bull CALREL

bull PF +CALREL

bull RELACS

bull

gt i

l

SENSITIVITY STUDIES

POSTERIOR DISTRIBUTION OF B

bull EFFECT OF CHANGE IN DISTRIBUTION OF PRIOR OF B

bull EFFECT OF REMOVAL OF ONE PLATFORM FROM EACH CATEGORY

bull EFFECT OF REMOVAL OF THREE DAMAGE CASES

bull EFFECT OF ADDITIONAL 13 SIMILAR PLATFORMS

-I

Prior and Posterior Distributions of B Variation In COV of PriorofB fromOl to04 -- -

4500

-4000

-

- -

3500 --middot - bull bull - bull bull Prior-r ICOV =02

3000

- - - - Prior-r ICOV =03 -

bull -~ e 2soo -

bull bull bull bull bull bull bull Pnor-r ICOV =o 4t frJ ~ 2000 Posterior rlCOV =0 2

bull bull Postenor rlCOV =0 3 bull bull 1500 bull middot bull - shy -middot- Posterior rlCOV =04

middot~ I 000 - bull

-middotmiddot ~ ~middotmiddot M -~

- middot 0 500

wbullbullbull _ I middot~ middotmiddotmiddotmiddotmiddot middotmiddotmiddotmiddotmiddot - -~ ~ __ middot_ _ bull -shy0 000

0 000 0200 0 400 0 600 0 800 IOOO 1200 1400 1600 1800 2000 -

- - b

-

=shy

Posterior Distribution of Bias Factor (B) bull Effect of Variation In COV of Prior of B

bull

bullPrior and Posterior Distributions of B Evaluating Effect of Removal of One Platform from Each Category from the Base Case of 13 Platforms

400

~ ~ 350

j IA i

3 00 bull I bull I ---Prior-r I bull l

250 1 I 1

I bull bull ---Posterior r lall 13 casesshy- ~E 2 00 bull bull bull bull bull bull bull Posterior r without STISIK

- ~ 1 bull - bull I - bull bull - bull bull Posterior r Iwithout T2I I 50 I

J-tt~i - bull - bull - Posterior r Iwithout T2S

tOO [-

I I ~ ~ I 0 50

0 00

___ r J i

l~T

~--

0 00 0 20 0 40 0 60 080 I 00 120 140 160 110 2 00

b

zshy

Posterior Distribution of Bias Factor (8) bull Effect of Removal of middotPlatform from Any Category

bullPrior and Posterior Distributions or B Evaluating Effect or Removal or 3 Damage Cases from the Base Case or 13 Platrorms

4

35

3

25

e L 2 ~ bulle shy

I 15

bull I

05

0

0 02 04 06 I 2 I 4 16 IS 2

I I

I

I

I Ibull

J -tmiddot L--

bullI

~

I

fl

II ~

I I v I

middot bull bull

bull L

~ ll- I-~lt

OS I

b

I

bull

1middot bullbullbullbull

~ __

bull

- - - - Prior-r

- - - - Postenor flall failure cases

Postenor flall success cases

--- Posterior flall 13 cases

---Posterior fI success failure cases only

IPosterior or B Mean =t11 COV =0121

-

_I

Posterir Distribution or Bias Factor (B) bull Effect or Removal or 3-Damage Cases

-

bullPrior and Posterior Distributions or B Evaluating Effect or Additional 13 Similar Observations as

Used in Andrew JIP The POsterior or B from Andrew-JIP Considered as New Prior or B

- 500

450

400

350

bull bull

bull bull bull bull J bull

bull bull f ~ I

I ~ Ibull bull

bullbull bull

l bull I -bullbull bull bull

I

I bull bull

bull I I~~bull bull bull

bull ~

I bull bull

~ ~_ ~~~

- - New Prior r bull

Posterior 114 new failure cases ~ 300- bull bull bull bull bull bull bull Posterior f 13 new damage casesl 150 ~

- bull bull - bull bull Posterior 116 new success cases i 2 00

Postenor 1113 new success failure -150 damage cases

I 00

0 50 -

0 00 0 00 0 20 0 40 0 60 0 80 I 00 120 I 40 160 110 2 00

Posterior ol B Mean= 120 COV= 0071 b

shy

shy

_rshy

Posterior Distribution of Bias Factor (B) bull Effect of Additional 13-Similar Observations

1

1 I

WEIGHTING APPROACH bull

bull ASSESSMENT OF PLATFORMS DATA

Select Platforms Most Mappable to the Analyzed Platforms

Unlikely SurvivalDamage Cases Identification

Based on Cut-Off Criteria Agemiddot Hs etc

bull INFORMATION REQUIRED FOR COMPARISON

bull CATEGORIZATION OF ADDITIONALmiddot PLATFORMS

bull UPDATING OF CALIBRATION RESULTS

--

OVERVIEW OF PLATFORMS SURVIVEDDAMAGED -~

Operator -

- Platforms

Survived CS)

Platforms

Damaged CD)

Platforms Used In

Phase IJIP -

JI~ ~Bill~IRBDll -

Amoco -

Chevron -

Exxon

Mob II

Phllllps

Shell

Unocal -

30

143

72 22

7

39 25shy -

1

12

bull

1 bull

1

1(S)+1 (D)

3 (S) +3(D)

2 (S)

bull shy

bull

bull

bull

Total Partlclnants Platforms 338 - 15 10C6S+4D) -

Other Operators

310

13

3 (D)

Total Platforms 648 28 13C6S+7D) -

Summary of Andrew JIP bull Phase I Platforms

Plairshy w v- N-ber Lea-le 1 shy car Hind_ o-w N-bull Deplh fl fl IJpe la bull n Yeadon Jadtet lbelledtf liefalit -- _ -OD

IL H - -shy _ICllshy IL IL - - rQClto ~ _

K STUIK 137 1963 8 365 608S4 Cellardeck

-middot STl-shy 170 1964 4 -~ K 39 62269 Cellardeck - --ST134W 137 1911 4 43 608S4 Smvlnlshy- _ - -

WD90A 184 1964 8 K 267Sbull34 83 SOS31 SUboftllar deck SmvMI

MC311 343 1978 8 x S7 62 - MC397 468 1991 4 x shy S33bull6SO 65911 -shySmeSmvlvol -- gt -

r-- T23ST52 63 1969 4 ton- shyKshy 52594 S03 -T25SSl39 62 1969 4 Kshy SOS94 S063 - middot ton-shy__ -

bullbull a ton--shyST161A 118 1964 8 K - 347S 51495 Cellarclecti

r~

ST1771 142 1965 8 K 39 60167 CeDordeck failme

STISIH 137 1964 8 K 395 608S4 CeDardeck Failme

K ST130A 140 1958 a SS 6095 I lblv Pai1uno - T21ST72 61 1969 4 Kshy S4S94 4974 I lhlv Pai1uno -

- -

JE

INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON

bull WATER DEPTH

bull bull YEAR OF INSTALLATIONDESIGN

bull PLATFORM ORIENTATION

bull PLATFORM LATITUDE AND LONGITUDE

bull KEY SKETCHES OF PLATFORM STRUCTURAL FRAMING

Number of Legs Plies Leg Batter Number of Conductors bull Deck Elevations (Including sub-cellar deck If any) Jacket Framing Elevations Brace type In Vertical Frames Sizes of Primary Members (leg pile vertical frames braces primary horizontals) Joint Details (can gap etc) LegPiie Annulus Grouting Status

bull middotPOST ANDREW INSPECTION REPORT Significant Wave Height Hs (ft)

bull INFORMATION ON DAMAGES PRIOR TO ANDREW

bull RESULTS FROM IN-HOUSE STUDIES IF AVAILABLE

bull ~RIGINAL DESIGN CRITERIA INFORMATION IF AVAILABLE

bull SOIL SHEAR STRENGTH PROFILE IF AVAILABLE

Platforms Identified for Evaluating Therr Likelihood of Falling Under Unexpected Survival Category

OPlRATOR

middotADA

BLOClt

llmucnJllE

NAME

YEAll

INSTAUJD

WA1ER

DBPTll

DISTANCE

JlllOM SHOllE

(MILU)

Bl

(M)

AMOCO 111113

STl11

WDI075

A

bull D

St

17

120

1n

44

20

10 I

CllllVllQfl STlllO c 61 110 21 11 STIUO D Q 1110 21 11

STl134 p Q 137 29 11

ST0151 G 61 137 11

ST0151 I 3 121 11

ST0151 J Q 140 11

STl151 It 137 11

ST0151 PJl001 Q UT 11

ST0151 PROD-2 Q UT 11

ST017 A 3 140 35 11

WDIUT c 65 214 34 10

llXXON STI055 II 51 6T 14 I

STl15 A t3 34

WDI073

WD0073

Amiddotshy Q

1111 1111

22

22

I

I

WDI073 c 1n 22 I

- WDOCl93-shy II

65 1110 2T A 3 194 31 MOBU 111121

SS IOl7

A B

501

50

50

31

T

I

T 6

MtlllPllY SS114 B T 50 T T

PlllllIPS SS0149 A 5 55 33 T

SAMllDAN STln A 107 T T

SBllLL WDl103 A 65 223 2T WDl103 B 65 221 2T WD0104 c 65 221 2T WD0104 D 6T 20 2T

UNOCAL SS020I

SS 020I SSl20I

SS 0209

SS0209

II p

PURll

A

B

3

n 61

103

97

97

95

100

34

34

34

34 34

bullbull bullbullbull

- SS0209 D u 95 34 I

- SS 1209 SSl253

SS0253

G A

c

6T

2

100

165

1T5

34

54

54

bullbull I

I I

I I

SOUTH MARCH

~LAND

N

SOUTH PaTOI ENE ISWltD AREA

MMS Defined Limits ror Offshore Structures Requiring Inspections Following Andrew

t---Jbulln3 T25bull I bullTtl

bull1$1Htl1Abull bull1SIK1779bull

~

middot-DaTA AREA~T

CRANO rsLE

AREA bull1300

bullll4W

=shy

bullPath of Hurricane Andrew with Platforms Used In Calibration

Likelihood Function vs b All Success Case STISI K STIJOQ ST134W WD90A MCJll MCJ97

I 0000

0 9000

0 8000

07000

-Bo6000

tiiii 0 5000 -

04000

0 3000

0 2000

01000

0 0000

0 05 25

ri1rt middot-middot-middot-middot-middot-

I 5 2

b

---11 middot151K(blsu~ccsI

- bull - bull - fk 130Q(bfU~lCSS)

lie bull134W(blsuccc]

lk-WD90A[blsultCS]

- - - lk middotMC397(bfUtlCS]

--o-- lk-MCJll(bbuctcsJ

--- ll (bl 6 out of 6 Mlttcs case]

_

( Likelihood Functions middotSuccess Cases - p - 1 c I

CATEGORIZATION OF ADDITIONAL PLATFORMS

I I I

bull CATEGORIES

Unexpected Survivals

Expected Survivals

bull APPROACH

Identify Corresponding Platform(s) Evaluated In the JIP

Determine Max Wave Height at Platform Location

Evaluate Whether Wave Hit the Deck

middot - Evaluate Likelihood of R and S to Differ from the Base Cases

Review Joint Information

Assess Special Features

bull IDENTIFY NUMBER OF PLATFORMS IN EACH CATEGORY

I

UPDATING OF CALIBRATION RESULTS

bull INCLUDE EFFECT OF ADDITIONAL PLATFORMS ON bullBbull

Consider Increased 1Uncertalntles In R and S

Develop bullNeW Ukellhood Functions for Different Categories

middotLet bullxbull Number of Ukely Unexpected Survival Cases and y Number of Ukely Expected Survival Cases

Determine Combined Likelihood Function lk (b Iadditional platforms) middot

bull REVISED LIKELIHOOD FUNCTION

lk(b I all cases) = lk(b I 9 platforms) middot x lk(b Ix platforms) x lk(b I y platforms)

bull ESTABLISH REVISED BIAS FACTOR B

f8 (b) u f8 (b) lk (b I all cases)

II

t t ~

shy

APPLICATIONS OF BIAS FACTORS

ii APPLICABLE TO PARTICULAR LOAD AND RESISTANCE RECIPE

bull USE BY THE API COMMITTEES AND MMS

Criteria Revisions

Recommended Design Guidelines Revisions bull

bullbull REASSESSMENT STUDIES AND DECISION MAKING

middot

bull RISK ASSESSMENT STUDIES BY THE OPERATORS

Individual Platforms

Fleet of Platforms

middot IMR Program Implementation

bullbull NOT TO BE USED IN DESIGN

I I fl J

EXAMPLE USE OF BIAS FACTORS

bull PHASE I EXAMPLE APPLICATION

bull EVALUATION OF ANNUAL PROBABILITY OF FAILURE

Establish Annual Maximum Seastate Data

Determine Coefficients to Define Base Shear

Determine Ultimate Capacity of Platform

Determine Probablllty of Failure Versus bullbbull

-bull DETERMINE UPDATED PROBABILITY OF FAILURE

Pf - P [S gt R]

Pf I P [~ gt Rb I b] f 8 (b) db

Pr - J P [S gt Rb I b] f bull 8 (b) db

bull TOTAL PROBABILITIES OF FAILURE

Total Pf - 1 bull Il [1 middotPf] - l Pf bull

Total Pf - 1 bull Il [1middot Pf] - l Pfmiddot

Total Pr = 1 - n [1middot PfbullJ = l Pfshy

I I I

EXAMPLE APPLICATION bull PHASE I

PLATFORM ST 151 K

bull GULF OF MEXICO

bull WATER DEPTH = 137 FT

bull ANNUAL MAXIMUM HS MEDIAN 175 ft COV 031

bull ASSUMED STORM DURATION 3 hoursmiddot

bull ASSUMED NUMBER OF WAVES 800

bull MEDIAN HS FOR STORM HOURS 1 2 AND 3

bull MEDIAN CURRENT VALUE bull

bull DETERMINE ULTIMATE CAPACITY FOR THREE DIRECTIONS

bull DETERMINE ANNUAL FAILURE PROBABILITIES

Broadsidemiddot Diagonal End-on

0037 0025 0007

0071 0044 0018

0022 0016 0004

DETERMINE TOTAL ANNUAL PROBABILITY OF FAILURE

Pf - 007

Pf - 013

Pr - 004

bull

EXAMPLE USE OF BIAS FACTORS - PHASE II middot

bull DETERMINE PREDICTIVE PROBABILITY OF FAILURE

Case - 1

Pf _ S S Pf (b b2) fa1a2 (b b2) db db2

g bull -1 (b1 R I S) - 1

92 - (b2 R2 I S) - 1

Case -2

Pf s ~ [bJ f11 (b) db

Pf [bJ p [g lt 0)

g - bull 1 (b1 R1 I S) - 1

1

bull1) l

I

BENEFITS OF ANDREW JIP bull PHASE II

lt

bull IMPROVED LOAD AND RESISTANCE RECIPE

bull IMPROVED CALIBRATION PROCEDURE

bull MULTIPLE BIAS FACTORS

bull PROCEDURE TO INCLUDE EFFECT OF OTHER PLATFORMS

bull PROCEDURE TO USE MULTIPLE BIAS FACTORS

  • Structure Bookmarks
    • ANDREW PHASE II JOINT1 INDUSTRY PROJECT
    • PROGRESS MEETING NUMBER 1
    • ANDREW PHASE II JOINT INDUSTRY PROJECT PROGRESS MEETING NUMBER 1 NOVEMBER 16 AGENDA
    • ANDREW PHASE II UST OF PARTICIPANTS
    • ANDREW PHASE I OBJECTIVES AND RESULTS
    • ANDREW PHASE II PROJECT OBJECTIVES
    • SUBSEQUENT PROJECTS
    • ANDREW PHASE II PROJECT TEAM
    • ANDREW PHASE II UPDATE MODELING middotAND ANALYSIS RECIPE
    • -ANDREW PHASE II CONDUCTOR RESISTANCE
    • ANDREW PHASE Il WA VE FORCE PROFILE
    • ANDREW PHASE II WA VE FORCE PROFILEmiddot
    • ANDREW PHASE II BRACE MODELING
    • ANDREW PHASE II FOUNDATIONS
    • ANDREW PHASE II middot middot JOINT MODELING
    • ANDREW PHASE Il JOINT MODELING OPTIONS DISCUSSION
    • ANDREW PHASE II JOINT MODEJJNG OPTIONS ILLUSTRATIONS
    • ANDREW PHASE II JOINT MODELING PHASE II METHOD
    • ANDREW PHASE II ANDREW STORM CONDmONS UPDATED
    • Platform Selection bullbullbull1111111111111
    • BasisCriteria for Selection bullbullbull1111111111111
    • CALIBRATION
    • WHERE TO INTRODUCE BIAS FACTOR (B
    • PHASE I -KEY CONCLUSIONS
    • bull CALIBRATION ISSUES PHASE II
    • PHASE II WORK SCOPE (CALIBRATION)
    • SOFTWARE
    • SENSITIVITY STUDIES POSTERIOR DISTRIBUTION OF B
    • WEIGHTING APPROACH bull
    • INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON
    • CATEGORIZATION OF ADDITIONAL PLATFORMS
    • UPDATING OF CALIBRATION RESULTS
    • APPLICATIONS OF BIAS FACTORS
    • EXAMPLE USE OF BIAS FACTORS
    • EXAMPLE APPLICATION bull PHASE I PLATFORM ST 151 K
    • EXAMPLE USE OF BIAS FACTORS -PHASE II middot
    • BENEFITS OF ANDREW JIP bull PHASE II
Page 6:  · . ' ANDREW PHASE II PROJECT OBJECTIVES . ' • ·UPDATE MODELING AND ANALYSIS RECIPE· JOINTS ' ' ·BRACES WAVE PROFILE FOUNDATIONS . ' ' CONDUCTORS· • ANALVZE PLATFORMS USING

bullbull

SUBSEQUENT PROJECTS

bull MMS INSPECTION PROJECT

bull APlMMS CAISSON PROJECT

bull APlMMS FOUNDATION PROJECT

bull TRIALSBENCHMARK JIP

I I It

ANDREW PHASE II

PROJECT TEAM

bull R L SMITH PROJECT MANAGER

bullbull F middot

J PUSKAR PROJECT ADVISOR

bull Rw LITTON TECHNICAL MANAGER

bull RAGGARWAL TECHNICAL DEVELOPER

bull D K DOLAN TECHNICAL ADVISOR middot

bull C A CORNELL RELIABILITY CONSULTANT

bull F MOSES RELIABILITY CONSULTANT

bull middotMSL ENGINEERING LTD TUBULAR JOINT CONSULTANT

q gtI

ANDREW PHASE II

UPDATE MODELING middotAND ANALYSIS RECIPE

bull CONDUCTOR RESISTANCE

bull WA VE FORCE PROFILE

bull BRACE MODELING

middotbull FOUNDATIONSmiddot

bull JOINT MODELING

~ I I

-ANDREW PHASE II

CONDUCTOR RESISTANCE

shybull RESISTANCE NOT INCLUDED IN PHASE I

bull CONTRIBUTION TO OVERALL RESISTANCE A PHASE II

STUDY ISSUE

middotbull middotCOMPARE FOR 4-LEG AND 8-LEG PLATFORM

bull CONSIDER CONDUCTOR POSmON IN GUIDES

- 13 LEFT 13 CENTER 13 RIGHT

Project ST151K Model Conductor version 1 Mon Nov 14 143140 1994

- ---- ---

--

117 s

111

7i

31

0 -5 -~ - ~()

- iltgt --lo

-(pO

-so

bull bullf OC)

-120

-ISo

-ISO

z

CAgt ~x Singl_e Conductor Model

bull

=I=

-_

-_

-gtshy

~ -

~

I

I~

- gt

~-----------____________________

Project STlSlK Model dir_2 version 7 Mon Nov 14 124200 1994

177 s

Ill

74

37

0- - e -zo - ~o

- Jo

-100

-10

-ISO

z

UV 1-+x

Project STlSlX Model Conductor Version 1 Mon Nov 14 140620 1994

bullmiddot

I

I

I I

I

I I

I I I I I

1 ==

-L

-L - L

-

- L

-

-

L

-

Uniform Displacement step 6 Inelastic Events Legend

Elastic

Strut Buckling Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield Beam Clmn FUlly Plastic Fracture

--

i~~~~~~~~~~~---~--~~-J___--JJ~_middot~~~=-

Project ST151K Model Conductor Version1 Mon Nov 14 142735 1994

---

~

I

I

middot

I

I

I

- shy- shy

- shy

- shy

4x z

CIV Uniform Displacement step 33 Inelastic Events Legend

bull Elasti~ Strut Buckling Strut Residual Strut Reloading Plastic StrutNLTruss Beam Clmn Initial Yield Beam Clmn-Fully Plastic Fracture

gt bull

Project STlSlK middot Model Conductor Version 2 Mon Nov 14 142025 1994

-----------------

bullf

I

I

I

II I I

I I I I

I ==

-

Variable Displacement step 9 Inelastic Events Legend

Elastic Strut Residual

bull

Strut Buckling Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield Beam Clmn Fully Plastic Fracture

--

Project ST151K Model Conductor Version 2 Mon Nov 14 142145 1994

I I I I

I I

I

I

I - - -

-Ishy-

-

Variable Displacement step 35 InelasticEvents Legend

Elastic Strut Buckling Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield Beam Clmn Fully Plastic Fracture

Project ST151K Model Conductormiddot

Version 1

Clgt - Cut Plane Force Fx Mon Nov 14 140051 1994

- -_ vii I fOIZM

0 _

~

N ID 0 bull 0

x 0

middotrt ~ () bull u 0 ~ 0 Ir

N bull

0

0 1 2 3 4 5 X Deck Disp [Ft)

I I 1 1 WbullTI- bullbull

lltcxi lr-C

011lt~~~~~~~---~~~~~~~~L~~~~~~~--~~~~~~~~i~~~~~~~--

Uniform Node Displacement

shy

Ji

r

Comparison of Pushover Curves for ST 151K Variable Load Pattern and Monotonic Load Pattern (Andrew)

Soll Improved 100

6000

5000 -i 4000

f

llS

3000 Ul

J 2000

10 c11 c1lgtolts 1000

0 v

bull ~

f

_ ----- --- ---shy

bull --Variable Lold

- - Andrew Load

I I I I I I

4 5 62 30 1

Deck Dlsplacement (ft)

bullbull r 1 II

~

ANDREW PHASE Il

WA VE FORCE PROFILE

bull DECK WA VE FORCE ~

PHASE I USED SECTION 17 (1056H degCd)

PHASE Il USE REVISED SECTION 17 (1000H Cd)

bull PHASE I USED SINGLE FORCE PROFILE FOR PUSHOVER

- USED WAVE HEIGHT CLOSE TO ULTIMATE CAPACITY

bull PHASE J

Il WILL STUDY THE

USE OF VARIABLE

PROFILE ~

- IF FORCE LEVEL INCREASES (OR DECREASES) BY A FACTOR EQUIVALENT TO A 2 FOOT CHANGE IN WA VE HEIGHT THE INCREMENTAL FO~CE PA1IERN IS CHANGED ACCORDINGLY (CAP FEATURE)

- INITIAL EXAMPLE

f~

a

middotn

a

I N -a a a amiddot

+

~ a

l

a w gt 0c 0 gt a I shyc -CJ wc w gtc(

== c( I shyZ w E W a 0 z- 8

ii sect

~ ~ bullT 0

I if

x ~

bullbullbull

II ~------------________

ANDREW PHASE II

WA VE FORCE PROFILEmiddot

CAP AUTOMATIC APPLICATION FEATURE

STAT

fR1 0 lt f lt1

STAT

STAT

f(R3-R2) middot f lt1

Project ST151K Model Wave_Loading Version 1 rue Nov 8 164107 1994 PT

E~ nss FT

SL 14amp FT

a 111 FT

EL 14 li

E-1 37 FT

El o FT

1middot1middot ______________ _ ________ __ _ __

0

200 400 600 800 1000 12000 Force at Elevatloii (kip)

200

a ll_

Wave Force at Major Elevations for Discrete Wave Heights

180

bullHmiddot40ft

Hmiddot495ft

H bull 51 ft

Hmiddot55ft_

H-59ft I

Andrew (H bull 60 86 ft) II

i I

_11---__

- middot- Hmiddot30ft

-

- bull -

middot bull bull bull middot

---Hmiddot53ft

bullbullbullbullHmiddot57ft

-----H-63ft

Elevation Above Mudllne 100

(ft)

180

140

120

80

60

40

20

Loading at Major Elevations

---H-49511

-deg-Hmiddot608611

0 200 400 600 800 1000 1200

Force It Elevatlon (kip)

--

Base Shear vs Wave Height ST151K

6000

5000 1---------Ult1mateCapacitybull 4890 kip (100 Improved Soll)-----------~--

4000a a lii ~ 3000

81 ampI

2000

1000

0 +-~~~--i~~~~~~~~~-+-~~~--lf--~~~~~~~~-+-~~~--l

30 35 40 45 50 55 60 65

Wave Height (ft)

bull middot

~

bull

II

2

g

~

It)

0

cCD E 0 E a c E euro~ CDshygtItgtmiddoto ~ti c -a Gic CD gt ~

-= c -Cll

I z

bullO 0i i ~ g 0

() CI -bull(q-ll) iuawow llu1wniiaAO

11

shy

116

114

112

g 110

bullc = 108

~ ~ 106 -sect 104

102

100

98

96

-

6555 60

Wave Height vs Center of Action ST151K

30 35 40 45 50

Wave Height (ft)

t

Comparison of Pushover Curves for ST 151K Variable Load Pattern and Monotonic Load Pattern (Andrew)

Soll Improved 100

--- --------- --- ----6000

5000

- 4000 8 - I

1 3000 Ul

81 m 2000

_J -shy

1000

---Variable Load

- - Andrew Load

0 f-~~~~~r--~~~~-t-~~~~~-t-~~~~~+-~~~~----t~~~~~-1

0 1 - 2 3 4 5 6

Deck Displacement (ft)

-middotshy

1 t I I ~

ANDREW PHASE II

BRACE MODELING

bull MARSHALL STRUTS USED FORPHASE I

bull PHASE II WILL ASSESS DIFFERENCE IN USING BUCKLING BEAMS AND MARSHALL STRUTS

middot bull COMPARE FOR 2D FRAME

bull COMPARE FOR PLATFORM 177 WHERE LOWER

BAY EXHIBITS LARGE BRACE MOMENTS

Project andrew Model beamframe Version 1 Mon Nov 14 145447 1994

~

middot Frame Test buckling beam-column model

-

Project andrew Model strutframe Version 1 Mon Nov 14 145754 1994

bull

Frame Test strut model

middot-middot _____________--_

_middotmiddotr_~ --ll--~-middot__

Buckling Beam bull

Strut (K=O 65)

Lateral Displacement

ANDREW PHASE II

FOUNDATIONS

bull ~LICIT NONLINEAR API FOUNDATION USED INiPHASE I

bull FOUNDATIONS

JIP SELECTED MINI-VANE DATA FOR

STRENGTH PROFILE

bull COMMENTS ON BENCHMARK COMPARISONS

- SIGNIFICANT VARIABILITY DUE TO FOUNDATION MODELING METHODS

II Ii

z

ctV 4x SPlO caissonfy-42s soil-api(lf-097step-114)

Inelastic Events Legend Elastic Strut Buckling

Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield

--------shyBeam Clmn Fully Plastic Fracture

~--~~~~~~~~~--~----~__i~~---~~~~__--~

Pro)ect apipile Model South Pelto 10 Version Wed May 18072235 1994 - -

1bullt

+

t

f i bullbullbull

bullbull bullbull

I I

5

~x SPlO Caissonfy=42s soil=api(lf=lOOstep=137)

Inelastic Events Legend

Elastic Strut Bucklilg

Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield

earn Clmn Full~ Plas1c Fracture

ANDREW PHASE II middot middot

JOINT MODELING

middotbull API RIGID-PLASTIC USED IN PHASE I

a ELASTIC-BRITTLE USED IN PHASE IB

amiddot MSL WILL PROVIDE NONLINEAR FORCE-DEFORMATION

AND MOMENT-DEFORMATION RELATIONS FOR PHASE II

a MODELING OPTIONS FOR PHASE II -

l

- UNCO~LEDmiddotONE DIMENSIONAL MEMBER END JOINT MODELING

J Ii

ANDREW PHASE Il

JOINT MODELING OPTIONS DISCUSSION

bull JD COUPLED INTERACTION SURFACE

- IDEAL

- DATA NOT AVAILABLE TO CALIBRATE

bull 2D COUPLED INTERACTION SURFACE

- SIGNIFICANT MODELING IMPROVEMENT

- DATAAVAILABLE

- IMPLEMENTATION PRACTICAL BUT NEEDS DEVELOPMENT AND CALIBRATION

bull 2D UNCOUPLED INTERACTION SURFACE

- SIGNIFICANT MODELING IMPROVEMENT

- DATA AVAILABLE

- IMPLEMENTATION PRACTICAL AND AVAILABLE

ANDREW PHASE II

JOINT MODEJJNG OPTIONS ILLUSTRATIONS

P

e

ANDREW PHASE II

JOINT MODELING PHASE II METHOD

e

I I I

ANDREW PHASE II

ANDREW STORM CONDmONS UPDATED

bull -PHASE I USED OCEANWEATHER HINDCAST-(Nov 1992)

bull middotmiddotPHASE II WILL USE THE REVISED HINDCAST (Nov 1994)

bull lt c lt I lt

2-2

l---4-~~~+-~~-+-~~~1--~~-+-~~--l--~T--+-~~--i24

f-tJVtMfiz~ 1Cfl (r STvb(

Plollol bullbull 5middotNOYmiddot9214bull1J Irbullbull Illbull [ANDREVl~NDREVCUSTbullMAXll 5-NOY-1992 10bull54 - 9208

HEIGHT lml

I~ MAXI

I L 1~middotamp SZ~- ~A I 13c

I J) ( r I X~0iWJ~0gtJ gt L gt gt I 1c I 120

-- N

~ deg

I I I I I I e C I ~=-41 122 -94 -92 -90 -88 -86 -84 -82 -80

-

Figure 7 contours of maximum hindcast significant wave height

_

~

~tl ~

NOltNttIJ 17lt ltNS S iJJ)y

Maximum Significant Wave Height - Hurricane Andrew 1992

30 ~

26 ~ I ~-~~~- == -shy -~ 35-_

24deg1--shy --------~- middot-middot-------middotmiddotmiddot----middot------middot-middot~middot- - - ~-----1---middot

22middot ~~~~~~~-~ -94 -92 -90 -88 -84deg-86 -82 -somiddot

c

-

Platform Selection bullbullbull1111111111111

0 BasisCriteria for Selection middot

0 Review of Phase I Platforms middot 0 Alternative Platforms

0 PMB Recommendations

0 Participants Comments

gt

L

l I

BasisCriteria for Selection bullbullbull1111111111111

0 Benefit for Calibration - Classification (UnexpectedExpected)

- Failure Mode (StructuresFoundations)

- Damaged StructuresKnown response

- Better Representation of Full ~opulation

0 Availability of Data - Existing Models

- Inspection Data

- Damage Data

PREDICTED BEHAVIOR

-

OBSERVED

BEHAVIOR

-SURVIVED DAMAGED FAILED

SURVIVED

ST134W WD90A

-MC311 -

MC397 -

--

-

-

ST151K ST130Q

-

-

-

DAMAGED - -

-

-

--

-

-

ST52 SS139 ST177B -

ST161A

-_

-

FAILED

-

-

-

ST151H ST130A ST72

Operator - Platfonns - Platfonns Survived Damagedamp

- Falled -

bull

-

-

Jle eBDIGIPllDbl -Amocoshy 30 1 Chevron -

143 12 Exxon 72 Mobll 22 1 Phllllps 7 bull

-Shell 39 Unocal 25 1

Total Partlclnts Platfonns 338 15

Other Operators bull- 310 13

-

Total Platfonns - -

648 28

fbml

Platforms Considered 6 7

tie of Total Platfonns 1 25

OVERVIEW OF PLATFORMS SURVIVEDDAMAGEDFAILED

-

shy

v

Platforms Evaluated in Phase I

-

-Platfonn Water Year Number Analysis Avallabillty of Availability Platform

- Name Depth rl of Model DetaDed rl ClassUlcatlon

Installation legs Inspection Soll Report Based on Information Phase I Calibration - -ft

- -- Suaill flalfnrm ~BSi -STISIK 137 1963 8 CAP Yes Yes Un eel Sarvlval

-STIJOO 170 1964 4 CAPmiddotshy Yes ST134data u ST134W 137 1981 4 CAP Yes Yesmiddot E

-

edSarvlval -

middotsurvival

WD90A 184 - 1964 8 CAP Yes E -

- Sarvlvill

MC3ll 343 1978 8 KARMA

MC397 468 1991 4 KARMA -

bull ~reglmmiddot- -

Yes Sure Survival

bull Yes - Sare Sanlval-

-T23STS2 63 1969 4 CAP

T2SSS139 62 1969 4 CAP -

Yes

Yes

Yes Ex

Yes Une

bull SurvivalLikely to DamalI

middot bull SanlvalLlkely to DamalI

ST161A 118 1964 8 USFOS

Eail11a pound1al(bullllm Clss ST177B 142 --1965 8 CAP

STISIH 137 1964 - 8 CAP C

STIJOA 140 19S8 8 CAP - -

Yes

Yes

ollamed

Yes

- Yes E

-

at about 1 mUe

Yes

ST134data

bull SafvlvalLlkelv to Dam- bull- -

E middotrahre

Emeried Failure -

Likely FaUare

T21ST72 61 1969 4 CAP Collamed No Likely FaUure

Platforms Identified for Evaluating Their Likelihood of Falling Under UnexpeCted Survival Category

CllEllATOa

AllB

llLOCll

Sl1lvcnJllE

NAME

YEAR

INSTAUJCD

WATER

IJBPJB

middotmiddotDISTANCE

ftOMSllOllE

(MIUIS)

Ill

(M)

AMOCO BllW

STllI

WDI075

A

bull D

4

n uo 17J

44

32

bull 20

10

bull CllEVllOH STtllO c I 1bull 21 11

STl130 D 0 161 21 11 STllJ4 I 0 137 D - 11

STIUI G I 137 32 11

STIUI I 0 121 32 11

STIUI

J 0 140 32 11

STIUI It 4 137 32 11

STllSI PROl)I 0 137 32 11

STllSI PROD-2 0 137 32 11

STll7 A 0

140 35 11

WDl117 c 65 214 34 10

DXON STshy E 51 14 bull STl165 A t3 bull 34 WD0073

WD0073

WD0073

A _ c

4

62

4

IA IA

17J

22

22

22

bullbullbull WD003-shy E 65 161 Z7

A 0 IN 31 MOBD Bl12I

ssn A bull

50

50

31

7

bull 7

MIJllPllY SS114 B 7 50 7 7

PlllUIPS SSl14t A 55 33 7

SAMEDAN ST17J A 4 117 7 bull 7

SJIELL WD0103 A 65 223 27 WDllOJ bull 65 221 27 WDll04 c 65 221 27 WDOI04 D 67 2 27

UNOCAL

SSl20I

SSl20I

bull SSl20I

SS0209

SS0209

SS0209

SS0209

SS 0253

SS 0253

E

I

PtJllE

A

bull D

G A

c

0

4

4

7J

61

0

0

69

103

97

97

9S

100

95

100

165

175

34

34

34

34

34

34

34

54

54

bullbullbullbullbullbullbullbullbull

Platforms Recommended for Phase II

llalr y_ w- Aot Av- _ I 1 Doplla Model nm bull bull _ bull I I H11 s18-1 Irht lbwl~ bullbullbull(-illtdlel OM lllMtD

II ---

-STISIIC 137 1963 I CAP Yeo Yeo I Yeo

STtshy -170 1964 4 CAP Yeo ST134dlU I Yeo

STl34W 137 Yeo Yeo -1981 4 CAP D Yeo

WD90A 114 1964 I CAP Yeo - D

MC311 30 1971 I KARMA Yeo -sre-

MC397 middot 461 1991 4 ICARMA Yeo Sireshy - - r- -

__ T13ST5l 63 1969 4 CAP Yeo Yeobull middot- ~n- D Yeo _ Tl5SSl39 62 1969 4 CAP Yeo Yeo I Yeo

_ STIT711 142 1965 I CAP Yeo I Yeo middot-1shySTl61A Ill 1964 I USFOS Yeo Yeo - D bulloshy

- r-

STISIH 137 1964 I CAP y D y -_ STl31A 140 1951 I CAP Yeo STl34dlU I Yeo

T21Sll 61 1969 4 CAP No rar-middot I Yeo

~-rmiddot-middot SS209A 95 1967 16 CAP Yeo D -SSU4R

11 UI A 41 -in D SS 2741AIM CAP bull

_ _ -shy - -shy- -shy

CALIBRATION

bull PHASEIAPPROACH

bull CALIBRATION ISSUES

bull PHASE II WORK SCOPE

bull SOFlWARE AVAILABILITY

bull EFFECT OF ADDITIONAL PLATFORMS

bull WEIGHTING APPROACH

bull EXAMPLE APPLICATION middot

(l=Ofce Mode] Wind

~

fcurrent l -middot

~SW ~leldl4

ultimate resistance force I first

design load Illgt

i _

deflection

-

bull

4

Historical Hurrlcane Dariiage and Survival of Platforms

Components Required for Calibration (Petrauskas 1992)

PRIOR Distribution

PLATFORM EXPERIENCE

BAYESIAN Survivals and Failures

UPDATING

Waves Winds and Currents bull

POSTERIOR Distribution

- A_middotJA ~ Survival Data Failure Data

Result Result

Load Resistance

middotfl

-

Bayesian Updaing middotGeneral Description (Petrauskas 1992)

L Imiddot

WHERE TO INTRODUCE BIAS FACTOR (B

On Resistance (Rmue - B (RPREDICreD I

middotOn Loads (Smue middot - B (SPRE01cTEo

On Safety Margin (RSmue - B (RSPREDICTED

CALIBRATION METHODOLOGY

gt

r

FIELD OBSERVATIONS

Observed Behavior

survivals Damages

Failures

r

bull

CAPACITY ANALYSIS

Predicted Load and Resistance Predicted Failure Modes

Survival Damage Failure cases

RELIABILITY ANALYSIS

- Probability of Failure ~

Ukellhood Failure

BAYESIAN UPDATING

Prior Distnbution of B Combined Likelihood Function

Posterior Dlstnbution of B

BIAS FACTOR

Mean Value cov bull

Represents Modeling Uncertainties

Prior Distribution of B Likelihood Functionmiddot Observing Success case STISIK and Posterior Distribution of B

2 00

uo

160

140

120

e 100 L $ uoIshy

0 60 ~ ~ 040

020

0 00

0 00 0 50

_ I 00

b

I 50 2 00

-shy lk [bl STISIK)

--shy Pnor-f(b)

--shy Pos1enor r(bl ISIK)

Prior or B Mean bull 10 COV bull 030

Posterlor of Bbull Meanbull 136COV 0 16

2 50

Likelihood Function and Posterior Distribution Example Platform STISIK

- -

~

- Prior and Posterior Distributions or B Failure cases (ST177B STlSlH ST130A T21) Damage cases (T23 T25 ST161A) Success cases (STlSlK ST130Q ST134W WD90A MC311 MC397)

400

350

300

2 so 0-

I

bull bull

bull bull gt

- bullbull ~ bull bull I bull

_ bull bull bullbull I bull bull bullII V -

bull bull

I bull ~ bull

bullbull bullI bull bullI bull bull bull

middot -~

bull__ ~ middotI middotmiddot

L 2 00

e ~ I SO

-I 00

- 0 so

000 gt

0 00 020 0 40 060 0 80 I 00 120 140 160 110 200

b -

-

shy

Prior-r

- - - - Posterior fLlll failure cases

bull bull bull bull bull bull bull Posterior flall damage cases

- bull bull - bull middot Postenor flall success cases

Posterior flall success fulure damage cases

-

Posterior of B Mean= 119 COV 0101

Posterior Distribution or Blas Factor (8) bull Different Categories and All Cases Combined

shy

l

PHASE I - KEY CONCLUSIONS

bull GLOBAL CORRECTION FACTOR

bull 15 TO 20 CONSERVATISM MEAN)

bull UNEXPECTED SURVIVALS HAVE GREATEST IMPACT

bull DAMAGED CASES MORE INFORMATIVE

bull CALIBRATION ISSUES PHASE II

bull

EFFECT OF IMPROVED CAPACITY ANALYSIS RECIPE

bull APPLICABILITY TO MULTIPLE FAILURE MODES

bull NO FOUNDATION DAMAGE CASE INCLUDED

bull IMPROVED REPRESENTATION OF FULL ANDREW DISTRIBUTION

-bull NUMBER OF PLATFORMS CONSIDERED

I 1 I

I t ) bull ~

PHASE II WORK SCOPE (CALIBRATION)

bull

bull ESTABLISH MULTIPLE BIAS FACTORS

Use Nine Base Case Platforms

middot - Capacity Analysis for Revised recipe

Foundation Blas Factor from Foundation JIP

bull DEVELOP WEIGHTING APPROACH

bull UPDATE MULTIPLE BIAS FACTORS

bullEffect of Platforms Not In the Nine Base Cases

bull EXAMPLE APPLICATION OF BIAS FACTORS I

bullbull

PHASE I PHASE II

CAPACITYbull ANALYSISRu bull

---------------------1----middot

Jc1 - iJbullraquoJ dbull

FttlS frt[f~l ratdhJJ

-

LIKELIHOOD FUNCTION

DEVELOPMENT

lk Jb Jfall1N) bull Jb)

111 Ill Isurvtvoq bull 1 - bl

-middot -------------------shy ~---------------------middot-middot

r111ir8 Cb1 111 Cbl BAYESIAN

UPDATING

~cove

R1R2 bull middot---------------------~----

Jblbull P(g cOJ

1 bull JbRI (MBSI I bull 1

~middot

llcJllJfalluro) bull P(gcOJ Ill 111 b21fallun) bull P ((g1 CO) U 1112 c OD

middot--------------------shy ----------------------shy bull

r111111r8 Cbl 111 Ill i-1nro1

deg ~1b2)laquottfrr (b1b21-lnfo) re Ill bull 1-11middot b2) bull lodegamp

middot

i1 covamp i2 cov~i

-

shy

--

Likelihood Function vs b All Success Case STISI K STl30Q STl34W WD90A MC311 MC397

I 0000

0 9000

OIOOO

0 7000

~ 0amp000

ta osoOO

~ 0 4000

0 3000

0 2000

0 1000

bull 0 0000

bull

I

middot

middot bull I

Ibull

-sibull ~ -middot-middot-middot-middot -middot-

----7----+-----~ I

Ibull bull bull I

rbull

--- Jk ISIK[blsuccess)

- bull - bull - lk 130Q[blsuccess) bull

lk 134W[blsuccess)

Jk-WD90A(blsuccc~I

- - - lk MC397[blsucccss)

--ltgt-- lk middotMC31 l[blsucccss)

--- Jk [bl 6 out of6succcsHascs)

05 I 5 2 250

b

Likelihood Functionsmiddot Success Cases - Phbullsc L

shy

bullbull bull bull bull bull bull -------------------~

bull bull bullbull bullbull bullbull bullbull bullbull bullbull bull

bull bullbull

bullbullbullbull

bullbull

bullbullbullbullbullbullbull

bullbullbullbullbullbull

bullbullbullbullbullbullbull

z 0

t z ) LL

c 0 0c i ~ i

J

~ sect ~--------------~ e l5

I

I

SOFTWARE

bull PF PROGRAM

bull CALREL

bull PF +CALREL

bull RELACS

bull

gt i

l

SENSITIVITY STUDIES

POSTERIOR DISTRIBUTION OF B

bull EFFECT OF CHANGE IN DISTRIBUTION OF PRIOR OF B

bull EFFECT OF REMOVAL OF ONE PLATFORM FROM EACH CATEGORY

bull EFFECT OF REMOVAL OF THREE DAMAGE CASES

bull EFFECT OF ADDITIONAL 13 SIMILAR PLATFORMS

-I

Prior and Posterior Distributions of B Variation In COV of PriorofB fromOl to04 -- -

4500

-4000

-

- -

3500 --middot - bull bull - bull bull Prior-r ICOV =02

3000

- - - - Prior-r ICOV =03 -

bull -~ e 2soo -

bull bull bull bull bull bull bull Pnor-r ICOV =o 4t frJ ~ 2000 Posterior rlCOV =0 2

bull bull Postenor rlCOV =0 3 bull bull 1500 bull middot bull - shy -middot- Posterior rlCOV =04

middot~ I 000 - bull

-middotmiddot ~ ~middotmiddot M -~

- middot 0 500

wbullbullbull _ I middot~ middotmiddotmiddotmiddotmiddot middotmiddotmiddotmiddotmiddot - -~ ~ __ middot_ _ bull -shy0 000

0 000 0200 0 400 0 600 0 800 IOOO 1200 1400 1600 1800 2000 -

- - b

-

=shy

Posterior Distribution of Bias Factor (B) bull Effect of Variation In COV of Prior of B

bull

bullPrior and Posterior Distributions of B Evaluating Effect of Removal of One Platform from Each Category from the Base Case of 13 Platforms

400

~ ~ 350

j IA i

3 00 bull I bull I ---Prior-r I bull l

250 1 I 1

I bull bull ---Posterior r lall 13 casesshy- ~E 2 00 bull bull bull bull bull bull bull Posterior r without STISIK

- ~ 1 bull - bull I - bull bull - bull bull Posterior r Iwithout T2I I 50 I

J-tt~i - bull - bull - Posterior r Iwithout T2S

tOO [-

I I ~ ~ I 0 50

0 00

___ r J i

l~T

~--

0 00 0 20 0 40 0 60 080 I 00 120 140 160 110 2 00

b

zshy

Posterior Distribution of Bias Factor (8) bull Effect of Removal of middotPlatform from Any Category

bullPrior and Posterior Distributions or B Evaluating Effect or Removal or 3 Damage Cases from the Base Case or 13 Platrorms

4

35

3

25

e L 2 ~ bulle shy

I 15

bull I

05

0

0 02 04 06 I 2 I 4 16 IS 2

I I

I

I

I Ibull

J -tmiddot L--

bullI

~

I

fl

II ~

I I v I

middot bull bull

bull L

~ ll- I-~lt

OS I

b

I

bull

1middot bullbullbullbull

~ __

bull

- - - - Prior-r

- - - - Postenor flall failure cases

Postenor flall success cases

--- Posterior flall 13 cases

---Posterior fI success failure cases only

IPosterior or B Mean =t11 COV =0121

-

_I

Posterir Distribution or Bias Factor (B) bull Effect or Removal or 3-Damage Cases

-

bullPrior and Posterior Distributions or B Evaluating Effect or Additional 13 Similar Observations as

Used in Andrew JIP The POsterior or B from Andrew-JIP Considered as New Prior or B

- 500

450

400

350

bull bull

bull bull bull bull J bull

bull bull f ~ I

I ~ Ibull bull

bullbull bull

l bull I -bullbull bull bull

I

I bull bull

bull I I~~bull bull bull

bull ~

I bull bull

~ ~_ ~~~

- - New Prior r bull

Posterior 114 new failure cases ~ 300- bull bull bull bull bull bull bull Posterior f 13 new damage casesl 150 ~

- bull bull - bull bull Posterior 116 new success cases i 2 00

Postenor 1113 new success failure -150 damage cases

I 00

0 50 -

0 00 0 00 0 20 0 40 0 60 0 80 I 00 120 I 40 160 110 2 00

Posterior ol B Mean= 120 COV= 0071 b

shy

shy

_rshy

Posterior Distribution of Bias Factor (B) bull Effect of Additional 13-Similar Observations

1

1 I

WEIGHTING APPROACH bull

bull ASSESSMENT OF PLATFORMS DATA

Select Platforms Most Mappable to the Analyzed Platforms

Unlikely SurvivalDamage Cases Identification

Based on Cut-Off Criteria Agemiddot Hs etc

bull INFORMATION REQUIRED FOR COMPARISON

bull CATEGORIZATION OF ADDITIONALmiddot PLATFORMS

bull UPDATING OF CALIBRATION RESULTS

--

OVERVIEW OF PLATFORMS SURVIVEDDAMAGED -~

Operator -

- Platforms

Survived CS)

Platforms

Damaged CD)

Platforms Used In

Phase IJIP -

JI~ ~Bill~IRBDll -

Amoco -

Chevron -

Exxon

Mob II

Phllllps

Shell

Unocal -

30

143

72 22

7

39 25shy -

1

12

bull

1 bull

1

1(S)+1 (D)

3 (S) +3(D)

2 (S)

bull shy

bull

bull

bull

Total Partlclnants Platforms 338 - 15 10C6S+4D) -

Other Operators

310

13

3 (D)

Total Platforms 648 28 13C6S+7D) -

Summary of Andrew JIP bull Phase I Platforms

Plairshy w v- N-ber Lea-le 1 shy car Hind_ o-w N-bull Deplh fl fl IJpe la bull n Yeadon Jadtet lbelledtf liefalit -- _ -OD

IL H - -shy _ICllshy IL IL - - rQClto ~ _

K STUIK 137 1963 8 365 608S4 Cellardeck

-middot STl-shy 170 1964 4 -~ K 39 62269 Cellardeck - --ST134W 137 1911 4 43 608S4 Smvlnlshy- _ - -

WD90A 184 1964 8 K 267Sbull34 83 SOS31 SUboftllar deck SmvMI

MC311 343 1978 8 x S7 62 - MC397 468 1991 4 x shy S33bull6SO 65911 -shySmeSmvlvol -- gt -

r-- T23ST52 63 1969 4 ton- shyKshy 52594 S03 -T25SSl39 62 1969 4 Kshy SOS94 S063 - middot ton-shy__ -

bullbull a ton--shyST161A 118 1964 8 K - 347S 51495 Cellarclecti

r~

ST1771 142 1965 8 K 39 60167 CeDordeck failme

STISIH 137 1964 8 K 395 608S4 CeDardeck Failme

K ST130A 140 1958 a SS 6095 I lblv Pai1uno - T21ST72 61 1969 4 Kshy S4S94 4974 I lhlv Pai1uno -

- -

JE

INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON

bull WATER DEPTH

bull bull YEAR OF INSTALLATIONDESIGN

bull PLATFORM ORIENTATION

bull PLATFORM LATITUDE AND LONGITUDE

bull KEY SKETCHES OF PLATFORM STRUCTURAL FRAMING

Number of Legs Plies Leg Batter Number of Conductors bull Deck Elevations (Including sub-cellar deck If any) Jacket Framing Elevations Brace type In Vertical Frames Sizes of Primary Members (leg pile vertical frames braces primary horizontals) Joint Details (can gap etc) LegPiie Annulus Grouting Status

bull middotPOST ANDREW INSPECTION REPORT Significant Wave Height Hs (ft)

bull INFORMATION ON DAMAGES PRIOR TO ANDREW

bull RESULTS FROM IN-HOUSE STUDIES IF AVAILABLE

bull ~RIGINAL DESIGN CRITERIA INFORMATION IF AVAILABLE

bull SOIL SHEAR STRENGTH PROFILE IF AVAILABLE

Platforms Identified for Evaluating Therr Likelihood of Falling Under Unexpected Survival Category

OPlRATOR

middotADA

BLOClt

llmucnJllE

NAME

YEAll

INSTAUJD

WA1ER

DBPTll

DISTANCE

JlllOM SHOllE

(MILU)

Bl

(M)

AMOCO 111113

STl11

WDI075

A

bull D

St

17

120

1n

44

20

10 I

CllllVllQfl STlllO c 61 110 21 11 STIUO D Q 1110 21 11

STl134 p Q 137 29 11

ST0151 G 61 137 11

ST0151 I 3 121 11

ST0151 J Q 140 11

STl151 It 137 11

ST0151 PJl001 Q UT 11

ST0151 PROD-2 Q UT 11

ST017 A 3 140 35 11

WDIUT c 65 214 34 10

llXXON STI055 II 51 6T 14 I

STl15 A t3 34

WDI073

WD0073

Amiddotshy Q

1111 1111

22

22

I

I

WDI073 c 1n 22 I

- WDOCl93-shy II

65 1110 2T A 3 194 31 MOBU 111121

SS IOl7

A B

501

50

50

31

T

I

T 6

MtlllPllY SS114 B T 50 T T

PlllllIPS SS0149 A 5 55 33 T

SAMllDAN STln A 107 T T

SBllLL WDl103 A 65 223 2T WDl103 B 65 221 2T WD0104 c 65 221 2T WD0104 D 6T 20 2T

UNOCAL SS020I

SS 020I SSl20I

SS 0209

SS0209

II p

PURll

A

B

3

n 61

103

97

97

95

100

34

34

34

34 34

bullbull bullbullbull

- SS0209 D u 95 34 I

- SS 1209 SSl253

SS0253

G A

c

6T

2

100

165

1T5

34

54

54

bullbull I

I I

I I

SOUTH MARCH

~LAND

N

SOUTH PaTOI ENE ISWltD AREA

MMS Defined Limits ror Offshore Structures Requiring Inspections Following Andrew

t---Jbulln3 T25bull I bullTtl

bull1$1Htl1Abull bull1SIK1779bull

~

middot-DaTA AREA~T

CRANO rsLE

AREA bull1300

bullll4W

=shy

bullPath of Hurricane Andrew with Platforms Used In Calibration

Likelihood Function vs b All Success Case STISI K STIJOQ ST134W WD90A MCJll MCJ97

I 0000

0 9000

0 8000

07000

-Bo6000

tiiii 0 5000 -

04000

0 3000

0 2000

01000

0 0000

0 05 25

ri1rt middot-middot-middot-middot-middot-

I 5 2

b

---11 middot151K(blsu~ccsI

- bull - bull - fk 130Q(bfU~lCSS)

lie bull134W(blsuccc]

lk-WD90A[blsultCS]

- - - lk middotMC397(bfUtlCS]

--o-- lk-MCJll(bbuctcsJ

--- ll (bl 6 out of 6 Mlttcs case]

_

( Likelihood Functions middotSuccess Cases - p - 1 c I

CATEGORIZATION OF ADDITIONAL PLATFORMS

I I I

bull CATEGORIES

Unexpected Survivals

Expected Survivals

bull APPROACH

Identify Corresponding Platform(s) Evaluated In the JIP

Determine Max Wave Height at Platform Location

Evaluate Whether Wave Hit the Deck

middot - Evaluate Likelihood of R and S to Differ from the Base Cases

Review Joint Information

Assess Special Features

bull IDENTIFY NUMBER OF PLATFORMS IN EACH CATEGORY

I

UPDATING OF CALIBRATION RESULTS

bull INCLUDE EFFECT OF ADDITIONAL PLATFORMS ON bullBbull

Consider Increased 1Uncertalntles In R and S

Develop bullNeW Ukellhood Functions for Different Categories

middotLet bullxbull Number of Ukely Unexpected Survival Cases and y Number of Ukely Expected Survival Cases

Determine Combined Likelihood Function lk (b Iadditional platforms) middot

bull REVISED LIKELIHOOD FUNCTION

lk(b I all cases) = lk(b I 9 platforms) middot x lk(b Ix platforms) x lk(b I y platforms)

bull ESTABLISH REVISED BIAS FACTOR B

f8 (b) u f8 (b) lk (b I all cases)

II

t t ~

shy

APPLICATIONS OF BIAS FACTORS

ii APPLICABLE TO PARTICULAR LOAD AND RESISTANCE RECIPE

bull USE BY THE API COMMITTEES AND MMS

Criteria Revisions

Recommended Design Guidelines Revisions bull

bullbull REASSESSMENT STUDIES AND DECISION MAKING

middot

bull RISK ASSESSMENT STUDIES BY THE OPERATORS

Individual Platforms

Fleet of Platforms

middot IMR Program Implementation

bullbull NOT TO BE USED IN DESIGN

I I fl J

EXAMPLE USE OF BIAS FACTORS

bull PHASE I EXAMPLE APPLICATION

bull EVALUATION OF ANNUAL PROBABILITY OF FAILURE

Establish Annual Maximum Seastate Data

Determine Coefficients to Define Base Shear

Determine Ultimate Capacity of Platform

Determine Probablllty of Failure Versus bullbbull

-bull DETERMINE UPDATED PROBABILITY OF FAILURE

Pf - P [S gt R]

Pf I P [~ gt Rb I b] f 8 (b) db

Pr - J P [S gt Rb I b] f bull 8 (b) db

bull TOTAL PROBABILITIES OF FAILURE

Total Pf - 1 bull Il [1 middotPf] - l Pf bull

Total Pf - 1 bull Il [1middot Pf] - l Pfmiddot

Total Pr = 1 - n [1middot PfbullJ = l Pfshy

I I I

EXAMPLE APPLICATION bull PHASE I

PLATFORM ST 151 K

bull GULF OF MEXICO

bull WATER DEPTH = 137 FT

bull ANNUAL MAXIMUM HS MEDIAN 175 ft COV 031

bull ASSUMED STORM DURATION 3 hoursmiddot

bull ASSUMED NUMBER OF WAVES 800

bull MEDIAN HS FOR STORM HOURS 1 2 AND 3

bull MEDIAN CURRENT VALUE bull

bull DETERMINE ULTIMATE CAPACITY FOR THREE DIRECTIONS

bull DETERMINE ANNUAL FAILURE PROBABILITIES

Broadsidemiddot Diagonal End-on

0037 0025 0007

0071 0044 0018

0022 0016 0004

DETERMINE TOTAL ANNUAL PROBABILITY OF FAILURE

Pf - 007

Pf - 013

Pr - 004

bull

EXAMPLE USE OF BIAS FACTORS - PHASE II middot

bull DETERMINE PREDICTIVE PROBABILITY OF FAILURE

Case - 1

Pf _ S S Pf (b b2) fa1a2 (b b2) db db2

g bull -1 (b1 R I S) - 1

92 - (b2 R2 I S) - 1

Case -2

Pf s ~ [bJ f11 (b) db

Pf [bJ p [g lt 0)

g - bull 1 (b1 R1 I S) - 1

1

bull1) l

I

BENEFITS OF ANDREW JIP bull PHASE II

lt

bull IMPROVED LOAD AND RESISTANCE RECIPE

bull IMPROVED CALIBRATION PROCEDURE

bull MULTIPLE BIAS FACTORS

bull PROCEDURE TO INCLUDE EFFECT OF OTHER PLATFORMS

bull PROCEDURE TO USE MULTIPLE BIAS FACTORS

  • Structure Bookmarks
    • ANDREW PHASE II JOINT1 INDUSTRY PROJECT
    • PROGRESS MEETING NUMBER 1
    • ANDREW PHASE II JOINT INDUSTRY PROJECT PROGRESS MEETING NUMBER 1 NOVEMBER 16 AGENDA
    • ANDREW PHASE II UST OF PARTICIPANTS
    • ANDREW PHASE I OBJECTIVES AND RESULTS
    • ANDREW PHASE II PROJECT OBJECTIVES
    • SUBSEQUENT PROJECTS
    • ANDREW PHASE II PROJECT TEAM
    • ANDREW PHASE II UPDATE MODELING middotAND ANALYSIS RECIPE
    • -ANDREW PHASE II CONDUCTOR RESISTANCE
    • ANDREW PHASE Il WA VE FORCE PROFILE
    • ANDREW PHASE II WA VE FORCE PROFILEmiddot
    • ANDREW PHASE II BRACE MODELING
    • ANDREW PHASE II FOUNDATIONS
    • ANDREW PHASE II middot middot JOINT MODELING
    • ANDREW PHASE Il JOINT MODELING OPTIONS DISCUSSION
    • ANDREW PHASE II JOINT MODEJJNG OPTIONS ILLUSTRATIONS
    • ANDREW PHASE II JOINT MODELING PHASE II METHOD
    • ANDREW PHASE II ANDREW STORM CONDmONS UPDATED
    • Platform Selection bullbullbull1111111111111
    • BasisCriteria for Selection bullbullbull1111111111111
    • CALIBRATION
    • WHERE TO INTRODUCE BIAS FACTOR (B
    • PHASE I -KEY CONCLUSIONS
    • bull CALIBRATION ISSUES PHASE II
    • PHASE II WORK SCOPE (CALIBRATION)
    • SOFTWARE
    • SENSITIVITY STUDIES POSTERIOR DISTRIBUTION OF B
    • WEIGHTING APPROACH bull
    • INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON
    • CATEGORIZATION OF ADDITIONAL PLATFORMS
    • UPDATING OF CALIBRATION RESULTS
    • APPLICATIONS OF BIAS FACTORS
    • EXAMPLE USE OF BIAS FACTORS
    • EXAMPLE APPLICATION bull PHASE I PLATFORM ST 151 K
    • EXAMPLE USE OF BIAS FACTORS -PHASE II middot
    • BENEFITS OF ANDREW JIP bull PHASE II
Page 7:  · . ' ANDREW PHASE II PROJECT OBJECTIVES . ' • ·UPDATE MODELING AND ANALYSIS RECIPE· JOINTS ' ' ·BRACES WAVE PROFILE FOUNDATIONS . ' ' CONDUCTORS· • ANALVZE PLATFORMS USING

I I It

ANDREW PHASE II

PROJECT TEAM

bull R L SMITH PROJECT MANAGER

bullbull F middot

J PUSKAR PROJECT ADVISOR

bull Rw LITTON TECHNICAL MANAGER

bull RAGGARWAL TECHNICAL DEVELOPER

bull D K DOLAN TECHNICAL ADVISOR middot

bull C A CORNELL RELIABILITY CONSULTANT

bull F MOSES RELIABILITY CONSULTANT

bull middotMSL ENGINEERING LTD TUBULAR JOINT CONSULTANT

q gtI

ANDREW PHASE II

UPDATE MODELING middotAND ANALYSIS RECIPE

bull CONDUCTOR RESISTANCE

bull WA VE FORCE PROFILE

bull BRACE MODELING

middotbull FOUNDATIONSmiddot

bull JOINT MODELING

~ I I

-ANDREW PHASE II

CONDUCTOR RESISTANCE

shybull RESISTANCE NOT INCLUDED IN PHASE I

bull CONTRIBUTION TO OVERALL RESISTANCE A PHASE II

STUDY ISSUE

middotbull middotCOMPARE FOR 4-LEG AND 8-LEG PLATFORM

bull CONSIDER CONDUCTOR POSmON IN GUIDES

- 13 LEFT 13 CENTER 13 RIGHT

Project ST151K Model Conductor version 1 Mon Nov 14 143140 1994

- ---- ---

--

117 s

111

7i

31

0 -5 -~ - ~()

- iltgt --lo

-(pO

-so

bull bullf OC)

-120

-ISo

-ISO

z

CAgt ~x Singl_e Conductor Model

bull

=I=

-_

-_

-gtshy

~ -

~

I

I~

- gt

~-----------____________________

Project STlSlK Model dir_2 version 7 Mon Nov 14 124200 1994

177 s

Ill

74

37

0- - e -zo - ~o

- Jo

-100

-10

-ISO

z

UV 1-+x

Project STlSlX Model Conductor Version 1 Mon Nov 14 140620 1994

bullmiddot

I

I

I I

I

I I

I I I I I

1 ==

-L

-L - L

-

- L

-

-

L

-

Uniform Displacement step 6 Inelastic Events Legend

Elastic

Strut Buckling Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield Beam Clmn FUlly Plastic Fracture

--

i~~~~~~~~~~~---~--~~-J___--JJ~_middot~~~=-

Project ST151K Model Conductor Version1 Mon Nov 14 142735 1994

---

~

I

I

middot

I

I

I

- shy- shy

- shy

- shy

4x z

CIV Uniform Displacement step 33 Inelastic Events Legend

bull Elasti~ Strut Buckling Strut Residual Strut Reloading Plastic StrutNLTruss Beam Clmn Initial Yield Beam Clmn-Fully Plastic Fracture

gt bull

Project STlSlK middot Model Conductor Version 2 Mon Nov 14 142025 1994

-----------------

bullf

I

I

I

II I I

I I I I

I ==

-

Variable Displacement step 9 Inelastic Events Legend

Elastic Strut Residual

bull

Strut Buckling Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield Beam Clmn Fully Plastic Fracture

--

Project ST151K Model Conductor Version 2 Mon Nov 14 142145 1994

I I I I

I I

I

I

I - - -

-Ishy-

-

Variable Displacement step 35 InelasticEvents Legend

Elastic Strut Buckling Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield Beam Clmn Fully Plastic Fracture

Project ST151K Model Conductormiddot

Version 1

Clgt - Cut Plane Force Fx Mon Nov 14 140051 1994

- -_ vii I fOIZM

0 _

~

N ID 0 bull 0

x 0

middotrt ~ () bull u 0 ~ 0 Ir

N bull

0

0 1 2 3 4 5 X Deck Disp [Ft)

I I 1 1 WbullTI- bullbull

lltcxi lr-C

011lt~~~~~~~---~~~~~~~~L~~~~~~~--~~~~~~~~i~~~~~~~--

Uniform Node Displacement

shy

Ji

r

Comparison of Pushover Curves for ST 151K Variable Load Pattern and Monotonic Load Pattern (Andrew)

Soll Improved 100

6000

5000 -i 4000

f

llS

3000 Ul

J 2000

10 c11 c1lgtolts 1000

0 v

bull ~

f

_ ----- --- ---shy

bull --Variable Lold

- - Andrew Load

I I I I I I

4 5 62 30 1

Deck Dlsplacement (ft)

bullbull r 1 II

~

ANDREW PHASE Il

WA VE FORCE PROFILE

bull DECK WA VE FORCE ~

PHASE I USED SECTION 17 (1056H degCd)

PHASE Il USE REVISED SECTION 17 (1000H Cd)

bull PHASE I USED SINGLE FORCE PROFILE FOR PUSHOVER

- USED WAVE HEIGHT CLOSE TO ULTIMATE CAPACITY

bull PHASE J

Il WILL STUDY THE

USE OF VARIABLE

PROFILE ~

- IF FORCE LEVEL INCREASES (OR DECREASES) BY A FACTOR EQUIVALENT TO A 2 FOOT CHANGE IN WA VE HEIGHT THE INCREMENTAL FO~CE PA1IERN IS CHANGED ACCORDINGLY (CAP FEATURE)

- INITIAL EXAMPLE

f~

a

middotn

a

I N -a a a amiddot

+

~ a

l

a w gt 0c 0 gt a I shyc -CJ wc w gtc(

== c( I shyZ w E W a 0 z- 8

ii sect

~ ~ bullT 0

I if

x ~

bullbullbull

II ~------------________

ANDREW PHASE II

WA VE FORCE PROFILEmiddot

CAP AUTOMATIC APPLICATION FEATURE

STAT

fR1 0 lt f lt1

STAT

STAT

f(R3-R2) middot f lt1

Project ST151K Model Wave_Loading Version 1 rue Nov 8 164107 1994 PT

E~ nss FT

SL 14amp FT

a 111 FT

EL 14 li

E-1 37 FT

El o FT

1middot1middot ______________ _ ________ __ _ __

0

200 400 600 800 1000 12000 Force at Elevatloii (kip)

200

a ll_

Wave Force at Major Elevations for Discrete Wave Heights

180

bullHmiddot40ft

Hmiddot495ft

H bull 51 ft

Hmiddot55ft_

H-59ft I

Andrew (H bull 60 86 ft) II

i I

_11---__

- middot- Hmiddot30ft

-

- bull -

middot bull bull bull middot

---Hmiddot53ft

bullbullbullbullHmiddot57ft

-----H-63ft

Elevation Above Mudllne 100

(ft)

180

140

120

80

60

40

20

Loading at Major Elevations

---H-49511

-deg-Hmiddot608611

0 200 400 600 800 1000 1200

Force It Elevatlon (kip)

--

Base Shear vs Wave Height ST151K

6000

5000 1---------Ult1mateCapacitybull 4890 kip (100 Improved Soll)-----------~--

4000a a lii ~ 3000

81 ampI

2000

1000

0 +-~~~--i~~~~~~~~~-+-~~~--lf--~~~~~~~~-+-~~~--l

30 35 40 45 50 55 60 65

Wave Height (ft)

bull middot

~

bull

II

2

g

~

It)

0

cCD E 0 E a c E euro~ CDshygtItgtmiddoto ~ti c -a Gic CD gt ~

-= c -Cll

I z

bullO 0i i ~ g 0

() CI -bull(q-ll) iuawow llu1wniiaAO

11

shy

116

114

112

g 110

bullc = 108

~ ~ 106 -sect 104

102

100

98

96

-

6555 60

Wave Height vs Center of Action ST151K

30 35 40 45 50

Wave Height (ft)

t

Comparison of Pushover Curves for ST 151K Variable Load Pattern and Monotonic Load Pattern (Andrew)

Soll Improved 100

--- --------- --- ----6000

5000

- 4000 8 - I

1 3000 Ul

81 m 2000

_J -shy

1000

---Variable Load

- - Andrew Load

0 f-~~~~~r--~~~~-t-~~~~~-t-~~~~~+-~~~~----t~~~~~-1

0 1 - 2 3 4 5 6

Deck Displacement (ft)

-middotshy

1 t I I ~

ANDREW PHASE II

BRACE MODELING

bull MARSHALL STRUTS USED FORPHASE I

bull PHASE II WILL ASSESS DIFFERENCE IN USING BUCKLING BEAMS AND MARSHALL STRUTS

middot bull COMPARE FOR 2D FRAME

bull COMPARE FOR PLATFORM 177 WHERE LOWER

BAY EXHIBITS LARGE BRACE MOMENTS

Project andrew Model beamframe Version 1 Mon Nov 14 145447 1994

~

middot Frame Test buckling beam-column model

-

Project andrew Model strutframe Version 1 Mon Nov 14 145754 1994

bull

Frame Test strut model

middot-middot _____________--_

_middotmiddotr_~ --ll--~-middot__

Buckling Beam bull

Strut (K=O 65)

Lateral Displacement

ANDREW PHASE II

FOUNDATIONS

bull ~LICIT NONLINEAR API FOUNDATION USED INiPHASE I

bull FOUNDATIONS

JIP SELECTED MINI-VANE DATA FOR

STRENGTH PROFILE

bull COMMENTS ON BENCHMARK COMPARISONS

- SIGNIFICANT VARIABILITY DUE TO FOUNDATION MODELING METHODS

II Ii

z

ctV 4x SPlO caissonfy-42s soil-api(lf-097step-114)

Inelastic Events Legend Elastic Strut Buckling

Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield

--------shyBeam Clmn Fully Plastic Fracture

~--~~~~~~~~~--~----~__i~~---~~~~__--~

Pro)ect apipile Model South Pelto 10 Version Wed May 18072235 1994 - -

1bullt

+

t

f i bullbullbull

bullbull bullbull

I I

5

~x SPlO Caissonfy=42s soil=api(lf=lOOstep=137)

Inelastic Events Legend

Elastic Strut Bucklilg

Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield

earn Clmn Full~ Plas1c Fracture

ANDREW PHASE II middot middot

JOINT MODELING

middotbull API RIGID-PLASTIC USED IN PHASE I

a ELASTIC-BRITTLE USED IN PHASE IB

amiddot MSL WILL PROVIDE NONLINEAR FORCE-DEFORMATION

AND MOMENT-DEFORMATION RELATIONS FOR PHASE II

a MODELING OPTIONS FOR PHASE II -

l

- UNCO~LEDmiddotONE DIMENSIONAL MEMBER END JOINT MODELING

J Ii

ANDREW PHASE Il

JOINT MODELING OPTIONS DISCUSSION

bull JD COUPLED INTERACTION SURFACE

- IDEAL

- DATA NOT AVAILABLE TO CALIBRATE

bull 2D COUPLED INTERACTION SURFACE

- SIGNIFICANT MODELING IMPROVEMENT

- DATAAVAILABLE

- IMPLEMENTATION PRACTICAL BUT NEEDS DEVELOPMENT AND CALIBRATION

bull 2D UNCOUPLED INTERACTION SURFACE

- SIGNIFICANT MODELING IMPROVEMENT

- DATA AVAILABLE

- IMPLEMENTATION PRACTICAL AND AVAILABLE

ANDREW PHASE II

JOINT MODEJJNG OPTIONS ILLUSTRATIONS

P

e

ANDREW PHASE II

JOINT MODELING PHASE II METHOD

e

I I I

ANDREW PHASE II

ANDREW STORM CONDmONS UPDATED

bull -PHASE I USED OCEANWEATHER HINDCAST-(Nov 1992)

bull middotmiddotPHASE II WILL USE THE REVISED HINDCAST (Nov 1994)

bull lt c lt I lt

2-2

l---4-~~~+-~~-+-~~~1--~~-+-~~--l--~T--+-~~--i24

f-tJVtMfiz~ 1Cfl (r STvb(

Plollol bullbull 5middotNOYmiddot9214bull1J Irbullbull Illbull [ANDREVl~NDREVCUSTbullMAXll 5-NOY-1992 10bull54 - 9208

HEIGHT lml

I~ MAXI

I L 1~middotamp SZ~- ~A I 13c

I J) ( r I X~0iWJ~0gtJ gt L gt gt I 1c I 120

-- N

~ deg

I I I I I I e C I ~=-41 122 -94 -92 -90 -88 -86 -84 -82 -80

-

Figure 7 contours of maximum hindcast significant wave height

_

~

~tl ~

NOltNttIJ 17lt ltNS S iJJ)y

Maximum Significant Wave Height - Hurricane Andrew 1992

30 ~

26 ~ I ~-~~~- == -shy -~ 35-_

24deg1--shy --------~- middot-middot-------middotmiddotmiddot----middot------middot-middot~middot- - - ~-----1---middot

22middot ~~~~~~~-~ -94 -92 -90 -88 -84deg-86 -82 -somiddot

c

-

Platform Selection bullbullbull1111111111111

0 BasisCriteria for Selection middot

0 Review of Phase I Platforms middot 0 Alternative Platforms

0 PMB Recommendations

0 Participants Comments

gt

L

l I

BasisCriteria for Selection bullbullbull1111111111111

0 Benefit for Calibration - Classification (UnexpectedExpected)

- Failure Mode (StructuresFoundations)

- Damaged StructuresKnown response

- Better Representation of Full ~opulation

0 Availability of Data - Existing Models

- Inspection Data

- Damage Data

PREDICTED BEHAVIOR

-

OBSERVED

BEHAVIOR

-SURVIVED DAMAGED FAILED

SURVIVED

ST134W WD90A

-MC311 -

MC397 -

--

-

-

ST151K ST130Q

-

-

-

DAMAGED - -

-

-

--

-

-

ST52 SS139 ST177B -

ST161A

-_

-

FAILED

-

-

-

ST151H ST130A ST72

Operator - Platfonns - Platfonns Survived Damagedamp

- Falled -

bull

-

-

Jle eBDIGIPllDbl -Amocoshy 30 1 Chevron -

143 12 Exxon 72 Mobll 22 1 Phllllps 7 bull

-Shell 39 Unocal 25 1

Total Partlclnts Platfonns 338 15

Other Operators bull- 310 13

-

Total Platfonns - -

648 28

fbml

Platforms Considered 6 7

tie of Total Platfonns 1 25

OVERVIEW OF PLATFORMS SURVIVEDDAMAGEDFAILED

-

shy

v

Platforms Evaluated in Phase I

-

-Platfonn Water Year Number Analysis Avallabillty of Availability Platform

- Name Depth rl of Model DetaDed rl ClassUlcatlon

Installation legs Inspection Soll Report Based on Information Phase I Calibration - -ft

- -- Suaill flalfnrm ~BSi -STISIK 137 1963 8 CAP Yes Yes Un eel Sarvlval

-STIJOO 170 1964 4 CAPmiddotshy Yes ST134data u ST134W 137 1981 4 CAP Yes Yesmiddot E

-

edSarvlval -

middotsurvival

WD90A 184 - 1964 8 CAP Yes E -

- Sarvlvill

MC3ll 343 1978 8 KARMA

MC397 468 1991 4 KARMA -

bull ~reglmmiddot- -

Yes Sure Survival

bull Yes - Sare Sanlval-

-T23STS2 63 1969 4 CAP

T2SSS139 62 1969 4 CAP -

Yes

Yes

Yes Ex

Yes Une

bull SurvivalLikely to DamalI

middot bull SanlvalLlkely to DamalI

ST161A 118 1964 8 USFOS

Eail11a pound1al(bullllm Clss ST177B 142 --1965 8 CAP

STISIH 137 1964 - 8 CAP C

STIJOA 140 19S8 8 CAP - -

Yes

Yes

ollamed

Yes

- Yes E

-

at about 1 mUe

Yes

ST134data

bull SafvlvalLlkelv to Dam- bull- -

E middotrahre

Emeried Failure -

Likely FaUare

T21ST72 61 1969 4 CAP Collamed No Likely FaUure

Platforms Identified for Evaluating Their Likelihood of Falling Under UnexpeCted Survival Category

CllEllATOa

AllB

llLOCll

Sl1lvcnJllE

NAME

YEAR

INSTAUJCD

WATER

IJBPJB

middotmiddotDISTANCE

ftOMSllOllE

(MIUIS)

Ill

(M)

AMOCO BllW

STllI

WDI075

A

bull D

4

n uo 17J

44

32

bull 20

10

bull CllEVllOH STtllO c I 1bull 21 11

STl130 D 0 161 21 11 STllJ4 I 0 137 D - 11

STIUI G I 137 32 11

STIUI I 0 121 32 11

STIUI

J 0 140 32 11

STIUI It 4 137 32 11

STllSI PROl)I 0 137 32 11

STllSI PROD-2 0 137 32 11

STll7 A 0

140 35 11

WDl117 c 65 214 34 10

DXON STshy E 51 14 bull STl165 A t3 bull 34 WD0073

WD0073

WD0073

A _ c

4

62

4

IA IA

17J

22

22

22

bullbullbull WD003-shy E 65 161 Z7

A 0 IN 31 MOBD Bl12I

ssn A bull

50

50

31

7

bull 7

MIJllPllY SS114 B 7 50 7 7

PlllUIPS SSl14t A 55 33 7

SAMEDAN ST17J A 4 117 7 bull 7

SJIELL WD0103 A 65 223 27 WDllOJ bull 65 221 27 WDll04 c 65 221 27 WDOI04 D 67 2 27

UNOCAL

SSl20I

SSl20I

bull SSl20I

SS0209

SS0209

SS0209

SS0209

SS 0253

SS 0253

E

I

PtJllE

A

bull D

G A

c

0

4

4

7J

61

0

0

69

103

97

97

9S

100

95

100

165

175

34

34

34

34

34

34

34

54

54

bullbullbullbullbullbullbullbullbull

Platforms Recommended for Phase II

llalr y_ w- Aot Av- _ I 1 Doplla Model nm bull bull _ bull I I H11 s18-1 Irht lbwl~ bullbullbull(-illtdlel OM lllMtD

II ---

-STISIIC 137 1963 I CAP Yeo Yeo I Yeo

STtshy -170 1964 4 CAP Yeo ST134dlU I Yeo

STl34W 137 Yeo Yeo -1981 4 CAP D Yeo

WD90A 114 1964 I CAP Yeo - D

MC311 30 1971 I KARMA Yeo -sre-

MC397 middot 461 1991 4 ICARMA Yeo Sireshy - - r- -

__ T13ST5l 63 1969 4 CAP Yeo Yeobull middot- ~n- D Yeo _ Tl5SSl39 62 1969 4 CAP Yeo Yeo I Yeo

_ STIT711 142 1965 I CAP Yeo I Yeo middot-1shySTl61A Ill 1964 I USFOS Yeo Yeo - D bulloshy

- r-

STISIH 137 1964 I CAP y D y -_ STl31A 140 1951 I CAP Yeo STl34dlU I Yeo

T21Sll 61 1969 4 CAP No rar-middot I Yeo

~-rmiddot-middot SS209A 95 1967 16 CAP Yeo D -SSU4R

11 UI A 41 -in D SS 2741AIM CAP bull

_ _ -shy - -shy- -shy

CALIBRATION

bull PHASEIAPPROACH

bull CALIBRATION ISSUES

bull PHASE II WORK SCOPE

bull SOFlWARE AVAILABILITY

bull EFFECT OF ADDITIONAL PLATFORMS

bull WEIGHTING APPROACH

bull EXAMPLE APPLICATION middot

(l=Ofce Mode] Wind

~

fcurrent l -middot

~SW ~leldl4

ultimate resistance force I first

design load Illgt

i _

deflection

-

bull

4

Historical Hurrlcane Dariiage and Survival of Platforms

Components Required for Calibration (Petrauskas 1992)

PRIOR Distribution

PLATFORM EXPERIENCE

BAYESIAN Survivals and Failures

UPDATING

Waves Winds and Currents bull

POSTERIOR Distribution

- A_middotJA ~ Survival Data Failure Data

Result Result

Load Resistance

middotfl

-

Bayesian Updaing middotGeneral Description (Petrauskas 1992)

L Imiddot

WHERE TO INTRODUCE BIAS FACTOR (B

On Resistance (Rmue - B (RPREDICreD I

middotOn Loads (Smue middot - B (SPRE01cTEo

On Safety Margin (RSmue - B (RSPREDICTED

CALIBRATION METHODOLOGY

gt

r

FIELD OBSERVATIONS

Observed Behavior

survivals Damages

Failures

r

bull

CAPACITY ANALYSIS

Predicted Load and Resistance Predicted Failure Modes

Survival Damage Failure cases

RELIABILITY ANALYSIS

- Probability of Failure ~

Ukellhood Failure

BAYESIAN UPDATING

Prior Distnbution of B Combined Likelihood Function

Posterior Dlstnbution of B

BIAS FACTOR

Mean Value cov bull

Represents Modeling Uncertainties

Prior Distribution of B Likelihood Functionmiddot Observing Success case STISIK and Posterior Distribution of B

2 00

uo

160

140

120

e 100 L $ uoIshy

0 60 ~ ~ 040

020

0 00

0 00 0 50

_ I 00

b

I 50 2 00

-shy lk [bl STISIK)

--shy Pnor-f(b)

--shy Pos1enor r(bl ISIK)

Prior or B Mean bull 10 COV bull 030

Posterlor of Bbull Meanbull 136COV 0 16

2 50

Likelihood Function and Posterior Distribution Example Platform STISIK

- -

~

- Prior and Posterior Distributions or B Failure cases (ST177B STlSlH ST130A T21) Damage cases (T23 T25 ST161A) Success cases (STlSlK ST130Q ST134W WD90A MC311 MC397)

400

350

300

2 so 0-

I

bull bull

bull bull gt

- bullbull ~ bull bull I bull

_ bull bull bullbull I bull bull bullII V -

bull bull

I bull ~ bull

bullbull bullI bull bullI bull bull bull

middot -~

bull__ ~ middotI middotmiddot

L 2 00

e ~ I SO

-I 00

- 0 so

000 gt

0 00 020 0 40 060 0 80 I 00 120 140 160 110 200

b -

-

shy

Prior-r

- - - - Posterior fLlll failure cases

bull bull bull bull bull bull bull Posterior flall damage cases

- bull bull - bull middot Postenor flall success cases

Posterior flall success fulure damage cases

-

Posterior of B Mean= 119 COV 0101

Posterior Distribution or Blas Factor (8) bull Different Categories and All Cases Combined

shy

l

PHASE I - KEY CONCLUSIONS

bull GLOBAL CORRECTION FACTOR

bull 15 TO 20 CONSERVATISM MEAN)

bull UNEXPECTED SURVIVALS HAVE GREATEST IMPACT

bull DAMAGED CASES MORE INFORMATIVE

bull CALIBRATION ISSUES PHASE II

bull

EFFECT OF IMPROVED CAPACITY ANALYSIS RECIPE

bull APPLICABILITY TO MULTIPLE FAILURE MODES

bull NO FOUNDATION DAMAGE CASE INCLUDED

bull IMPROVED REPRESENTATION OF FULL ANDREW DISTRIBUTION

-bull NUMBER OF PLATFORMS CONSIDERED

I 1 I

I t ) bull ~

PHASE II WORK SCOPE (CALIBRATION)

bull

bull ESTABLISH MULTIPLE BIAS FACTORS

Use Nine Base Case Platforms

middot - Capacity Analysis for Revised recipe

Foundation Blas Factor from Foundation JIP

bull DEVELOP WEIGHTING APPROACH

bull UPDATE MULTIPLE BIAS FACTORS

bullEffect of Platforms Not In the Nine Base Cases

bull EXAMPLE APPLICATION OF BIAS FACTORS I

bullbull

PHASE I PHASE II

CAPACITYbull ANALYSISRu bull

---------------------1----middot

Jc1 - iJbullraquoJ dbull

FttlS frt[f~l ratdhJJ

-

LIKELIHOOD FUNCTION

DEVELOPMENT

lk Jb Jfall1N) bull Jb)

111 Ill Isurvtvoq bull 1 - bl

-middot -------------------shy ~---------------------middot-middot

r111ir8 Cb1 111 Cbl BAYESIAN

UPDATING

~cove

R1R2 bull middot---------------------~----

Jblbull P(g cOJ

1 bull JbRI (MBSI I bull 1

~middot

llcJllJfalluro) bull P(gcOJ Ill 111 b21fallun) bull P ((g1 CO) U 1112 c OD

middot--------------------shy ----------------------shy bull

r111111r8 Cbl 111 Ill i-1nro1

deg ~1b2)laquottfrr (b1b21-lnfo) re Ill bull 1-11middot b2) bull lodegamp

middot

i1 covamp i2 cov~i

-

shy

--

Likelihood Function vs b All Success Case STISI K STl30Q STl34W WD90A MC311 MC397

I 0000

0 9000

OIOOO

0 7000

~ 0amp000

ta osoOO

~ 0 4000

0 3000

0 2000

0 1000

bull 0 0000

bull

I

middot

middot bull I

Ibull

-sibull ~ -middot-middot-middot-middot -middot-

----7----+-----~ I

Ibull bull bull I

rbull

--- Jk ISIK[blsuccess)

- bull - bull - lk 130Q[blsuccess) bull

lk 134W[blsuccess)

Jk-WD90A(blsuccc~I

- - - lk MC397[blsucccss)

--ltgt-- lk middotMC31 l[blsucccss)

--- Jk [bl 6 out of6succcsHascs)

05 I 5 2 250

b

Likelihood Functionsmiddot Success Cases - Phbullsc L

shy

bullbull bull bull bull bull bull -------------------~

bull bull bullbull bullbull bullbull bullbull bullbull bullbull bull

bull bullbull

bullbullbullbull

bullbull

bullbullbullbullbullbullbull

bullbullbullbullbullbull

bullbullbullbullbullbullbull

z 0

t z ) LL

c 0 0c i ~ i

J

~ sect ~--------------~ e l5

I

I

SOFTWARE

bull PF PROGRAM

bull CALREL

bull PF +CALREL

bull RELACS

bull

gt i

l

SENSITIVITY STUDIES

POSTERIOR DISTRIBUTION OF B

bull EFFECT OF CHANGE IN DISTRIBUTION OF PRIOR OF B

bull EFFECT OF REMOVAL OF ONE PLATFORM FROM EACH CATEGORY

bull EFFECT OF REMOVAL OF THREE DAMAGE CASES

bull EFFECT OF ADDITIONAL 13 SIMILAR PLATFORMS

-I

Prior and Posterior Distributions of B Variation In COV of PriorofB fromOl to04 -- -

4500

-4000

-

- -

3500 --middot - bull bull - bull bull Prior-r ICOV =02

3000

- - - - Prior-r ICOV =03 -

bull -~ e 2soo -

bull bull bull bull bull bull bull Pnor-r ICOV =o 4t frJ ~ 2000 Posterior rlCOV =0 2

bull bull Postenor rlCOV =0 3 bull bull 1500 bull middot bull - shy -middot- Posterior rlCOV =04

middot~ I 000 - bull

-middotmiddot ~ ~middotmiddot M -~

- middot 0 500

wbullbullbull _ I middot~ middotmiddotmiddotmiddotmiddot middotmiddotmiddotmiddotmiddot - -~ ~ __ middot_ _ bull -shy0 000

0 000 0200 0 400 0 600 0 800 IOOO 1200 1400 1600 1800 2000 -

- - b

-

=shy

Posterior Distribution of Bias Factor (B) bull Effect of Variation In COV of Prior of B

bull

bullPrior and Posterior Distributions of B Evaluating Effect of Removal of One Platform from Each Category from the Base Case of 13 Platforms

400

~ ~ 350

j IA i

3 00 bull I bull I ---Prior-r I bull l

250 1 I 1

I bull bull ---Posterior r lall 13 casesshy- ~E 2 00 bull bull bull bull bull bull bull Posterior r without STISIK

- ~ 1 bull - bull I - bull bull - bull bull Posterior r Iwithout T2I I 50 I

J-tt~i - bull - bull - Posterior r Iwithout T2S

tOO [-

I I ~ ~ I 0 50

0 00

___ r J i

l~T

~--

0 00 0 20 0 40 0 60 080 I 00 120 140 160 110 2 00

b

zshy

Posterior Distribution of Bias Factor (8) bull Effect of Removal of middotPlatform from Any Category

bullPrior and Posterior Distributions or B Evaluating Effect or Removal or 3 Damage Cases from the Base Case or 13 Platrorms

4

35

3

25

e L 2 ~ bulle shy

I 15

bull I

05

0

0 02 04 06 I 2 I 4 16 IS 2

I I

I

I

I Ibull

J -tmiddot L--

bullI

~

I

fl

II ~

I I v I

middot bull bull

bull L

~ ll- I-~lt

OS I

b

I

bull

1middot bullbullbullbull

~ __

bull

- - - - Prior-r

- - - - Postenor flall failure cases

Postenor flall success cases

--- Posterior flall 13 cases

---Posterior fI success failure cases only

IPosterior or B Mean =t11 COV =0121

-

_I

Posterir Distribution or Bias Factor (B) bull Effect or Removal or 3-Damage Cases

-

bullPrior and Posterior Distributions or B Evaluating Effect or Additional 13 Similar Observations as

Used in Andrew JIP The POsterior or B from Andrew-JIP Considered as New Prior or B

- 500

450

400

350

bull bull

bull bull bull bull J bull

bull bull f ~ I

I ~ Ibull bull

bullbull bull

l bull I -bullbull bull bull

I

I bull bull

bull I I~~bull bull bull

bull ~

I bull bull

~ ~_ ~~~

- - New Prior r bull

Posterior 114 new failure cases ~ 300- bull bull bull bull bull bull bull Posterior f 13 new damage casesl 150 ~

- bull bull - bull bull Posterior 116 new success cases i 2 00

Postenor 1113 new success failure -150 damage cases

I 00

0 50 -

0 00 0 00 0 20 0 40 0 60 0 80 I 00 120 I 40 160 110 2 00

Posterior ol B Mean= 120 COV= 0071 b

shy

shy

_rshy

Posterior Distribution of Bias Factor (B) bull Effect of Additional 13-Similar Observations

1

1 I

WEIGHTING APPROACH bull

bull ASSESSMENT OF PLATFORMS DATA

Select Platforms Most Mappable to the Analyzed Platforms

Unlikely SurvivalDamage Cases Identification

Based on Cut-Off Criteria Agemiddot Hs etc

bull INFORMATION REQUIRED FOR COMPARISON

bull CATEGORIZATION OF ADDITIONALmiddot PLATFORMS

bull UPDATING OF CALIBRATION RESULTS

--

OVERVIEW OF PLATFORMS SURVIVEDDAMAGED -~

Operator -

- Platforms

Survived CS)

Platforms

Damaged CD)

Platforms Used In

Phase IJIP -

JI~ ~Bill~IRBDll -

Amoco -

Chevron -

Exxon

Mob II

Phllllps

Shell

Unocal -

30

143

72 22

7

39 25shy -

1

12

bull

1 bull

1

1(S)+1 (D)

3 (S) +3(D)

2 (S)

bull shy

bull

bull

bull

Total Partlclnants Platforms 338 - 15 10C6S+4D) -

Other Operators

310

13

3 (D)

Total Platforms 648 28 13C6S+7D) -

Summary of Andrew JIP bull Phase I Platforms

Plairshy w v- N-ber Lea-le 1 shy car Hind_ o-w N-bull Deplh fl fl IJpe la bull n Yeadon Jadtet lbelledtf liefalit -- _ -OD

IL H - -shy _ICllshy IL IL - - rQClto ~ _

K STUIK 137 1963 8 365 608S4 Cellardeck

-middot STl-shy 170 1964 4 -~ K 39 62269 Cellardeck - --ST134W 137 1911 4 43 608S4 Smvlnlshy- _ - -

WD90A 184 1964 8 K 267Sbull34 83 SOS31 SUboftllar deck SmvMI

MC311 343 1978 8 x S7 62 - MC397 468 1991 4 x shy S33bull6SO 65911 -shySmeSmvlvol -- gt -

r-- T23ST52 63 1969 4 ton- shyKshy 52594 S03 -T25SSl39 62 1969 4 Kshy SOS94 S063 - middot ton-shy__ -

bullbull a ton--shyST161A 118 1964 8 K - 347S 51495 Cellarclecti

r~

ST1771 142 1965 8 K 39 60167 CeDordeck failme

STISIH 137 1964 8 K 395 608S4 CeDardeck Failme

K ST130A 140 1958 a SS 6095 I lblv Pai1uno - T21ST72 61 1969 4 Kshy S4S94 4974 I lhlv Pai1uno -

- -

JE

INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON

bull WATER DEPTH

bull bull YEAR OF INSTALLATIONDESIGN

bull PLATFORM ORIENTATION

bull PLATFORM LATITUDE AND LONGITUDE

bull KEY SKETCHES OF PLATFORM STRUCTURAL FRAMING

Number of Legs Plies Leg Batter Number of Conductors bull Deck Elevations (Including sub-cellar deck If any) Jacket Framing Elevations Brace type In Vertical Frames Sizes of Primary Members (leg pile vertical frames braces primary horizontals) Joint Details (can gap etc) LegPiie Annulus Grouting Status

bull middotPOST ANDREW INSPECTION REPORT Significant Wave Height Hs (ft)

bull INFORMATION ON DAMAGES PRIOR TO ANDREW

bull RESULTS FROM IN-HOUSE STUDIES IF AVAILABLE

bull ~RIGINAL DESIGN CRITERIA INFORMATION IF AVAILABLE

bull SOIL SHEAR STRENGTH PROFILE IF AVAILABLE

Platforms Identified for Evaluating Therr Likelihood of Falling Under Unexpected Survival Category

OPlRATOR

middotADA

BLOClt

llmucnJllE

NAME

YEAll

INSTAUJD

WA1ER

DBPTll

DISTANCE

JlllOM SHOllE

(MILU)

Bl

(M)

AMOCO 111113

STl11

WDI075

A

bull D

St

17

120

1n

44

20

10 I

CllllVllQfl STlllO c 61 110 21 11 STIUO D Q 1110 21 11

STl134 p Q 137 29 11

ST0151 G 61 137 11

ST0151 I 3 121 11

ST0151 J Q 140 11

STl151 It 137 11

ST0151 PJl001 Q UT 11

ST0151 PROD-2 Q UT 11

ST017 A 3 140 35 11

WDIUT c 65 214 34 10

llXXON STI055 II 51 6T 14 I

STl15 A t3 34

WDI073

WD0073

Amiddotshy Q

1111 1111

22

22

I

I

WDI073 c 1n 22 I

- WDOCl93-shy II

65 1110 2T A 3 194 31 MOBU 111121

SS IOl7

A B

501

50

50

31

T

I

T 6

MtlllPllY SS114 B T 50 T T

PlllllIPS SS0149 A 5 55 33 T

SAMllDAN STln A 107 T T

SBllLL WDl103 A 65 223 2T WDl103 B 65 221 2T WD0104 c 65 221 2T WD0104 D 6T 20 2T

UNOCAL SS020I

SS 020I SSl20I

SS 0209

SS0209

II p

PURll

A

B

3

n 61

103

97

97

95

100

34

34

34

34 34

bullbull bullbullbull

- SS0209 D u 95 34 I

- SS 1209 SSl253

SS0253

G A

c

6T

2

100

165

1T5

34

54

54

bullbull I

I I

I I

SOUTH MARCH

~LAND

N

SOUTH PaTOI ENE ISWltD AREA

MMS Defined Limits ror Offshore Structures Requiring Inspections Following Andrew

t---Jbulln3 T25bull I bullTtl

bull1$1Htl1Abull bull1SIK1779bull

~

middot-DaTA AREA~T

CRANO rsLE

AREA bull1300

bullll4W

=shy

bullPath of Hurricane Andrew with Platforms Used In Calibration

Likelihood Function vs b All Success Case STISI K STIJOQ ST134W WD90A MCJll MCJ97

I 0000

0 9000

0 8000

07000

-Bo6000

tiiii 0 5000 -

04000

0 3000

0 2000

01000

0 0000

0 05 25

ri1rt middot-middot-middot-middot-middot-

I 5 2

b

---11 middot151K(blsu~ccsI

- bull - bull - fk 130Q(bfU~lCSS)

lie bull134W(blsuccc]

lk-WD90A[blsultCS]

- - - lk middotMC397(bfUtlCS]

--o-- lk-MCJll(bbuctcsJ

--- ll (bl 6 out of 6 Mlttcs case]

_

( Likelihood Functions middotSuccess Cases - p - 1 c I

CATEGORIZATION OF ADDITIONAL PLATFORMS

I I I

bull CATEGORIES

Unexpected Survivals

Expected Survivals

bull APPROACH

Identify Corresponding Platform(s) Evaluated In the JIP

Determine Max Wave Height at Platform Location

Evaluate Whether Wave Hit the Deck

middot - Evaluate Likelihood of R and S to Differ from the Base Cases

Review Joint Information

Assess Special Features

bull IDENTIFY NUMBER OF PLATFORMS IN EACH CATEGORY

I

UPDATING OF CALIBRATION RESULTS

bull INCLUDE EFFECT OF ADDITIONAL PLATFORMS ON bullBbull

Consider Increased 1Uncertalntles In R and S

Develop bullNeW Ukellhood Functions for Different Categories

middotLet bullxbull Number of Ukely Unexpected Survival Cases and y Number of Ukely Expected Survival Cases

Determine Combined Likelihood Function lk (b Iadditional platforms) middot

bull REVISED LIKELIHOOD FUNCTION

lk(b I all cases) = lk(b I 9 platforms) middot x lk(b Ix platforms) x lk(b I y platforms)

bull ESTABLISH REVISED BIAS FACTOR B

f8 (b) u f8 (b) lk (b I all cases)

II

t t ~

shy

APPLICATIONS OF BIAS FACTORS

ii APPLICABLE TO PARTICULAR LOAD AND RESISTANCE RECIPE

bull USE BY THE API COMMITTEES AND MMS

Criteria Revisions

Recommended Design Guidelines Revisions bull

bullbull REASSESSMENT STUDIES AND DECISION MAKING

middot

bull RISK ASSESSMENT STUDIES BY THE OPERATORS

Individual Platforms

Fleet of Platforms

middot IMR Program Implementation

bullbull NOT TO BE USED IN DESIGN

I I fl J

EXAMPLE USE OF BIAS FACTORS

bull PHASE I EXAMPLE APPLICATION

bull EVALUATION OF ANNUAL PROBABILITY OF FAILURE

Establish Annual Maximum Seastate Data

Determine Coefficients to Define Base Shear

Determine Ultimate Capacity of Platform

Determine Probablllty of Failure Versus bullbbull

-bull DETERMINE UPDATED PROBABILITY OF FAILURE

Pf - P [S gt R]

Pf I P [~ gt Rb I b] f 8 (b) db

Pr - J P [S gt Rb I b] f bull 8 (b) db

bull TOTAL PROBABILITIES OF FAILURE

Total Pf - 1 bull Il [1 middotPf] - l Pf bull

Total Pf - 1 bull Il [1middot Pf] - l Pfmiddot

Total Pr = 1 - n [1middot PfbullJ = l Pfshy

I I I

EXAMPLE APPLICATION bull PHASE I

PLATFORM ST 151 K

bull GULF OF MEXICO

bull WATER DEPTH = 137 FT

bull ANNUAL MAXIMUM HS MEDIAN 175 ft COV 031

bull ASSUMED STORM DURATION 3 hoursmiddot

bull ASSUMED NUMBER OF WAVES 800

bull MEDIAN HS FOR STORM HOURS 1 2 AND 3

bull MEDIAN CURRENT VALUE bull

bull DETERMINE ULTIMATE CAPACITY FOR THREE DIRECTIONS

bull DETERMINE ANNUAL FAILURE PROBABILITIES

Broadsidemiddot Diagonal End-on

0037 0025 0007

0071 0044 0018

0022 0016 0004

DETERMINE TOTAL ANNUAL PROBABILITY OF FAILURE

Pf - 007

Pf - 013

Pr - 004

bull

EXAMPLE USE OF BIAS FACTORS - PHASE II middot

bull DETERMINE PREDICTIVE PROBABILITY OF FAILURE

Case - 1

Pf _ S S Pf (b b2) fa1a2 (b b2) db db2

g bull -1 (b1 R I S) - 1

92 - (b2 R2 I S) - 1

Case -2

Pf s ~ [bJ f11 (b) db

Pf [bJ p [g lt 0)

g - bull 1 (b1 R1 I S) - 1

1

bull1) l

I

BENEFITS OF ANDREW JIP bull PHASE II

lt

bull IMPROVED LOAD AND RESISTANCE RECIPE

bull IMPROVED CALIBRATION PROCEDURE

bull MULTIPLE BIAS FACTORS

bull PROCEDURE TO INCLUDE EFFECT OF OTHER PLATFORMS

bull PROCEDURE TO USE MULTIPLE BIAS FACTORS

  • Structure Bookmarks
    • ANDREW PHASE II JOINT1 INDUSTRY PROJECT
    • PROGRESS MEETING NUMBER 1
    • ANDREW PHASE II JOINT INDUSTRY PROJECT PROGRESS MEETING NUMBER 1 NOVEMBER 16 AGENDA
    • ANDREW PHASE II UST OF PARTICIPANTS
    • ANDREW PHASE I OBJECTIVES AND RESULTS
    • ANDREW PHASE II PROJECT OBJECTIVES
    • SUBSEQUENT PROJECTS
    • ANDREW PHASE II PROJECT TEAM
    • ANDREW PHASE II UPDATE MODELING middotAND ANALYSIS RECIPE
    • -ANDREW PHASE II CONDUCTOR RESISTANCE
    • ANDREW PHASE Il WA VE FORCE PROFILE
    • ANDREW PHASE II WA VE FORCE PROFILEmiddot
    • ANDREW PHASE II BRACE MODELING
    • ANDREW PHASE II FOUNDATIONS
    • ANDREW PHASE II middot middot JOINT MODELING
    • ANDREW PHASE Il JOINT MODELING OPTIONS DISCUSSION
    • ANDREW PHASE II JOINT MODEJJNG OPTIONS ILLUSTRATIONS
    • ANDREW PHASE II JOINT MODELING PHASE II METHOD
    • ANDREW PHASE II ANDREW STORM CONDmONS UPDATED
    • Platform Selection bullbullbull1111111111111
    • BasisCriteria for Selection bullbullbull1111111111111
    • CALIBRATION
    • WHERE TO INTRODUCE BIAS FACTOR (B
    • PHASE I -KEY CONCLUSIONS
    • bull CALIBRATION ISSUES PHASE II
    • PHASE II WORK SCOPE (CALIBRATION)
    • SOFTWARE
    • SENSITIVITY STUDIES POSTERIOR DISTRIBUTION OF B
    • WEIGHTING APPROACH bull
    • INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON
    • CATEGORIZATION OF ADDITIONAL PLATFORMS
    • UPDATING OF CALIBRATION RESULTS
    • APPLICATIONS OF BIAS FACTORS
    • EXAMPLE USE OF BIAS FACTORS
    • EXAMPLE APPLICATION bull PHASE I PLATFORM ST 151 K
    • EXAMPLE USE OF BIAS FACTORS -PHASE II middot
    • BENEFITS OF ANDREW JIP bull PHASE II
Page 8:  · . ' ANDREW PHASE II PROJECT OBJECTIVES . ' • ·UPDATE MODELING AND ANALYSIS RECIPE· JOINTS ' ' ·BRACES WAVE PROFILE FOUNDATIONS . ' ' CONDUCTORS· • ANALVZE PLATFORMS USING

q gtI

ANDREW PHASE II

UPDATE MODELING middotAND ANALYSIS RECIPE

bull CONDUCTOR RESISTANCE

bull WA VE FORCE PROFILE

bull BRACE MODELING

middotbull FOUNDATIONSmiddot

bull JOINT MODELING

~ I I

-ANDREW PHASE II

CONDUCTOR RESISTANCE

shybull RESISTANCE NOT INCLUDED IN PHASE I

bull CONTRIBUTION TO OVERALL RESISTANCE A PHASE II

STUDY ISSUE

middotbull middotCOMPARE FOR 4-LEG AND 8-LEG PLATFORM

bull CONSIDER CONDUCTOR POSmON IN GUIDES

- 13 LEFT 13 CENTER 13 RIGHT

Project ST151K Model Conductor version 1 Mon Nov 14 143140 1994

- ---- ---

--

117 s

111

7i

31

0 -5 -~ - ~()

- iltgt --lo

-(pO

-so

bull bullf OC)

-120

-ISo

-ISO

z

CAgt ~x Singl_e Conductor Model

bull

=I=

-_

-_

-gtshy

~ -

~

I

I~

- gt

~-----------____________________

Project STlSlK Model dir_2 version 7 Mon Nov 14 124200 1994

177 s

Ill

74

37

0- - e -zo - ~o

- Jo

-100

-10

-ISO

z

UV 1-+x

Project STlSlX Model Conductor Version 1 Mon Nov 14 140620 1994

bullmiddot

I

I

I I

I

I I

I I I I I

1 ==

-L

-L - L

-

- L

-

-

L

-

Uniform Displacement step 6 Inelastic Events Legend

Elastic

Strut Buckling Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield Beam Clmn FUlly Plastic Fracture

--

i~~~~~~~~~~~---~--~~-J___--JJ~_middot~~~=-

Project ST151K Model Conductor Version1 Mon Nov 14 142735 1994

---

~

I

I

middot

I

I

I

- shy- shy

- shy

- shy

4x z

CIV Uniform Displacement step 33 Inelastic Events Legend

bull Elasti~ Strut Buckling Strut Residual Strut Reloading Plastic StrutNLTruss Beam Clmn Initial Yield Beam Clmn-Fully Plastic Fracture

gt bull

Project STlSlK middot Model Conductor Version 2 Mon Nov 14 142025 1994

-----------------

bullf

I

I

I

II I I

I I I I

I ==

-

Variable Displacement step 9 Inelastic Events Legend

Elastic Strut Residual

bull

Strut Buckling Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield Beam Clmn Fully Plastic Fracture

--

Project ST151K Model Conductor Version 2 Mon Nov 14 142145 1994

I I I I

I I

I

I

I - - -

-Ishy-

-

Variable Displacement step 35 InelasticEvents Legend

Elastic Strut Buckling Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield Beam Clmn Fully Plastic Fracture

Project ST151K Model Conductormiddot

Version 1

Clgt - Cut Plane Force Fx Mon Nov 14 140051 1994

- -_ vii I fOIZM

0 _

~

N ID 0 bull 0

x 0

middotrt ~ () bull u 0 ~ 0 Ir

N bull

0

0 1 2 3 4 5 X Deck Disp [Ft)

I I 1 1 WbullTI- bullbull

lltcxi lr-C

011lt~~~~~~~---~~~~~~~~L~~~~~~~--~~~~~~~~i~~~~~~~--

Uniform Node Displacement

shy

Ji

r

Comparison of Pushover Curves for ST 151K Variable Load Pattern and Monotonic Load Pattern (Andrew)

Soll Improved 100

6000

5000 -i 4000

f

llS

3000 Ul

J 2000

10 c11 c1lgtolts 1000

0 v

bull ~

f

_ ----- --- ---shy

bull --Variable Lold

- - Andrew Load

I I I I I I

4 5 62 30 1

Deck Dlsplacement (ft)

bullbull r 1 II

~

ANDREW PHASE Il

WA VE FORCE PROFILE

bull DECK WA VE FORCE ~

PHASE I USED SECTION 17 (1056H degCd)

PHASE Il USE REVISED SECTION 17 (1000H Cd)

bull PHASE I USED SINGLE FORCE PROFILE FOR PUSHOVER

- USED WAVE HEIGHT CLOSE TO ULTIMATE CAPACITY

bull PHASE J

Il WILL STUDY THE

USE OF VARIABLE

PROFILE ~

- IF FORCE LEVEL INCREASES (OR DECREASES) BY A FACTOR EQUIVALENT TO A 2 FOOT CHANGE IN WA VE HEIGHT THE INCREMENTAL FO~CE PA1IERN IS CHANGED ACCORDINGLY (CAP FEATURE)

- INITIAL EXAMPLE

f~

a

middotn

a

I N -a a a amiddot

+

~ a

l

a w gt 0c 0 gt a I shyc -CJ wc w gtc(

== c( I shyZ w E W a 0 z- 8

ii sect

~ ~ bullT 0

I if

x ~

bullbullbull

II ~------------________

ANDREW PHASE II

WA VE FORCE PROFILEmiddot

CAP AUTOMATIC APPLICATION FEATURE

STAT

fR1 0 lt f lt1

STAT

STAT

f(R3-R2) middot f lt1

Project ST151K Model Wave_Loading Version 1 rue Nov 8 164107 1994 PT

E~ nss FT

SL 14amp FT

a 111 FT

EL 14 li

E-1 37 FT

El o FT

1middot1middot ______________ _ ________ __ _ __

0

200 400 600 800 1000 12000 Force at Elevatloii (kip)

200

a ll_

Wave Force at Major Elevations for Discrete Wave Heights

180

bullHmiddot40ft

Hmiddot495ft

H bull 51 ft

Hmiddot55ft_

H-59ft I

Andrew (H bull 60 86 ft) II

i I

_11---__

- middot- Hmiddot30ft

-

- bull -

middot bull bull bull middot

---Hmiddot53ft

bullbullbullbullHmiddot57ft

-----H-63ft

Elevation Above Mudllne 100

(ft)

180

140

120

80

60

40

20

Loading at Major Elevations

---H-49511

-deg-Hmiddot608611

0 200 400 600 800 1000 1200

Force It Elevatlon (kip)

--

Base Shear vs Wave Height ST151K

6000

5000 1---------Ult1mateCapacitybull 4890 kip (100 Improved Soll)-----------~--

4000a a lii ~ 3000

81 ampI

2000

1000

0 +-~~~--i~~~~~~~~~-+-~~~--lf--~~~~~~~~-+-~~~--l

30 35 40 45 50 55 60 65

Wave Height (ft)

bull middot

~

bull

II

2

g

~

It)

0

cCD E 0 E a c E euro~ CDshygtItgtmiddoto ~ti c -a Gic CD gt ~

-= c -Cll

I z

bullO 0i i ~ g 0

() CI -bull(q-ll) iuawow llu1wniiaAO

11

shy

116

114

112

g 110

bullc = 108

~ ~ 106 -sect 104

102

100

98

96

-

6555 60

Wave Height vs Center of Action ST151K

30 35 40 45 50

Wave Height (ft)

t

Comparison of Pushover Curves for ST 151K Variable Load Pattern and Monotonic Load Pattern (Andrew)

Soll Improved 100

--- --------- --- ----6000

5000

- 4000 8 - I

1 3000 Ul

81 m 2000

_J -shy

1000

---Variable Load

- - Andrew Load

0 f-~~~~~r--~~~~-t-~~~~~-t-~~~~~+-~~~~----t~~~~~-1

0 1 - 2 3 4 5 6

Deck Displacement (ft)

-middotshy

1 t I I ~

ANDREW PHASE II

BRACE MODELING

bull MARSHALL STRUTS USED FORPHASE I

bull PHASE II WILL ASSESS DIFFERENCE IN USING BUCKLING BEAMS AND MARSHALL STRUTS

middot bull COMPARE FOR 2D FRAME

bull COMPARE FOR PLATFORM 177 WHERE LOWER

BAY EXHIBITS LARGE BRACE MOMENTS

Project andrew Model beamframe Version 1 Mon Nov 14 145447 1994

~

middot Frame Test buckling beam-column model

-

Project andrew Model strutframe Version 1 Mon Nov 14 145754 1994

bull

Frame Test strut model

middot-middot _____________--_

_middotmiddotr_~ --ll--~-middot__

Buckling Beam bull

Strut (K=O 65)

Lateral Displacement

ANDREW PHASE II

FOUNDATIONS

bull ~LICIT NONLINEAR API FOUNDATION USED INiPHASE I

bull FOUNDATIONS

JIP SELECTED MINI-VANE DATA FOR

STRENGTH PROFILE

bull COMMENTS ON BENCHMARK COMPARISONS

- SIGNIFICANT VARIABILITY DUE TO FOUNDATION MODELING METHODS

II Ii

z

ctV 4x SPlO caissonfy-42s soil-api(lf-097step-114)

Inelastic Events Legend Elastic Strut Buckling

Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield

--------shyBeam Clmn Fully Plastic Fracture

~--~~~~~~~~~--~----~__i~~---~~~~__--~

Pro)ect apipile Model South Pelto 10 Version Wed May 18072235 1994 - -

1bullt

+

t

f i bullbullbull

bullbull bullbull

I I

5

~x SPlO Caissonfy=42s soil=api(lf=lOOstep=137)

Inelastic Events Legend

Elastic Strut Bucklilg

Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield

earn Clmn Full~ Plas1c Fracture

ANDREW PHASE II middot middot

JOINT MODELING

middotbull API RIGID-PLASTIC USED IN PHASE I

a ELASTIC-BRITTLE USED IN PHASE IB

amiddot MSL WILL PROVIDE NONLINEAR FORCE-DEFORMATION

AND MOMENT-DEFORMATION RELATIONS FOR PHASE II

a MODELING OPTIONS FOR PHASE II -

l

- UNCO~LEDmiddotONE DIMENSIONAL MEMBER END JOINT MODELING

J Ii

ANDREW PHASE Il

JOINT MODELING OPTIONS DISCUSSION

bull JD COUPLED INTERACTION SURFACE

- IDEAL

- DATA NOT AVAILABLE TO CALIBRATE

bull 2D COUPLED INTERACTION SURFACE

- SIGNIFICANT MODELING IMPROVEMENT

- DATAAVAILABLE

- IMPLEMENTATION PRACTICAL BUT NEEDS DEVELOPMENT AND CALIBRATION

bull 2D UNCOUPLED INTERACTION SURFACE

- SIGNIFICANT MODELING IMPROVEMENT

- DATA AVAILABLE

- IMPLEMENTATION PRACTICAL AND AVAILABLE

ANDREW PHASE II

JOINT MODEJJNG OPTIONS ILLUSTRATIONS

P

e

ANDREW PHASE II

JOINT MODELING PHASE II METHOD

e

I I I

ANDREW PHASE II

ANDREW STORM CONDmONS UPDATED

bull -PHASE I USED OCEANWEATHER HINDCAST-(Nov 1992)

bull middotmiddotPHASE II WILL USE THE REVISED HINDCAST (Nov 1994)

bull lt c lt I lt

2-2

l---4-~~~+-~~-+-~~~1--~~-+-~~--l--~T--+-~~--i24

f-tJVtMfiz~ 1Cfl (r STvb(

Plollol bullbull 5middotNOYmiddot9214bull1J Irbullbull Illbull [ANDREVl~NDREVCUSTbullMAXll 5-NOY-1992 10bull54 - 9208

HEIGHT lml

I~ MAXI

I L 1~middotamp SZ~- ~A I 13c

I J) ( r I X~0iWJ~0gtJ gt L gt gt I 1c I 120

-- N

~ deg

I I I I I I e C I ~=-41 122 -94 -92 -90 -88 -86 -84 -82 -80

-

Figure 7 contours of maximum hindcast significant wave height

_

~

~tl ~

NOltNttIJ 17lt ltNS S iJJ)y

Maximum Significant Wave Height - Hurricane Andrew 1992

30 ~

26 ~ I ~-~~~- == -shy -~ 35-_

24deg1--shy --------~- middot-middot-------middotmiddotmiddot----middot------middot-middot~middot- - - ~-----1---middot

22middot ~~~~~~~-~ -94 -92 -90 -88 -84deg-86 -82 -somiddot

c

-

Platform Selection bullbullbull1111111111111

0 BasisCriteria for Selection middot

0 Review of Phase I Platforms middot 0 Alternative Platforms

0 PMB Recommendations

0 Participants Comments

gt

L

l I

BasisCriteria for Selection bullbullbull1111111111111

0 Benefit for Calibration - Classification (UnexpectedExpected)

- Failure Mode (StructuresFoundations)

- Damaged StructuresKnown response

- Better Representation of Full ~opulation

0 Availability of Data - Existing Models

- Inspection Data

- Damage Data

PREDICTED BEHAVIOR

-

OBSERVED

BEHAVIOR

-SURVIVED DAMAGED FAILED

SURVIVED

ST134W WD90A

-MC311 -

MC397 -

--

-

-

ST151K ST130Q

-

-

-

DAMAGED - -

-

-

--

-

-

ST52 SS139 ST177B -

ST161A

-_

-

FAILED

-

-

-

ST151H ST130A ST72

Operator - Platfonns - Platfonns Survived Damagedamp

- Falled -

bull

-

-

Jle eBDIGIPllDbl -Amocoshy 30 1 Chevron -

143 12 Exxon 72 Mobll 22 1 Phllllps 7 bull

-Shell 39 Unocal 25 1

Total Partlclnts Platfonns 338 15

Other Operators bull- 310 13

-

Total Platfonns - -

648 28

fbml

Platforms Considered 6 7

tie of Total Platfonns 1 25

OVERVIEW OF PLATFORMS SURVIVEDDAMAGEDFAILED

-

shy

v

Platforms Evaluated in Phase I

-

-Platfonn Water Year Number Analysis Avallabillty of Availability Platform

- Name Depth rl of Model DetaDed rl ClassUlcatlon

Installation legs Inspection Soll Report Based on Information Phase I Calibration - -ft

- -- Suaill flalfnrm ~BSi -STISIK 137 1963 8 CAP Yes Yes Un eel Sarvlval

-STIJOO 170 1964 4 CAPmiddotshy Yes ST134data u ST134W 137 1981 4 CAP Yes Yesmiddot E

-

edSarvlval -

middotsurvival

WD90A 184 - 1964 8 CAP Yes E -

- Sarvlvill

MC3ll 343 1978 8 KARMA

MC397 468 1991 4 KARMA -

bull ~reglmmiddot- -

Yes Sure Survival

bull Yes - Sare Sanlval-

-T23STS2 63 1969 4 CAP

T2SSS139 62 1969 4 CAP -

Yes

Yes

Yes Ex

Yes Une

bull SurvivalLikely to DamalI

middot bull SanlvalLlkely to DamalI

ST161A 118 1964 8 USFOS

Eail11a pound1al(bullllm Clss ST177B 142 --1965 8 CAP

STISIH 137 1964 - 8 CAP C

STIJOA 140 19S8 8 CAP - -

Yes

Yes

ollamed

Yes

- Yes E

-

at about 1 mUe

Yes

ST134data

bull SafvlvalLlkelv to Dam- bull- -

E middotrahre

Emeried Failure -

Likely FaUare

T21ST72 61 1969 4 CAP Collamed No Likely FaUure

Platforms Identified for Evaluating Their Likelihood of Falling Under UnexpeCted Survival Category

CllEllATOa

AllB

llLOCll

Sl1lvcnJllE

NAME

YEAR

INSTAUJCD

WATER

IJBPJB

middotmiddotDISTANCE

ftOMSllOllE

(MIUIS)

Ill

(M)

AMOCO BllW

STllI

WDI075

A

bull D

4

n uo 17J

44

32

bull 20

10

bull CllEVllOH STtllO c I 1bull 21 11

STl130 D 0 161 21 11 STllJ4 I 0 137 D - 11

STIUI G I 137 32 11

STIUI I 0 121 32 11

STIUI

J 0 140 32 11

STIUI It 4 137 32 11

STllSI PROl)I 0 137 32 11

STllSI PROD-2 0 137 32 11

STll7 A 0

140 35 11

WDl117 c 65 214 34 10

DXON STshy E 51 14 bull STl165 A t3 bull 34 WD0073

WD0073

WD0073

A _ c

4

62

4

IA IA

17J

22

22

22

bullbullbull WD003-shy E 65 161 Z7

A 0 IN 31 MOBD Bl12I

ssn A bull

50

50

31

7

bull 7

MIJllPllY SS114 B 7 50 7 7

PlllUIPS SSl14t A 55 33 7

SAMEDAN ST17J A 4 117 7 bull 7

SJIELL WD0103 A 65 223 27 WDllOJ bull 65 221 27 WDll04 c 65 221 27 WDOI04 D 67 2 27

UNOCAL

SSl20I

SSl20I

bull SSl20I

SS0209

SS0209

SS0209

SS0209

SS 0253

SS 0253

E

I

PtJllE

A

bull D

G A

c

0

4

4

7J

61

0

0

69

103

97

97

9S

100

95

100

165

175

34

34

34

34

34

34

34

54

54

bullbullbullbullbullbullbullbullbull

Platforms Recommended for Phase II

llalr y_ w- Aot Av- _ I 1 Doplla Model nm bull bull _ bull I I H11 s18-1 Irht lbwl~ bullbullbull(-illtdlel OM lllMtD

II ---

-STISIIC 137 1963 I CAP Yeo Yeo I Yeo

STtshy -170 1964 4 CAP Yeo ST134dlU I Yeo

STl34W 137 Yeo Yeo -1981 4 CAP D Yeo

WD90A 114 1964 I CAP Yeo - D

MC311 30 1971 I KARMA Yeo -sre-

MC397 middot 461 1991 4 ICARMA Yeo Sireshy - - r- -

__ T13ST5l 63 1969 4 CAP Yeo Yeobull middot- ~n- D Yeo _ Tl5SSl39 62 1969 4 CAP Yeo Yeo I Yeo

_ STIT711 142 1965 I CAP Yeo I Yeo middot-1shySTl61A Ill 1964 I USFOS Yeo Yeo - D bulloshy

- r-

STISIH 137 1964 I CAP y D y -_ STl31A 140 1951 I CAP Yeo STl34dlU I Yeo

T21Sll 61 1969 4 CAP No rar-middot I Yeo

~-rmiddot-middot SS209A 95 1967 16 CAP Yeo D -SSU4R

11 UI A 41 -in D SS 2741AIM CAP bull

_ _ -shy - -shy- -shy

CALIBRATION

bull PHASEIAPPROACH

bull CALIBRATION ISSUES

bull PHASE II WORK SCOPE

bull SOFlWARE AVAILABILITY

bull EFFECT OF ADDITIONAL PLATFORMS

bull WEIGHTING APPROACH

bull EXAMPLE APPLICATION middot

(l=Ofce Mode] Wind

~

fcurrent l -middot

~SW ~leldl4

ultimate resistance force I first

design load Illgt

i _

deflection

-

bull

4

Historical Hurrlcane Dariiage and Survival of Platforms

Components Required for Calibration (Petrauskas 1992)

PRIOR Distribution

PLATFORM EXPERIENCE

BAYESIAN Survivals and Failures

UPDATING

Waves Winds and Currents bull

POSTERIOR Distribution

- A_middotJA ~ Survival Data Failure Data

Result Result

Load Resistance

middotfl

-

Bayesian Updaing middotGeneral Description (Petrauskas 1992)

L Imiddot

WHERE TO INTRODUCE BIAS FACTOR (B

On Resistance (Rmue - B (RPREDICreD I

middotOn Loads (Smue middot - B (SPRE01cTEo

On Safety Margin (RSmue - B (RSPREDICTED

CALIBRATION METHODOLOGY

gt

r

FIELD OBSERVATIONS

Observed Behavior

survivals Damages

Failures

r

bull

CAPACITY ANALYSIS

Predicted Load and Resistance Predicted Failure Modes

Survival Damage Failure cases

RELIABILITY ANALYSIS

- Probability of Failure ~

Ukellhood Failure

BAYESIAN UPDATING

Prior Distnbution of B Combined Likelihood Function

Posterior Dlstnbution of B

BIAS FACTOR

Mean Value cov bull

Represents Modeling Uncertainties

Prior Distribution of B Likelihood Functionmiddot Observing Success case STISIK and Posterior Distribution of B

2 00

uo

160

140

120

e 100 L $ uoIshy

0 60 ~ ~ 040

020

0 00

0 00 0 50

_ I 00

b

I 50 2 00

-shy lk [bl STISIK)

--shy Pnor-f(b)

--shy Pos1enor r(bl ISIK)

Prior or B Mean bull 10 COV bull 030

Posterlor of Bbull Meanbull 136COV 0 16

2 50

Likelihood Function and Posterior Distribution Example Platform STISIK

- -

~

- Prior and Posterior Distributions or B Failure cases (ST177B STlSlH ST130A T21) Damage cases (T23 T25 ST161A) Success cases (STlSlK ST130Q ST134W WD90A MC311 MC397)

400

350

300

2 so 0-

I

bull bull

bull bull gt

- bullbull ~ bull bull I bull

_ bull bull bullbull I bull bull bullII V -

bull bull

I bull ~ bull

bullbull bullI bull bullI bull bull bull

middot -~

bull__ ~ middotI middotmiddot

L 2 00

e ~ I SO

-I 00

- 0 so

000 gt

0 00 020 0 40 060 0 80 I 00 120 140 160 110 200

b -

-

shy

Prior-r

- - - - Posterior fLlll failure cases

bull bull bull bull bull bull bull Posterior flall damage cases

- bull bull - bull middot Postenor flall success cases

Posterior flall success fulure damage cases

-

Posterior of B Mean= 119 COV 0101

Posterior Distribution or Blas Factor (8) bull Different Categories and All Cases Combined

shy

l

PHASE I - KEY CONCLUSIONS

bull GLOBAL CORRECTION FACTOR

bull 15 TO 20 CONSERVATISM MEAN)

bull UNEXPECTED SURVIVALS HAVE GREATEST IMPACT

bull DAMAGED CASES MORE INFORMATIVE

bull CALIBRATION ISSUES PHASE II

bull

EFFECT OF IMPROVED CAPACITY ANALYSIS RECIPE

bull APPLICABILITY TO MULTIPLE FAILURE MODES

bull NO FOUNDATION DAMAGE CASE INCLUDED

bull IMPROVED REPRESENTATION OF FULL ANDREW DISTRIBUTION

-bull NUMBER OF PLATFORMS CONSIDERED

I 1 I

I t ) bull ~

PHASE II WORK SCOPE (CALIBRATION)

bull

bull ESTABLISH MULTIPLE BIAS FACTORS

Use Nine Base Case Platforms

middot - Capacity Analysis for Revised recipe

Foundation Blas Factor from Foundation JIP

bull DEVELOP WEIGHTING APPROACH

bull UPDATE MULTIPLE BIAS FACTORS

bullEffect of Platforms Not In the Nine Base Cases

bull EXAMPLE APPLICATION OF BIAS FACTORS I

bullbull

PHASE I PHASE II

CAPACITYbull ANALYSISRu bull

---------------------1----middot

Jc1 - iJbullraquoJ dbull

FttlS frt[f~l ratdhJJ

-

LIKELIHOOD FUNCTION

DEVELOPMENT

lk Jb Jfall1N) bull Jb)

111 Ill Isurvtvoq bull 1 - bl

-middot -------------------shy ~---------------------middot-middot

r111ir8 Cb1 111 Cbl BAYESIAN

UPDATING

~cove

R1R2 bull middot---------------------~----

Jblbull P(g cOJ

1 bull JbRI (MBSI I bull 1

~middot

llcJllJfalluro) bull P(gcOJ Ill 111 b21fallun) bull P ((g1 CO) U 1112 c OD

middot--------------------shy ----------------------shy bull

r111111r8 Cbl 111 Ill i-1nro1

deg ~1b2)laquottfrr (b1b21-lnfo) re Ill bull 1-11middot b2) bull lodegamp

middot

i1 covamp i2 cov~i

-

shy

--

Likelihood Function vs b All Success Case STISI K STl30Q STl34W WD90A MC311 MC397

I 0000

0 9000

OIOOO

0 7000

~ 0amp000

ta osoOO

~ 0 4000

0 3000

0 2000

0 1000

bull 0 0000

bull

I

middot

middot bull I

Ibull

-sibull ~ -middot-middot-middot-middot -middot-

----7----+-----~ I

Ibull bull bull I

rbull

--- Jk ISIK[blsuccess)

- bull - bull - lk 130Q[blsuccess) bull

lk 134W[blsuccess)

Jk-WD90A(blsuccc~I

- - - lk MC397[blsucccss)

--ltgt-- lk middotMC31 l[blsucccss)

--- Jk [bl 6 out of6succcsHascs)

05 I 5 2 250

b

Likelihood Functionsmiddot Success Cases - Phbullsc L

shy

bullbull bull bull bull bull bull -------------------~

bull bull bullbull bullbull bullbull bullbull bullbull bullbull bull

bull bullbull

bullbullbullbull

bullbull

bullbullbullbullbullbullbull

bullbullbullbullbullbull

bullbullbullbullbullbullbull

z 0

t z ) LL

c 0 0c i ~ i

J

~ sect ~--------------~ e l5

I

I

SOFTWARE

bull PF PROGRAM

bull CALREL

bull PF +CALREL

bull RELACS

bull

gt i

l

SENSITIVITY STUDIES

POSTERIOR DISTRIBUTION OF B

bull EFFECT OF CHANGE IN DISTRIBUTION OF PRIOR OF B

bull EFFECT OF REMOVAL OF ONE PLATFORM FROM EACH CATEGORY

bull EFFECT OF REMOVAL OF THREE DAMAGE CASES

bull EFFECT OF ADDITIONAL 13 SIMILAR PLATFORMS

-I

Prior and Posterior Distributions of B Variation In COV of PriorofB fromOl to04 -- -

4500

-4000

-

- -

3500 --middot - bull bull - bull bull Prior-r ICOV =02

3000

- - - - Prior-r ICOV =03 -

bull -~ e 2soo -

bull bull bull bull bull bull bull Pnor-r ICOV =o 4t frJ ~ 2000 Posterior rlCOV =0 2

bull bull Postenor rlCOV =0 3 bull bull 1500 bull middot bull - shy -middot- Posterior rlCOV =04

middot~ I 000 - bull

-middotmiddot ~ ~middotmiddot M -~

- middot 0 500

wbullbullbull _ I middot~ middotmiddotmiddotmiddotmiddot middotmiddotmiddotmiddotmiddot - -~ ~ __ middot_ _ bull -shy0 000

0 000 0200 0 400 0 600 0 800 IOOO 1200 1400 1600 1800 2000 -

- - b

-

=shy

Posterior Distribution of Bias Factor (B) bull Effect of Variation In COV of Prior of B

bull

bullPrior and Posterior Distributions of B Evaluating Effect of Removal of One Platform from Each Category from the Base Case of 13 Platforms

400

~ ~ 350

j IA i

3 00 bull I bull I ---Prior-r I bull l

250 1 I 1

I bull bull ---Posterior r lall 13 casesshy- ~E 2 00 bull bull bull bull bull bull bull Posterior r without STISIK

- ~ 1 bull - bull I - bull bull - bull bull Posterior r Iwithout T2I I 50 I

J-tt~i - bull - bull - Posterior r Iwithout T2S

tOO [-

I I ~ ~ I 0 50

0 00

___ r J i

l~T

~--

0 00 0 20 0 40 0 60 080 I 00 120 140 160 110 2 00

b

zshy

Posterior Distribution of Bias Factor (8) bull Effect of Removal of middotPlatform from Any Category

bullPrior and Posterior Distributions or B Evaluating Effect or Removal or 3 Damage Cases from the Base Case or 13 Platrorms

4

35

3

25

e L 2 ~ bulle shy

I 15

bull I

05

0

0 02 04 06 I 2 I 4 16 IS 2

I I

I

I

I Ibull

J -tmiddot L--

bullI

~

I

fl

II ~

I I v I

middot bull bull

bull L

~ ll- I-~lt

OS I

b

I

bull

1middot bullbullbullbull

~ __

bull

- - - - Prior-r

- - - - Postenor flall failure cases

Postenor flall success cases

--- Posterior flall 13 cases

---Posterior fI success failure cases only

IPosterior or B Mean =t11 COV =0121

-

_I

Posterir Distribution or Bias Factor (B) bull Effect or Removal or 3-Damage Cases

-

bullPrior and Posterior Distributions or B Evaluating Effect or Additional 13 Similar Observations as

Used in Andrew JIP The POsterior or B from Andrew-JIP Considered as New Prior or B

- 500

450

400

350

bull bull

bull bull bull bull J bull

bull bull f ~ I

I ~ Ibull bull

bullbull bull

l bull I -bullbull bull bull

I

I bull bull

bull I I~~bull bull bull

bull ~

I bull bull

~ ~_ ~~~

- - New Prior r bull

Posterior 114 new failure cases ~ 300- bull bull bull bull bull bull bull Posterior f 13 new damage casesl 150 ~

- bull bull - bull bull Posterior 116 new success cases i 2 00

Postenor 1113 new success failure -150 damage cases

I 00

0 50 -

0 00 0 00 0 20 0 40 0 60 0 80 I 00 120 I 40 160 110 2 00

Posterior ol B Mean= 120 COV= 0071 b

shy

shy

_rshy

Posterior Distribution of Bias Factor (B) bull Effect of Additional 13-Similar Observations

1

1 I

WEIGHTING APPROACH bull

bull ASSESSMENT OF PLATFORMS DATA

Select Platforms Most Mappable to the Analyzed Platforms

Unlikely SurvivalDamage Cases Identification

Based on Cut-Off Criteria Agemiddot Hs etc

bull INFORMATION REQUIRED FOR COMPARISON

bull CATEGORIZATION OF ADDITIONALmiddot PLATFORMS

bull UPDATING OF CALIBRATION RESULTS

--

OVERVIEW OF PLATFORMS SURVIVEDDAMAGED -~

Operator -

- Platforms

Survived CS)

Platforms

Damaged CD)

Platforms Used In

Phase IJIP -

JI~ ~Bill~IRBDll -

Amoco -

Chevron -

Exxon

Mob II

Phllllps

Shell

Unocal -

30

143

72 22

7

39 25shy -

1

12

bull

1 bull

1

1(S)+1 (D)

3 (S) +3(D)

2 (S)

bull shy

bull

bull

bull

Total Partlclnants Platforms 338 - 15 10C6S+4D) -

Other Operators

310

13

3 (D)

Total Platforms 648 28 13C6S+7D) -

Summary of Andrew JIP bull Phase I Platforms

Plairshy w v- N-ber Lea-le 1 shy car Hind_ o-w N-bull Deplh fl fl IJpe la bull n Yeadon Jadtet lbelledtf liefalit -- _ -OD

IL H - -shy _ICllshy IL IL - - rQClto ~ _

K STUIK 137 1963 8 365 608S4 Cellardeck

-middot STl-shy 170 1964 4 -~ K 39 62269 Cellardeck - --ST134W 137 1911 4 43 608S4 Smvlnlshy- _ - -

WD90A 184 1964 8 K 267Sbull34 83 SOS31 SUboftllar deck SmvMI

MC311 343 1978 8 x S7 62 - MC397 468 1991 4 x shy S33bull6SO 65911 -shySmeSmvlvol -- gt -

r-- T23ST52 63 1969 4 ton- shyKshy 52594 S03 -T25SSl39 62 1969 4 Kshy SOS94 S063 - middot ton-shy__ -

bullbull a ton--shyST161A 118 1964 8 K - 347S 51495 Cellarclecti

r~

ST1771 142 1965 8 K 39 60167 CeDordeck failme

STISIH 137 1964 8 K 395 608S4 CeDardeck Failme

K ST130A 140 1958 a SS 6095 I lblv Pai1uno - T21ST72 61 1969 4 Kshy S4S94 4974 I lhlv Pai1uno -

- -

JE

INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON

bull WATER DEPTH

bull bull YEAR OF INSTALLATIONDESIGN

bull PLATFORM ORIENTATION

bull PLATFORM LATITUDE AND LONGITUDE

bull KEY SKETCHES OF PLATFORM STRUCTURAL FRAMING

Number of Legs Plies Leg Batter Number of Conductors bull Deck Elevations (Including sub-cellar deck If any) Jacket Framing Elevations Brace type In Vertical Frames Sizes of Primary Members (leg pile vertical frames braces primary horizontals) Joint Details (can gap etc) LegPiie Annulus Grouting Status

bull middotPOST ANDREW INSPECTION REPORT Significant Wave Height Hs (ft)

bull INFORMATION ON DAMAGES PRIOR TO ANDREW

bull RESULTS FROM IN-HOUSE STUDIES IF AVAILABLE

bull ~RIGINAL DESIGN CRITERIA INFORMATION IF AVAILABLE

bull SOIL SHEAR STRENGTH PROFILE IF AVAILABLE

Platforms Identified for Evaluating Therr Likelihood of Falling Under Unexpected Survival Category

OPlRATOR

middotADA

BLOClt

llmucnJllE

NAME

YEAll

INSTAUJD

WA1ER

DBPTll

DISTANCE

JlllOM SHOllE

(MILU)

Bl

(M)

AMOCO 111113

STl11

WDI075

A

bull D

St

17

120

1n

44

20

10 I

CllllVllQfl STlllO c 61 110 21 11 STIUO D Q 1110 21 11

STl134 p Q 137 29 11

ST0151 G 61 137 11

ST0151 I 3 121 11

ST0151 J Q 140 11

STl151 It 137 11

ST0151 PJl001 Q UT 11

ST0151 PROD-2 Q UT 11

ST017 A 3 140 35 11

WDIUT c 65 214 34 10

llXXON STI055 II 51 6T 14 I

STl15 A t3 34

WDI073

WD0073

Amiddotshy Q

1111 1111

22

22

I

I

WDI073 c 1n 22 I

- WDOCl93-shy II

65 1110 2T A 3 194 31 MOBU 111121

SS IOl7

A B

501

50

50

31

T

I

T 6

MtlllPllY SS114 B T 50 T T

PlllllIPS SS0149 A 5 55 33 T

SAMllDAN STln A 107 T T

SBllLL WDl103 A 65 223 2T WDl103 B 65 221 2T WD0104 c 65 221 2T WD0104 D 6T 20 2T

UNOCAL SS020I

SS 020I SSl20I

SS 0209

SS0209

II p

PURll

A

B

3

n 61

103

97

97

95

100

34

34

34

34 34

bullbull bullbullbull

- SS0209 D u 95 34 I

- SS 1209 SSl253

SS0253

G A

c

6T

2

100

165

1T5

34

54

54

bullbull I

I I

I I

SOUTH MARCH

~LAND

N

SOUTH PaTOI ENE ISWltD AREA

MMS Defined Limits ror Offshore Structures Requiring Inspections Following Andrew

t---Jbulln3 T25bull I bullTtl

bull1$1Htl1Abull bull1SIK1779bull

~

middot-DaTA AREA~T

CRANO rsLE

AREA bull1300

bullll4W

=shy

bullPath of Hurricane Andrew with Platforms Used In Calibration

Likelihood Function vs b All Success Case STISI K STIJOQ ST134W WD90A MCJll MCJ97

I 0000

0 9000

0 8000

07000

-Bo6000

tiiii 0 5000 -

04000

0 3000

0 2000

01000

0 0000

0 05 25

ri1rt middot-middot-middot-middot-middot-

I 5 2

b

---11 middot151K(blsu~ccsI

- bull - bull - fk 130Q(bfU~lCSS)

lie bull134W(blsuccc]

lk-WD90A[blsultCS]

- - - lk middotMC397(bfUtlCS]

--o-- lk-MCJll(bbuctcsJ

--- ll (bl 6 out of 6 Mlttcs case]

_

( Likelihood Functions middotSuccess Cases - p - 1 c I

CATEGORIZATION OF ADDITIONAL PLATFORMS

I I I

bull CATEGORIES

Unexpected Survivals

Expected Survivals

bull APPROACH

Identify Corresponding Platform(s) Evaluated In the JIP

Determine Max Wave Height at Platform Location

Evaluate Whether Wave Hit the Deck

middot - Evaluate Likelihood of R and S to Differ from the Base Cases

Review Joint Information

Assess Special Features

bull IDENTIFY NUMBER OF PLATFORMS IN EACH CATEGORY

I

UPDATING OF CALIBRATION RESULTS

bull INCLUDE EFFECT OF ADDITIONAL PLATFORMS ON bullBbull

Consider Increased 1Uncertalntles In R and S

Develop bullNeW Ukellhood Functions for Different Categories

middotLet bullxbull Number of Ukely Unexpected Survival Cases and y Number of Ukely Expected Survival Cases

Determine Combined Likelihood Function lk (b Iadditional platforms) middot

bull REVISED LIKELIHOOD FUNCTION

lk(b I all cases) = lk(b I 9 platforms) middot x lk(b Ix platforms) x lk(b I y platforms)

bull ESTABLISH REVISED BIAS FACTOR B

f8 (b) u f8 (b) lk (b I all cases)

II

t t ~

shy

APPLICATIONS OF BIAS FACTORS

ii APPLICABLE TO PARTICULAR LOAD AND RESISTANCE RECIPE

bull USE BY THE API COMMITTEES AND MMS

Criteria Revisions

Recommended Design Guidelines Revisions bull

bullbull REASSESSMENT STUDIES AND DECISION MAKING

middot

bull RISK ASSESSMENT STUDIES BY THE OPERATORS

Individual Platforms

Fleet of Platforms

middot IMR Program Implementation

bullbull NOT TO BE USED IN DESIGN

I I fl J

EXAMPLE USE OF BIAS FACTORS

bull PHASE I EXAMPLE APPLICATION

bull EVALUATION OF ANNUAL PROBABILITY OF FAILURE

Establish Annual Maximum Seastate Data

Determine Coefficients to Define Base Shear

Determine Ultimate Capacity of Platform

Determine Probablllty of Failure Versus bullbbull

-bull DETERMINE UPDATED PROBABILITY OF FAILURE

Pf - P [S gt R]

Pf I P [~ gt Rb I b] f 8 (b) db

Pr - J P [S gt Rb I b] f bull 8 (b) db

bull TOTAL PROBABILITIES OF FAILURE

Total Pf - 1 bull Il [1 middotPf] - l Pf bull

Total Pf - 1 bull Il [1middot Pf] - l Pfmiddot

Total Pr = 1 - n [1middot PfbullJ = l Pfshy

I I I

EXAMPLE APPLICATION bull PHASE I

PLATFORM ST 151 K

bull GULF OF MEXICO

bull WATER DEPTH = 137 FT

bull ANNUAL MAXIMUM HS MEDIAN 175 ft COV 031

bull ASSUMED STORM DURATION 3 hoursmiddot

bull ASSUMED NUMBER OF WAVES 800

bull MEDIAN HS FOR STORM HOURS 1 2 AND 3

bull MEDIAN CURRENT VALUE bull

bull DETERMINE ULTIMATE CAPACITY FOR THREE DIRECTIONS

bull DETERMINE ANNUAL FAILURE PROBABILITIES

Broadsidemiddot Diagonal End-on

0037 0025 0007

0071 0044 0018

0022 0016 0004

DETERMINE TOTAL ANNUAL PROBABILITY OF FAILURE

Pf - 007

Pf - 013

Pr - 004

bull

EXAMPLE USE OF BIAS FACTORS - PHASE II middot

bull DETERMINE PREDICTIVE PROBABILITY OF FAILURE

Case - 1

Pf _ S S Pf (b b2) fa1a2 (b b2) db db2

g bull -1 (b1 R I S) - 1

92 - (b2 R2 I S) - 1

Case -2

Pf s ~ [bJ f11 (b) db

Pf [bJ p [g lt 0)

g - bull 1 (b1 R1 I S) - 1

1

bull1) l

I

BENEFITS OF ANDREW JIP bull PHASE II

lt

bull IMPROVED LOAD AND RESISTANCE RECIPE

bull IMPROVED CALIBRATION PROCEDURE

bull MULTIPLE BIAS FACTORS

bull PROCEDURE TO INCLUDE EFFECT OF OTHER PLATFORMS

bull PROCEDURE TO USE MULTIPLE BIAS FACTORS

  • Structure Bookmarks
    • ANDREW PHASE II JOINT1 INDUSTRY PROJECT
    • PROGRESS MEETING NUMBER 1
    • ANDREW PHASE II JOINT INDUSTRY PROJECT PROGRESS MEETING NUMBER 1 NOVEMBER 16 AGENDA
    • ANDREW PHASE II UST OF PARTICIPANTS
    • ANDREW PHASE I OBJECTIVES AND RESULTS
    • ANDREW PHASE II PROJECT OBJECTIVES
    • SUBSEQUENT PROJECTS
    • ANDREW PHASE II PROJECT TEAM
    • ANDREW PHASE II UPDATE MODELING middotAND ANALYSIS RECIPE
    • -ANDREW PHASE II CONDUCTOR RESISTANCE
    • ANDREW PHASE Il WA VE FORCE PROFILE
    • ANDREW PHASE II WA VE FORCE PROFILEmiddot
    • ANDREW PHASE II BRACE MODELING
    • ANDREW PHASE II FOUNDATIONS
    • ANDREW PHASE II middot middot JOINT MODELING
    • ANDREW PHASE Il JOINT MODELING OPTIONS DISCUSSION
    • ANDREW PHASE II JOINT MODEJJNG OPTIONS ILLUSTRATIONS
    • ANDREW PHASE II JOINT MODELING PHASE II METHOD
    • ANDREW PHASE II ANDREW STORM CONDmONS UPDATED
    • Platform Selection bullbullbull1111111111111
    • BasisCriteria for Selection bullbullbull1111111111111
    • CALIBRATION
    • WHERE TO INTRODUCE BIAS FACTOR (B
    • PHASE I -KEY CONCLUSIONS
    • bull CALIBRATION ISSUES PHASE II
    • PHASE II WORK SCOPE (CALIBRATION)
    • SOFTWARE
    • SENSITIVITY STUDIES POSTERIOR DISTRIBUTION OF B
    • WEIGHTING APPROACH bull
    • INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON
    • CATEGORIZATION OF ADDITIONAL PLATFORMS
    • UPDATING OF CALIBRATION RESULTS
    • APPLICATIONS OF BIAS FACTORS
    • EXAMPLE USE OF BIAS FACTORS
    • EXAMPLE APPLICATION bull PHASE I PLATFORM ST 151 K
    • EXAMPLE USE OF BIAS FACTORS -PHASE II middot
    • BENEFITS OF ANDREW JIP bull PHASE II
Page 9:  · . ' ANDREW PHASE II PROJECT OBJECTIVES . ' • ·UPDATE MODELING AND ANALYSIS RECIPE· JOINTS ' ' ·BRACES WAVE PROFILE FOUNDATIONS . ' ' CONDUCTORS· • ANALVZE PLATFORMS USING

~ I I

-ANDREW PHASE II

CONDUCTOR RESISTANCE

shybull RESISTANCE NOT INCLUDED IN PHASE I

bull CONTRIBUTION TO OVERALL RESISTANCE A PHASE II

STUDY ISSUE

middotbull middotCOMPARE FOR 4-LEG AND 8-LEG PLATFORM

bull CONSIDER CONDUCTOR POSmON IN GUIDES

- 13 LEFT 13 CENTER 13 RIGHT

Project ST151K Model Conductor version 1 Mon Nov 14 143140 1994

- ---- ---

--

117 s

111

7i

31

0 -5 -~ - ~()

- iltgt --lo

-(pO

-so

bull bullf OC)

-120

-ISo

-ISO

z

CAgt ~x Singl_e Conductor Model

bull

=I=

-_

-_

-gtshy

~ -

~

I

I~

- gt

~-----------____________________

Project STlSlK Model dir_2 version 7 Mon Nov 14 124200 1994

177 s

Ill

74

37

0- - e -zo - ~o

- Jo

-100

-10

-ISO

z

UV 1-+x

Project STlSlX Model Conductor Version 1 Mon Nov 14 140620 1994

bullmiddot

I

I

I I

I

I I

I I I I I

1 ==

-L

-L - L

-

- L

-

-

L

-

Uniform Displacement step 6 Inelastic Events Legend

Elastic

Strut Buckling Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield Beam Clmn FUlly Plastic Fracture

--

i~~~~~~~~~~~---~--~~-J___--JJ~_middot~~~=-

Project ST151K Model Conductor Version1 Mon Nov 14 142735 1994

---

~

I

I

middot

I

I

I

- shy- shy

- shy

- shy

4x z

CIV Uniform Displacement step 33 Inelastic Events Legend

bull Elasti~ Strut Buckling Strut Residual Strut Reloading Plastic StrutNLTruss Beam Clmn Initial Yield Beam Clmn-Fully Plastic Fracture

gt bull

Project STlSlK middot Model Conductor Version 2 Mon Nov 14 142025 1994

-----------------

bullf

I

I

I

II I I

I I I I

I ==

-

Variable Displacement step 9 Inelastic Events Legend

Elastic Strut Residual

bull

Strut Buckling Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield Beam Clmn Fully Plastic Fracture

--

Project ST151K Model Conductor Version 2 Mon Nov 14 142145 1994

I I I I

I I

I

I

I - - -

-Ishy-

-

Variable Displacement step 35 InelasticEvents Legend

Elastic Strut Buckling Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield Beam Clmn Fully Plastic Fracture

Project ST151K Model Conductormiddot

Version 1

Clgt - Cut Plane Force Fx Mon Nov 14 140051 1994

- -_ vii I fOIZM

0 _

~

N ID 0 bull 0

x 0

middotrt ~ () bull u 0 ~ 0 Ir

N bull

0

0 1 2 3 4 5 X Deck Disp [Ft)

I I 1 1 WbullTI- bullbull

lltcxi lr-C

011lt~~~~~~~---~~~~~~~~L~~~~~~~--~~~~~~~~i~~~~~~~--

Uniform Node Displacement

shy

Ji

r

Comparison of Pushover Curves for ST 151K Variable Load Pattern and Monotonic Load Pattern (Andrew)

Soll Improved 100

6000

5000 -i 4000

f

llS

3000 Ul

J 2000

10 c11 c1lgtolts 1000

0 v

bull ~

f

_ ----- --- ---shy

bull --Variable Lold

- - Andrew Load

I I I I I I

4 5 62 30 1

Deck Dlsplacement (ft)

bullbull r 1 II

~

ANDREW PHASE Il

WA VE FORCE PROFILE

bull DECK WA VE FORCE ~

PHASE I USED SECTION 17 (1056H degCd)

PHASE Il USE REVISED SECTION 17 (1000H Cd)

bull PHASE I USED SINGLE FORCE PROFILE FOR PUSHOVER

- USED WAVE HEIGHT CLOSE TO ULTIMATE CAPACITY

bull PHASE J

Il WILL STUDY THE

USE OF VARIABLE

PROFILE ~

- IF FORCE LEVEL INCREASES (OR DECREASES) BY A FACTOR EQUIVALENT TO A 2 FOOT CHANGE IN WA VE HEIGHT THE INCREMENTAL FO~CE PA1IERN IS CHANGED ACCORDINGLY (CAP FEATURE)

- INITIAL EXAMPLE

f~

a

middotn

a

I N -a a a amiddot

+

~ a

l

a w gt 0c 0 gt a I shyc -CJ wc w gtc(

== c( I shyZ w E W a 0 z- 8

ii sect

~ ~ bullT 0

I if

x ~

bullbullbull

II ~------------________

ANDREW PHASE II

WA VE FORCE PROFILEmiddot

CAP AUTOMATIC APPLICATION FEATURE

STAT

fR1 0 lt f lt1

STAT

STAT

f(R3-R2) middot f lt1

Project ST151K Model Wave_Loading Version 1 rue Nov 8 164107 1994 PT

E~ nss FT

SL 14amp FT

a 111 FT

EL 14 li

E-1 37 FT

El o FT

1middot1middot ______________ _ ________ __ _ __

0

200 400 600 800 1000 12000 Force at Elevatloii (kip)

200

a ll_

Wave Force at Major Elevations for Discrete Wave Heights

180

bullHmiddot40ft

Hmiddot495ft

H bull 51 ft

Hmiddot55ft_

H-59ft I

Andrew (H bull 60 86 ft) II

i I

_11---__

- middot- Hmiddot30ft

-

- bull -

middot bull bull bull middot

---Hmiddot53ft

bullbullbullbullHmiddot57ft

-----H-63ft

Elevation Above Mudllne 100

(ft)

180

140

120

80

60

40

20

Loading at Major Elevations

---H-49511

-deg-Hmiddot608611

0 200 400 600 800 1000 1200

Force It Elevatlon (kip)

--

Base Shear vs Wave Height ST151K

6000

5000 1---------Ult1mateCapacitybull 4890 kip (100 Improved Soll)-----------~--

4000a a lii ~ 3000

81 ampI

2000

1000

0 +-~~~--i~~~~~~~~~-+-~~~--lf--~~~~~~~~-+-~~~--l

30 35 40 45 50 55 60 65

Wave Height (ft)

bull middot

~

bull

II

2

g

~

It)

0

cCD E 0 E a c E euro~ CDshygtItgtmiddoto ~ti c -a Gic CD gt ~

-= c -Cll

I z

bullO 0i i ~ g 0

() CI -bull(q-ll) iuawow llu1wniiaAO

11

shy

116

114

112

g 110

bullc = 108

~ ~ 106 -sect 104

102

100

98

96

-

6555 60

Wave Height vs Center of Action ST151K

30 35 40 45 50

Wave Height (ft)

t

Comparison of Pushover Curves for ST 151K Variable Load Pattern and Monotonic Load Pattern (Andrew)

Soll Improved 100

--- --------- --- ----6000

5000

- 4000 8 - I

1 3000 Ul

81 m 2000

_J -shy

1000

---Variable Load

- - Andrew Load

0 f-~~~~~r--~~~~-t-~~~~~-t-~~~~~+-~~~~----t~~~~~-1

0 1 - 2 3 4 5 6

Deck Displacement (ft)

-middotshy

1 t I I ~

ANDREW PHASE II

BRACE MODELING

bull MARSHALL STRUTS USED FORPHASE I

bull PHASE II WILL ASSESS DIFFERENCE IN USING BUCKLING BEAMS AND MARSHALL STRUTS

middot bull COMPARE FOR 2D FRAME

bull COMPARE FOR PLATFORM 177 WHERE LOWER

BAY EXHIBITS LARGE BRACE MOMENTS

Project andrew Model beamframe Version 1 Mon Nov 14 145447 1994

~

middot Frame Test buckling beam-column model

-

Project andrew Model strutframe Version 1 Mon Nov 14 145754 1994

bull

Frame Test strut model

middot-middot _____________--_

_middotmiddotr_~ --ll--~-middot__

Buckling Beam bull

Strut (K=O 65)

Lateral Displacement

ANDREW PHASE II

FOUNDATIONS

bull ~LICIT NONLINEAR API FOUNDATION USED INiPHASE I

bull FOUNDATIONS

JIP SELECTED MINI-VANE DATA FOR

STRENGTH PROFILE

bull COMMENTS ON BENCHMARK COMPARISONS

- SIGNIFICANT VARIABILITY DUE TO FOUNDATION MODELING METHODS

II Ii

z

ctV 4x SPlO caissonfy-42s soil-api(lf-097step-114)

Inelastic Events Legend Elastic Strut Buckling

Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield

--------shyBeam Clmn Fully Plastic Fracture

~--~~~~~~~~~--~----~__i~~---~~~~__--~

Pro)ect apipile Model South Pelto 10 Version Wed May 18072235 1994 - -

1bullt

+

t

f i bullbullbull

bullbull bullbull

I I

5

~x SPlO Caissonfy=42s soil=api(lf=lOOstep=137)

Inelastic Events Legend

Elastic Strut Bucklilg

Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield

earn Clmn Full~ Plas1c Fracture

ANDREW PHASE II middot middot

JOINT MODELING

middotbull API RIGID-PLASTIC USED IN PHASE I

a ELASTIC-BRITTLE USED IN PHASE IB

amiddot MSL WILL PROVIDE NONLINEAR FORCE-DEFORMATION

AND MOMENT-DEFORMATION RELATIONS FOR PHASE II

a MODELING OPTIONS FOR PHASE II -

l

- UNCO~LEDmiddotONE DIMENSIONAL MEMBER END JOINT MODELING

J Ii

ANDREW PHASE Il

JOINT MODELING OPTIONS DISCUSSION

bull JD COUPLED INTERACTION SURFACE

- IDEAL

- DATA NOT AVAILABLE TO CALIBRATE

bull 2D COUPLED INTERACTION SURFACE

- SIGNIFICANT MODELING IMPROVEMENT

- DATAAVAILABLE

- IMPLEMENTATION PRACTICAL BUT NEEDS DEVELOPMENT AND CALIBRATION

bull 2D UNCOUPLED INTERACTION SURFACE

- SIGNIFICANT MODELING IMPROVEMENT

- DATA AVAILABLE

- IMPLEMENTATION PRACTICAL AND AVAILABLE

ANDREW PHASE II

JOINT MODEJJNG OPTIONS ILLUSTRATIONS

P

e

ANDREW PHASE II

JOINT MODELING PHASE II METHOD

e

I I I

ANDREW PHASE II

ANDREW STORM CONDmONS UPDATED

bull -PHASE I USED OCEANWEATHER HINDCAST-(Nov 1992)

bull middotmiddotPHASE II WILL USE THE REVISED HINDCAST (Nov 1994)

bull lt c lt I lt

2-2

l---4-~~~+-~~-+-~~~1--~~-+-~~--l--~T--+-~~--i24

f-tJVtMfiz~ 1Cfl (r STvb(

Plollol bullbull 5middotNOYmiddot9214bull1J Irbullbull Illbull [ANDREVl~NDREVCUSTbullMAXll 5-NOY-1992 10bull54 - 9208

HEIGHT lml

I~ MAXI

I L 1~middotamp SZ~- ~A I 13c

I J) ( r I X~0iWJ~0gtJ gt L gt gt I 1c I 120

-- N

~ deg

I I I I I I e C I ~=-41 122 -94 -92 -90 -88 -86 -84 -82 -80

-

Figure 7 contours of maximum hindcast significant wave height

_

~

~tl ~

NOltNttIJ 17lt ltNS S iJJ)y

Maximum Significant Wave Height - Hurricane Andrew 1992

30 ~

26 ~ I ~-~~~- == -shy -~ 35-_

24deg1--shy --------~- middot-middot-------middotmiddotmiddot----middot------middot-middot~middot- - - ~-----1---middot

22middot ~~~~~~~-~ -94 -92 -90 -88 -84deg-86 -82 -somiddot

c

-

Platform Selection bullbullbull1111111111111

0 BasisCriteria for Selection middot

0 Review of Phase I Platforms middot 0 Alternative Platforms

0 PMB Recommendations

0 Participants Comments

gt

L

l I

BasisCriteria for Selection bullbullbull1111111111111

0 Benefit for Calibration - Classification (UnexpectedExpected)

- Failure Mode (StructuresFoundations)

- Damaged StructuresKnown response

- Better Representation of Full ~opulation

0 Availability of Data - Existing Models

- Inspection Data

- Damage Data

PREDICTED BEHAVIOR

-

OBSERVED

BEHAVIOR

-SURVIVED DAMAGED FAILED

SURVIVED

ST134W WD90A

-MC311 -

MC397 -

--

-

-

ST151K ST130Q

-

-

-

DAMAGED - -

-

-

--

-

-

ST52 SS139 ST177B -

ST161A

-_

-

FAILED

-

-

-

ST151H ST130A ST72

Operator - Platfonns - Platfonns Survived Damagedamp

- Falled -

bull

-

-

Jle eBDIGIPllDbl -Amocoshy 30 1 Chevron -

143 12 Exxon 72 Mobll 22 1 Phllllps 7 bull

-Shell 39 Unocal 25 1

Total Partlclnts Platfonns 338 15

Other Operators bull- 310 13

-

Total Platfonns - -

648 28

fbml

Platforms Considered 6 7

tie of Total Platfonns 1 25

OVERVIEW OF PLATFORMS SURVIVEDDAMAGEDFAILED

-

shy

v

Platforms Evaluated in Phase I

-

-Platfonn Water Year Number Analysis Avallabillty of Availability Platform

- Name Depth rl of Model DetaDed rl ClassUlcatlon

Installation legs Inspection Soll Report Based on Information Phase I Calibration - -ft

- -- Suaill flalfnrm ~BSi -STISIK 137 1963 8 CAP Yes Yes Un eel Sarvlval

-STIJOO 170 1964 4 CAPmiddotshy Yes ST134data u ST134W 137 1981 4 CAP Yes Yesmiddot E

-

edSarvlval -

middotsurvival

WD90A 184 - 1964 8 CAP Yes E -

- Sarvlvill

MC3ll 343 1978 8 KARMA

MC397 468 1991 4 KARMA -

bull ~reglmmiddot- -

Yes Sure Survival

bull Yes - Sare Sanlval-

-T23STS2 63 1969 4 CAP

T2SSS139 62 1969 4 CAP -

Yes

Yes

Yes Ex

Yes Une

bull SurvivalLikely to DamalI

middot bull SanlvalLlkely to DamalI

ST161A 118 1964 8 USFOS

Eail11a pound1al(bullllm Clss ST177B 142 --1965 8 CAP

STISIH 137 1964 - 8 CAP C

STIJOA 140 19S8 8 CAP - -

Yes

Yes

ollamed

Yes

- Yes E

-

at about 1 mUe

Yes

ST134data

bull SafvlvalLlkelv to Dam- bull- -

E middotrahre

Emeried Failure -

Likely FaUare

T21ST72 61 1969 4 CAP Collamed No Likely FaUure

Platforms Identified for Evaluating Their Likelihood of Falling Under UnexpeCted Survival Category

CllEllATOa

AllB

llLOCll

Sl1lvcnJllE

NAME

YEAR

INSTAUJCD

WATER

IJBPJB

middotmiddotDISTANCE

ftOMSllOllE

(MIUIS)

Ill

(M)

AMOCO BllW

STllI

WDI075

A

bull D

4

n uo 17J

44

32

bull 20

10

bull CllEVllOH STtllO c I 1bull 21 11

STl130 D 0 161 21 11 STllJ4 I 0 137 D - 11

STIUI G I 137 32 11

STIUI I 0 121 32 11

STIUI

J 0 140 32 11

STIUI It 4 137 32 11

STllSI PROl)I 0 137 32 11

STllSI PROD-2 0 137 32 11

STll7 A 0

140 35 11

WDl117 c 65 214 34 10

DXON STshy E 51 14 bull STl165 A t3 bull 34 WD0073

WD0073

WD0073

A _ c

4

62

4

IA IA

17J

22

22

22

bullbullbull WD003-shy E 65 161 Z7

A 0 IN 31 MOBD Bl12I

ssn A bull

50

50

31

7

bull 7

MIJllPllY SS114 B 7 50 7 7

PlllUIPS SSl14t A 55 33 7

SAMEDAN ST17J A 4 117 7 bull 7

SJIELL WD0103 A 65 223 27 WDllOJ bull 65 221 27 WDll04 c 65 221 27 WDOI04 D 67 2 27

UNOCAL

SSl20I

SSl20I

bull SSl20I

SS0209

SS0209

SS0209

SS0209

SS 0253

SS 0253

E

I

PtJllE

A

bull D

G A

c

0

4

4

7J

61

0

0

69

103

97

97

9S

100

95

100

165

175

34

34

34

34

34

34

34

54

54

bullbullbullbullbullbullbullbullbull

Platforms Recommended for Phase II

llalr y_ w- Aot Av- _ I 1 Doplla Model nm bull bull _ bull I I H11 s18-1 Irht lbwl~ bullbullbull(-illtdlel OM lllMtD

II ---

-STISIIC 137 1963 I CAP Yeo Yeo I Yeo

STtshy -170 1964 4 CAP Yeo ST134dlU I Yeo

STl34W 137 Yeo Yeo -1981 4 CAP D Yeo

WD90A 114 1964 I CAP Yeo - D

MC311 30 1971 I KARMA Yeo -sre-

MC397 middot 461 1991 4 ICARMA Yeo Sireshy - - r- -

__ T13ST5l 63 1969 4 CAP Yeo Yeobull middot- ~n- D Yeo _ Tl5SSl39 62 1969 4 CAP Yeo Yeo I Yeo

_ STIT711 142 1965 I CAP Yeo I Yeo middot-1shySTl61A Ill 1964 I USFOS Yeo Yeo - D bulloshy

- r-

STISIH 137 1964 I CAP y D y -_ STl31A 140 1951 I CAP Yeo STl34dlU I Yeo

T21Sll 61 1969 4 CAP No rar-middot I Yeo

~-rmiddot-middot SS209A 95 1967 16 CAP Yeo D -SSU4R

11 UI A 41 -in D SS 2741AIM CAP bull

_ _ -shy - -shy- -shy

CALIBRATION

bull PHASEIAPPROACH

bull CALIBRATION ISSUES

bull PHASE II WORK SCOPE

bull SOFlWARE AVAILABILITY

bull EFFECT OF ADDITIONAL PLATFORMS

bull WEIGHTING APPROACH

bull EXAMPLE APPLICATION middot

(l=Ofce Mode] Wind

~

fcurrent l -middot

~SW ~leldl4

ultimate resistance force I first

design load Illgt

i _

deflection

-

bull

4

Historical Hurrlcane Dariiage and Survival of Platforms

Components Required for Calibration (Petrauskas 1992)

PRIOR Distribution

PLATFORM EXPERIENCE

BAYESIAN Survivals and Failures

UPDATING

Waves Winds and Currents bull

POSTERIOR Distribution

- A_middotJA ~ Survival Data Failure Data

Result Result

Load Resistance

middotfl

-

Bayesian Updaing middotGeneral Description (Petrauskas 1992)

L Imiddot

WHERE TO INTRODUCE BIAS FACTOR (B

On Resistance (Rmue - B (RPREDICreD I

middotOn Loads (Smue middot - B (SPRE01cTEo

On Safety Margin (RSmue - B (RSPREDICTED

CALIBRATION METHODOLOGY

gt

r

FIELD OBSERVATIONS

Observed Behavior

survivals Damages

Failures

r

bull

CAPACITY ANALYSIS

Predicted Load and Resistance Predicted Failure Modes

Survival Damage Failure cases

RELIABILITY ANALYSIS

- Probability of Failure ~

Ukellhood Failure

BAYESIAN UPDATING

Prior Distnbution of B Combined Likelihood Function

Posterior Dlstnbution of B

BIAS FACTOR

Mean Value cov bull

Represents Modeling Uncertainties

Prior Distribution of B Likelihood Functionmiddot Observing Success case STISIK and Posterior Distribution of B

2 00

uo

160

140

120

e 100 L $ uoIshy

0 60 ~ ~ 040

020

0 00

0 00 0 50

_ I 00

b

I 50 2 00

-shy lk [bl STISIK)

--shy Pnor-f(b)

--shy Pos1enor r(bl ISIK)

Prior or B Mean bull 10 COV bull 030

Posterlor of Bbull Meanbull 136COV 0 16

2 50

Likelihood Function and Posterior Distribution Example Platform STISIK

- -

~

- Prior and Posterior Distributions or B Failure cases (ST177B STlSlH ST130A T21) Damage cases (T23 T25 ST161A) Success cases (STlSlK ST130Q ST134W WD90A MC311 MC397)

400

350

300

2 so 0-

I

bull bull

bull bull gt

- bullbull ~ bull bull I bull

_ bull bull bullbull I bull bull bullII V -

bull bull

I bull ~ bull

bullbull bullI bull bullI bull bull bull

middot -~

bull__ ~ middotI middotmiddot

L 2 00

e ~ I SO

-I 00

- 0 so

000 gt

0 00 020 0 40 060 0 80 I 00 120 140 160 110 200

b -

-

shy

Prior-r

- - - - Posterior fLlll failure cases

bull bull bull bull bull bull bull Posterior flall damage cases

- bull bull - bull middot Postenor flall success cases

Posterior flall success fulure damage cases

-

Posterior of B Mean= 119 COV 0101

Posterior Distribution or Blas Factor (8) bull Different Categories and All Cases Combined

shy

l

PHASE I - KEY CONCLUSIONS

bull GLOBAL CORRECTION FACTOR

bull 15 TO 20 CONSERVATISM MEAN)

bull UNEXPECTED SURVIVALS HAVE GREATEST IMPACT

bull DAMAGED CASES MORE INFORMATIVE

bull CALIBRATION ISSUES PHASE II

bull

EFFECT OF IMPROVED CAPACITY ANALYSIS RECIPE

bull APPLICABILITY TO MULTIPLE FAILURE MODES

bull NO FOUNDATION DAMAGE CASE INCLUDED

bull IMPROVED REPRESENTATION OF FULL ANDREW DISTRIBUTION

-bull NUMBER OF PLATFORMS CONSIDERED

I 1 I

I t ) bull ~

PHASE II WORK SCOPE (CALIBRATION)

bull

bull ESTABLISH MULTIPLE BIAS FACTORS

Use Nine Base Case Platforms

middot - Capacity Analysis for Revised recipe

Foundation Blas Factor from Foundation JIP

bull DEVELOP WEIGHTING APPROACH

bull UPDATE MULTIPLE BIAS FACTORS

bullEffect of Platforms Not In the Nine Base Cases

bull EXAMPLE APPLICATION OF BIAS FACTORS I

bullbull

PHASE I PHASE II

CAPACITYbull ANALYSISRu bull

---------------------1----middot

Jc1 - iJbullraquoJ dbull

FttlS frt[f~l ratdhJJ

-

LIKELIHOOD FUNCTION

DEVELOPMENT

lk Jb Jfall1N) bull Jb)

111 Ill Isurvtvoq bull 1 - bl

-middot -------------------shy ~---------------------middot-middot

r111ir8 Cb1 111 Cbl BAYESIAN

UPDATING

~cove

R1R2 bull middot---------------------~----

Jblbull P(g cOJ

1 bull JbRI (MBSI I bull 1

~middot

llcJllJfalluro) bull P(gcOJ Ill 111 b21fallun) bull P ((g1 CO) U 1112 c OD

middot--------------------shy ----------------------shy bull

r111111r8 Cbl 111 Ill i-1nro1

deg ~1b2)laquottfrr (b1b21-lnfo) re Ill bull 1-11middot b2) bull lodegamp

middot

i1 covamp i2 cov~i

-

shy

--

Likelihood Function vs b All Success Case STISI K STl30Q STl34W WD90A MC311 MC397

I 0000

0 9000

OIOOO

0 7000

~ 0amp000

ta osoOO

~ 0 4000

0 3000

0 2000

0 1000

bull 0 0000

bull

I

middot

middot bull I

Ibull

-sibull ~ -middot-middot-middot-middot -middot-

----7----+-----~ I

Ibull bull bull I

rbull

--- Jk ISIK[blsuccess)

- bull - bull - lk 130Q[blsuccess) bull

lk 134W[blsuccess)

Jk-WD90A(blsuccc~I

- - - lk MC397[blsucccss)

--ltgt-- lk middotMC31 l[blsucccss)

--- Jk [bl 6 out of6succcsHascs)

05 I 5 2 250

b

Likelihood Functionsmiddot Success Cases - Phbullsc L

shy

bullbull bull bull bull bull bull -------------------~

bull bull bullbull bullbull bullbull bullbull bullbull bullbull bull

bull bullbull

bullbullbullbull

bullbull

bullbullbullbullbullbullbull

bullbullbullbullbullbull

bullbullbullbullbullbullbull

z 0

t z ) LL

c 0 0c i ~ i

J

~ sect ~--------------~ e l5

I

I

SOFTWARE

bull PF PROGRAM

bull CALREL

bull PF +CALREL

bull RELACS

bull

gt i

l

SENSITIVITY STUDIES

POSTERIOR DISTRIBUTION OF B

bull EFFECT OF CHANGE IN DISTRIBUTION OF PRIOR OF B

bull EFFECT OF REMOVAL OF ONE PLATFORM FROM EACH CATEGORY

bull EFFECT OF REMOVAL OF THREE DAMAGE CASES

bull EFFECT OF ADDITIONAL 13 SIMILAR PLATFORMS

-I

Prior and Posterior Distributions of B Variation In COV of PriorofB fromOl to04 -- -

4500

-4000

-

- -

3500 --middot - bull bull - bull bull Prior-r ICOV =02

3000

- - - - Prior-r ICOV =03 -

bull -~ e 2soo -

bull bull bull bull bull bull bull Pnor-r ICOV =o 4t frJ ~ 2000 Posterior rlCOV =0 2

bull bull Postenor rlCOV =0 3 bull bull 1500 bull middot bull - shy -middot- Posterior rlCOV =04

middot~ I 000 - bull

-middotmiddot ~ ~middotmiddot M -~

- middot 0 500

wbullbullbull _ I middot~ middotmiddotmiddotmiddotmiddot middotmiddotmiddotmiddotmiddot - -~ ~ __ middot_ _ bull -shy0 000

0 000 0200 0 400 0 600 0 800 IOOO 1200 1400 1600 1800 2000 -

- - b

-

=shy

Posterior Distribution of Bias Factor (B) bull Effect of Variation In COV of Prior of B

bull

bullPrior and Posterior Distributions of B Evaluating Effect of Removal of One Platform from Each Category from the Base Case of 13 Platforms

400

~ ~ 350

j IA i

3 00 bull I bull I ---Prior-r I bull l

250 1 I 1

I bull bull ---Posterior r lall 13 casesshy- ~E 2 00 bull bull bull bull bull bull bull Posterior r without STISIK

- ~ 1 bull - bull I - bull bull - bull bull Posterior r Iwithout T2I I 50 I

J-tt~i - bull - bull - Posterior r Iwithout T2S

tOO [-

I I ~ ~ I 0 50

0 00

___ r J i

l~T

~--

0 00 0 20 0 40 0 60 080 I 00 120 140 160 110 2 00

b

zshy

Posterior Distribution of Bias Factor (8) bull Effect of Removal of middotPlatform from Any Category

bullPrior and Posterior Distributions or B Evaluating Effect or Removal or 3 Damage Cases from the Base Case or 13 Platrorms

4

35

3

25

e L 2 ~ bulle shy

I 15

bull I

05

0

0 02 04 06 I 2 I 4 16 IS 2

I I

I

I

I Ibull

J -tmiddot L--

bullI

~

I

fl

II ~

I I v I

middot bull bull

bull L

~ ll- I-~lt

OS I

b

I

bull

1middot bullbullbullbull

~ __

bull

- - - - Prior-r

- - - - Postenor flall failure cases

Postenor flall success cases

--- Posterior flall 13 cases

---Posterior fI success failure cases only

IPosterior or B Mean =t11 COV =0121

-

_I

Posterir Distribution or Bias Factor (B) bull Effect or Removal or 3-Damage Cases

-

bullPrior and Posterior Distributions or B Evaluating Effect or Additional 13 Similar Observations as

Used in Andrew JIP The POsterior or B from Andrew-JIP Considered as New Prior or B

- 500

450

400

350

bull bull

bull bull bull bull J bull

bull bull f ~ I

I ~ Ibull bull

bullbull bull

l bull I -bullbull bull bull

I

I bull bull

bull I I~~bull bull bull

bull ~

I bull bull

~ ~_ ~~~

- - New Prior r bull

Posterior 114 new failure cases ~ 300- bull bull bull bull bull bull bull Posterior f 13 new damage casesl 150 ~

- bull bull - bull bull Posterior 116 new success cases i 2 00

Postenor 1113 new success failure -150 damage cases

I 00

0 50 -

0 00 0 00 0 20 0 40 0 60 0 80 I 00 120 I 40 160 110 2 00

Posterior ol B Mean= 120 COV= 0071 b

shy

shy

_rshy

Posterior Distribution of Bias Factor (B) bull Effect of Additional 13-Similar Observations

1

1 I

WEIGHTING APPROACH bull

bull ASSESSMENT OF PLATFORMS DATA

Select Platforms Most Mappable to the Analyzed Platforms

Unlikely SurvivalDamage Cases Identification

Based on Cut-Off Criteria Agemiddot Hs etc

bull INFORMATION REQUIRED FOR COMPARISON

bull CATEGORIZATION OF ADDITIONALmiddot PLATFORMS

bull UPDATING OF CALIBRATION RESULTS

--

OVERVIEW OF PLATFORMS SURVIVEDDAMAGED -~

Operator -

- Platforms

Survived CS)

Platforms

Damaged CD)

Platforms Used In

Phase IJIP -

JI~ ~Bill~IRBDll -

Amoco -

Chevron -

Exxon

Mob II

Phllllps

Shell

Unocal -

30

143

72 22

7

39 25shy -

1

12

bull

1 bull

1

1(S)+1 (D)

3 (S) +3(D)

2 (S)

bull shy

bull

bull

bull

Total Partlclnants Platforms 338 - 15 10C6S+4D) -

Other Operators

310

13

3 (D)

Total Platforms 648 28 13C6S+7D) -

Summary of Andrew JIP bull Phase I Platforms

Plairshy w v- N-ber Lea-le 1 shy car Hind_ o-w N-bull Deplh fl fl IJpe la bull n Yeadon Jadtet lbelledtf liefalit -- _ -OD

IL H - -shy _ICllshy IL IL - - rQClto ~ _

K STUIK 137 1963 8 365 608S4 Cellardeck

-middot STl-shy 170 1964 4 -~ K 39 62269 Cellardeck - --ST134W 137 1911 4 43 608S4 Smvlnlshy- _ - -

WD90A 184 1964 8 K 267Sbull34 83 SOS31 SUboftllar deck SmvMI

MC311 343 1978 8 x S7 62 - MC397 468 1991 4 x shy S33bull6SO 65911 -shySmeSmvlvol -- gt -

r-- T23ST52 63 1969 4 ton- shyKshy 52594 S03 -T25SSl39 62 1969 4 Kshy SOS94 S063 - middot ton-shy__ -

bullbull a ton--shyST161A 118 1964 8 K - 347S 51495 Cellarclecti

r~

ST1771 142 1965 8 K 39 60167 CeDordeck failme

STISIH 137 1964 8 K 395 608S4 CeDardeck Failme

K ST130A 140 1958 a SS 6095 I lblv Pai1uno - T21ST72 61 1969 4 Kshy S4S94 4974 I lhlv Pai1uno -

- -

JE

INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON

bull WATER DEPTH

bull bull YEAR OF INSTALLATIONDESIGN

bull PLATFORM ORIENTATION

bull PLATFORM LATITUDE AND LONGITUDE

bull KEY SKETCHES OF PLATFORM STRUCTURAL FRAMING

Number of Legs Plies Leg Batter Number of Conductors bull Deck Elevations (Including sub-cellar deck If any) Jacket Framing Elevations Brace type In Vertical Frames Sizes of Primary Members (leg pile vertical frames braces primary horizontals) Joint Details (can gap etc) LegPiie Annulus Grouting Status

bull middotPOST ANDREW INSPECTION REPORT Significant Wave Height Hs (ft)

bull INFORMATION ON DAMAGES PRIOR TO ANDREW

bull RESULTS FROM IN-HOUSE STUDIES IF AVAILABLE

bull ~RIGINAL DESIGN CRITERIA INFORMATION IF AVAILABLE

bull SOIL SHEAR STRENGTH PROFILE IF AVAILABLE

Platforms Identified for Evaluating Therr Likelihood of Falling Under Unexpected Survival Category

OPlRATOR

middotADA

BLOClt

llmucnJllE

NAME

YEAll

INSTAUJD

WA1ER

DBPTll

DISTANCE

JlllOM SHOllE

(MILU)

Bl

(M)

AMOCO 111113

STl11

WDI075

A

bull D

St

17

120

1n

44

20

10 I

CllllVllQfl STlllO c 61 110 21 11 STIUO D Q 1110 21 11

STl134 p Q 137 29 11

ST0151 G 61 137 11

ST0151 I 3 121 11

ST0151 J Q 140 11

STl151 It 137 11

ST0151 PJl001 Q UT 11

ST0151 PROD-2 Q UT 11

ST017 A 3 140 35 11

WDIUT c 65 214 34 10

llXXON STI055 II 51 6T 14 I

STl15 A t3 34

WDI073

WD0073

Amiddotshy Q

1111 1111

22

22

I

I

WDI073 c 1n 22 I

- WDOCl93-shy II

65 1110 2T A 3 194 31 MOBU 111121

SS IOl7

A B

501

50

50

31

T

I

T 6

MtlllPllY SS114 B T 50 T T

PlllllIPS SS0149 A 5 55 33 T

SAMllDAN STln A 107 T T

SBllLL WDl103 A 65 223 2T WDl103 B 65 221 2T WD0104 c 65 221 2T WD0104 D 6T 20 2T

UNOCAL SS020I

SS 020I SSl20I

SS 0209

SS0209

II p

PURll

A

B

3

n 61

103

97

97

95

100

34

34

34

34 34

bullbull bullbullbull

- SS0209 D u 95 34 I

- SS 1209 SSl253

SS0253

G A

c

6T

2

100

165

1T5

34

54

54

bullbull I

I I

I I

SOUTH MARCH

~LAND

N

SOUTH PaTOI ENE ISWltD AREA

MMS Defined Limits ror Offshore Structures Requiring Inspections Following Andrew

t---Jbulln3 T25bull I bullTtl

bull1$1Htl1Abull bull1SIK1779bull

~

middot-DaTA AREA~T

CRANO rsLE

AREA bull1300

bullll4W

=shy

bullPath of Hurricane Andrew with Platforms Used In Calibration

Likelihood Function vs b All Success Case STISI K STIJOQ ST134W WD90A MCJll MCJ97

I 0000

0 9000

0 8000

07000

-Bo6000

tiiii 0 5000 -

04000

0 3000

0 2000

01000

0 0000

0 05 25

ri1rt middot-middot-middot-middot-middot-

I 5 2

b

---11 middot151K(blsu~ccsI

- bull - bull - fk 130Q(bfU~lCSS)

lie bull134W(blsuccc]

lk-WD90A[blsultCS]

- - - lk middotMC397(bfUtlCS]

--o-- lk-MCJll(bbuctcsJ

--- ll (bl 6 out of 6 Mlttcs case]

_

( Likelihood Functions middotSuccess Cases - p - 1 c I

CATEGORIZATION OF ADDITIONAL PLATFORMS

I I I

bull CATEGORIES

Unexpected Survivals

Expected Survivals

bull APPROACH

Identify Corresponding Platform(s) Evaluated In the JIP

Determine Max Wave Height at Platform Location

Evaluate Whether Wave Hit the Deck

middot - Evaluate Likelihood of R and S to Differ from the Base Cases

Review Joint Information

Assess Special Features

bull IDENTIFY NUMBER OF PLATFORMS IN EACH CATEGORY

I

UPDATING OF CALIBRATION RESULTS

bull INCLUDE EFFECT OF ADDITIONAL PLATFORMS ON bullBbull

Consider Increased 1Uncertalntles In R and S

Develop bullNeW Ukellhood Functions for Different Categories

middotLet bullxbull Number of Ukely Unexpected Survival Cases and y Number of Ukely Expected Survival Cases

Determine Combined Likelihood Function lk (b Iadditional platforms) middot

bull REVISED LIKELIHOOD FUNCTION

lk(b I all cases) = lk(b I 9 platforms) middot x lk(b Ix platforms) x lk(b I y platforms)

bull ESTABLISH REVISED BIAS FACTOR B

f8 (b) u f8 (b) lk (b I all cases)

II

t t ~

shy

APPLICATIONS OF BIAS FACTORS

ii APPLICABLE TO PARTICULAR LOAD AND RESISTANCE RECIPE

bull USE BY THE API COMMITTEES AND MMS

Criteria Revisions

Recommended Design Guidelines Revisions bull

bullbull REASSESSMENT STUDIES AND DECISION MAKING

middot

bull RISK ASSESSMENT STUDIES BY THE OPERATORS

Individual Platforms

Fleet of Platforms

middot IMR Program Implementation

bullbull NOT TO BE USED IN DESIGN

I I fl J

EXAMPLE USE OF BIAS FACTORS

bull PHASE I EXAMPLE APPLICATION

bull EVALUATION OF ANNUAL PROBABILITY OF FAILURE

Establish Annual Maximum Seastate Data

Determine Coefficients to Define Base Shear

Determine Ultimate Capacity of Platform

Determine Probablllty of Failure Versus bullbbull

-bull DETERMINE UPDATED PROBABILITY OF FAILURE

Pf - P [S gt R]

Pf I P [~ gt Rb I b] f 8 (b) db

Pr - J P [S gt Rb I b] f bull 8 (b) db

bull TOTAL PROBABILITIES OF FAILURE

Total Pf - 1 bull Il [1 middotPf] - l Pf bull

Total Pf - 1 bull Il [1middot Pf] - l Pfmiddot

Total Pr = 1 - n [1middot PfbullJ = l Pfshy

I I I

EXAMPLE APPLICATION bull PHASE I

PLATFORM ST 151 K

bull GULF OF MEXICO

bull WATER DEPTH = 137 FT

bull ANNUAL MAXIMUM HS MEDIAN 175 ft COV 031

bull ASSUMED STORM DURATION 3 hoursmiddot

bull ASSUMED NUMBER OF WAVES 800

bull MEDIAN HS FOR STORM HOURS 1 2 AND 3

bull MEDIAN CURRENT VALUE bull

bull DETERMINE ULTIMATE CAPACITY FOR THREE DIRECTIONS

bull DETERMINE ANNUAL FAILURE PROBABILITIES

Broadsidemiddot Diagonal End-on

0037 0025 0007

0071 0044 0018

0022 0016 0004

DETERMINE TOTAL ANNUAL PROBABILITY OF FAILURE

Pf - 007

Pf - 013

Pr - 004

bull

EXAMPLE USE OF BIAS FACTORS - PHASE II middot

bull DETERMINE PREDICTIVE PROBABILITY OF FAILURE

Case - 1

Pf _ S S Pf (b b2) fa1a2 (b b2) db db2

g bull -1 (b1 R I S) - 1

92 - (b2 R2 I S) - 1

Case -2

Pf s ~ [bJ f11 (b) db

Pf [bJ p [g lt 0)

g - bull 1 (b1 R1 I S) - 1

1

bull1) l

I

BENEFITS OF ANDREW JIP bull PHASE II

lt

bull IMPROVED LOAD AND RESISTANCE RECIPE

bull IMPROVED CALIBRATION PROCEDURE

bull MULTIPLE BIAS FACTORS

bull PROCEDURE TO INCLUDE EFFECT OF OTHER PLATFORMS

bull PROCEDURE TO USE MULTIPLE BIAS FACTORS

  • Structure Bookmarks
    • ANDREW PHASE II JOINT1 INDUSTRY PROJECT
    • PROGRESS MEETING NUMBER 1
    • ANDREW PHASE II JOINT INDUSTRY PROJECT PROGRESS MEETING NUMBER 1 NOVEMBER 16 AGENDA
    • ANDREW PHASE II UST OF PARTICIPANTS
    • ANDREW PHASE I OBJECTIVES AND RESULTS
    • ANDREW PHASE II PROJECT OBJECTIVES
    • SUBSEQUENT PROJECTS
    • ANDREW PHASE II PROJECT TEAM
    • ANDREW PHASE II UPDATE MODELING middotAND ANALYSIS RECIPE
    • -ANDREW PHASE II CONDUCTOR RESISTANCE
    • ANDREW PHASE Il WA VE FORCE PROFILE
    • ANDREW PHASE II WA VE FORCE PROFILEmiddot
    • ANDREW PHASE II BRACE MODELING
    • ANDREW PHASE II FOUNDATIONS
    • ANDREW PHASE II middot middot JOINT MODELING
    • ANDREW PHASE Il JOINT MODELING OPTIONS DISCUSSION
    • ANDREW PHASE II JOINT MODEJJNG OPTIONS ILLUSTRATIONS
    • ANDREW PHASE II JOINT MODELING PHASE II METHOD
    • ANDREW PHASE II ANDREW STORM CONDmONS UPDATED
    • Platform Selection bullbullbull1111111111111
    • BasisCriteria for Selection bullbullbull1111111111111
    • CALIBRATION
    • WHERE TO INTRODUCE BIAS FACTOR (B
    • PHASE I -KEY CONCLUSIONS
    • bull CALIBRATION ISSUES PHASE II
    • PHASE II WORK SCOPE (CALIBRATION)
    • SOFTWARE
    • SENSITIVITY STUDIES POSTERIOR DISTRIBUTION OF B
    • WEIGHTING APPROACH bull
    • INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON
    • CATEGORIZATION OF ADDITIONAL PLATFORMS
    • UPDATING OF CALIBRATION RESULTS
    • APPLICATIONS OF BIAS FACTORS
    • EXAMPLE USE OF BIAS FACTORS
    • EXAMPLE APPLICATION bull PHASE I PLATFORM ST 151 K
    • EXAMPLE USE OF BIAS FACTORS -PHASE II middot
    • BENEFITS OF ANDREW JIP bull PHASE II
Page 10:  · . ' ANDREW PHASE II PROJECT OBJECTIVES . ' • ·UPDATE MODELING AND ANALYSIS RECIPE· JOINTS ' ' ·BRACES WAVE PROFILE FOUNDATIONS . ' ' CONDUCTORS· • ANALVZE PLATFORMS USING

Project ST151K Model Conductor version 1 Mon Nov 14 143140 1994

- ---- ---

--

117 s

111

7i

31

0 -5 -~ - ~()

- iltgt --lo

-(pO

-so

bull bullf OC)

-120

-ISo

-ISO

z

CAgt ~x Singl_e Conductor Model

bull

=I=

-_

-_

-gtshy

~ -

~

I

I~

- gt

~-----------____________________

Project STlSlK Model dir_2 version 7 Mon Nov 14 124200 1994

177 s

Ill

74

37

0- - e -zo - ~o

- Jo

-100

-10

-ISO

z

UV 1-+x

Project STlSlX Model Conductor Version 1 Mon Nov 14 140620 1994

bullmiddot

I

I

I I

I

I I

I I I I I

1 ==

-L

-L - L

-

- L

-

-

L

-

Uniform Displacement step 6 Inelastic Events Legend

Elastic

Strut Buckling Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield Beam Clmn FUlly Plastic Fracture

--

i~~~~~~~~~~~---~--~~-J___--JJ~_middot~~~=-

Project ST151K Model Conductor Version1 Mon Nov 14 142735 1994

---

~

I

I

middot

I

I

I

- shy- shy

- shy

- shy

4x z

CIV Uniform Displacement step 33 Inelastic Events Legend

bull Elasti~ Strut Buckling Strut Residual Strut Reloading Plastic StrutNLTruss Beam Clmn Initial Yield Beam Clmn-Fully Plastic Fracture

gt bull

Project STlSlK middot Model Conductor Version 2 Mon Nov 14 142025 1994

-----------------

bullf

I

I

I

II I I

I I I I

I ==

-

Variable Displacement step 9 Inelastic Events Legend

Elastic Strut Residual

bull

Strut Buckling Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield Beam Clmn Fully Plastic Fracture

--

Project ST151K Model Conductor Version 2 Mon Nov 14 142145 1994

I I I I

I I

I

I

I - - -

-Ishy-

-

Variable Displacement step 35 InelasticEvents Legend

Elastic Strut Buckling Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield Beam Clmn Fully Plastic Fracture

Project ST151K Model Conductormiddot

Version 1

Clgt - Cut Plane Force Fx Mon Nov 14 140051 1994

- -_ vii I fOIZM

0 _

~

N ID 0 bull 0

x 0

middotrt ~ () bull u 0 ~ 0 Ir

N bull

0

0 1 2 3 4 5 X Deck Disp [Ft)

I I 1 1 WbullTI- bullbull

lltcxi lr-C

011lt~~~~~~~---~~~~~~~~L~~~~~~~--~~~~~~~~i~~~~~~~--

Uniform Node Displacement

shy

Ji

r

Comparison of Pushover Curves for ST 151K Variable Load Pattern and Monotonic Load Pattern (Andrew)

Soll Improved 100

6000

5000 -i 4000

f

llS

3000 Ul

J 2000

10 c11 c1lgtolts 1000

0 v

bull ~

f

_ ----- --- ---shy

bull --Variable Lold

- - Andrew Load

I I I I I I

4 5 62 30 1

Deck Dlsplacement (ft)

bullbull r 1 II

~

ANDREW PHASE Il

WA VE FORCE PROFILE

bull DECK WA VE FORCE ~

PHASE I USED SECTION 17 (1056H degCd)

PHASE Il USE REVISED SECTION 17 (1000H Cd)

bull PHASE I USED SINGLE FORCE PROFILE FOR PUSHOVER

- USED WAVE HEIGHT CLOSE TO ULTIMATE CAPACITY

bull PHASE J

Il WILL STUDY THE

USE OF VARIABLE

PROFILE ~

- IF FORCE LEVEL INCREASES (OR DECREASES) BY A FACTOR EQUIVALENT TO A 2 FOOT CHANGE IN WA VE HEIGHT THE INCREMENTAL FO~CE PA1IERN IS CHANGED ACCORDINGLY (CAP FEATURE)

- INITIAL EXAMPLE

f~

a

middotn

a

I N -a a a amiddot

+

~ a

l

a w gt 0c 0 gt a I shyc -CJ wc w gtc(

== c( I shyZ w E W a 0 z- 8

ii sect

~ ~ bullT 0

I if

x ~

bullbullbull

II ~------------________

ANDREW PHASE II

WA VE FORCE PROFILEmiddot

CAP AUTOMATIC APPLICATION FEATURE

STAT

fR1 0 lt f lt1

STAT

STAT

f(R3-R2) middot f lt1

Project ST151K Model Wave_Loading Version 1 rue Nov 8 164107 1994 PT

E~ nss FT

SL 14amp FT

a 111 FT

EL 14 li

E-1 37 FT

El o FT

1middot1middot ______________ _ ________ __ _ __

0

200 400 600 800 1000 12000 Force at Elevatloii (kip)

200

a ll_

Wave Force at Major Elevations for Discrete Wave Heights

180

bullHmiddot40ft

Hmiddot495ft

H bull 51 ft

Hmiddot55ft_

H-59ft I

Andrew (H bull 60 86 ft) II

i I

_11---__

- middot- Hmiddot30ft

-

- bull -

middot bull bull bull middot

---Hmiddot53ft

bullbullbullbullHmiddot57ft

-----H-63ft

Elevation Above Mudllne 100

(ft)

180

140

120

80

60

40

20

Loading at Major Elevations

---H-49511

-deg-Hmiddot608611

0 200 400 600 800 1000 1200

Force It Elevatlon (kip)

--

Base Shear vs Wave Height ST151K

6000

5000 1---------Ult1mateCapacitybull 4890 kip (100 Improved Soll)-----------~--

4000a a lii ~ 3000

81 ampI

2000

1000

0 +-~~~--i~~~~~~~~~-+-~~~--lf--~~~~~~~~-+-~~~--l

30 35 40 45 50 55 60 65

Wave Height (ft)

bull middot

~

bull

II

2

g

~

It)

0

cCD E 0 E a c E euro~ CDshygtItgtmiddoto ~ti c -a Gic CD gt ~

-= c -Cll

I z

bullO 0i i ~ g 0

() CI -bull(q-ll) iuawow llu1wniiaAO

11

shy

116

114

112

g 110

bullc = 108

~ ~ 106 -sect 104

102

100

98

96

-

6555 60

Wave Height vs Center of Action ST151K

30 35 40 45 50

Wave Height (ft)

t

Comparison of Pushover Curves for ST 151K Variable Load Pattern and Monotonic Load Pattern (Andrew)

Soll Improved 100

--- --------- --- ----6000

5000

- 4000 8 - I

1 3000 Ul

81 m 2000

_J -shy

1000

---Variable Load

- - Andrew Load

0 f-~~~~~r--~~~~-t-~~~~~-t-~~~~~+-~~~~----t~~~~~-1

0 1 - 2 3 4 5 6

Deck Displacement (ft)

-middotshy

1 t I I ~

ANDREW PHASE II

BRACE MODELING

bull MARSHALL STRUTS USED FORPHASE I

bull PHASE II WILL ASSESS DIFFERENCE IN USING BUCKLING BEAMS AND MARSHALL STRUTS

middot bull COMPARE FOR 2D FRAME

bull COMPARE FOR PLATFORM 177 WHERE LOWER

BAY EXHIBITS LARGE BRACE MOMENTS

Project andrew Model beamframe Version 1 Mon Nov 14 145447 1994

~

middot Frame Test buckling beam-column model

-

Project andrew Model strutframe Version 1 Mon Nov 14 145754 1994

bull

Frame Test strut model

middot-middot _____________--_

_middotmiddotr_~ --ll--~-middot__

Buckling Beam bull

Strut (K=O 65)

Lateral Displacement

ANDREW PHASE II

FOUNDATIONS

bull ~LICIT NONLINEAR API FOUNDATION USED INiPHASE I

bull FOUNDATIONS

JIP SELECTED MINI-VANE DATA FOR

STRENGTH PROFILE

bull COMMENTS ON BENCHMARK COMPARISONS

- SIGNIFICANT VARIABILITY DUE TO FOUNDATION MODELING METHODS

II Ii

z

ctV 4x SPlO caissonfy-42s soil-api(lf-097step-114)

Inelastic Events Legend Elastic Strut Buckling

Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield

--------shyBeam Clmn Fully Plastic Fracture

~--~~~~~~~~~--~----~__i~~---~~~~__--~

Pro)ect apipile Model South Pelto 10 Version Wed May 18072235 1994 - -

1bullt

+

t

f i bullbullbull

bullbull bullbull

I I

5

~x SPlO Caissonfy=42s soil=api(lf=lOOstep=137)

Inelastic Events Legend

Elastic Strut Bucklilg

Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield

earn Clmn Full~ Plas1c Fracture

ANDREW PHASE II middot middot

JOINT MODELING

middotbull API RIGID-PLASTIC USED IN PHASE I

a ELASTIC-BRITTLE USED IN PHASE IB

amiddot MSL WILL PROVIDE NONLINEAR FORCE-DEFORMATION

AND MOMENT-DEFORMATION RELATIONS FOR PHASE II

a MODELING OPTIONS FOR PHASE II -

l

- UNCO~LEDmiddotONE DIMENSIONAL MEMBER END JOINT MODELING

J Ii

ANDREW PHASE Il

JOINT MODELING OPTIONS DISCUSSION

bull JD COUPLED INTERACTION SURFACE

- IDEAL

- DATA NOT AVAILABLE TO CALIBRATE

bull 2D COUPLED INTERACTION SURFACE

- SIGNIFICANT MODELING IMPROVEMENT

- DATAAVAILABLE

- IMPLEMENTATION PRACTICAL BUT NEEDS DEVELOPMENT AND CALIBRATION

bull 2D UNCOUPLED INTERACTION SURFACE

- SIGNIFICANT MODELING IMPROVEMENT

- DATA AVAILABLE

- IMPLEMENTATION PRACTICAL AND AVAILABLE

ANDREW PHASE II

JOINT MODEJJNG OPTIONS ILLUSTRATIONS

P

e

ANDREW PHASE II

JOINT MODELING PHASE II METHOD

e

I I I

ANDREW PHASE II

ANDREW STORM CONDmONS UPDATED

bull -PHASE I USED OCEANWEATHER HINDCAST-(Nov 1992)

bull middotmiddotPHASE II WILL USE THE REVISED HINDCAST (Nov 1994)

bull lt c lt I lt

2-2

l---4-~~~+-~~-+-~~~1--~~-+-~~--l--~T--+-~~--i24

f-tJVtMfiz~ 1Cfl (r STvb(

Plollol bullbull 5middotNOYmiddot9214bull1J Irbullbull Illbull [ANDREVl~NDREVCUSTbullMAXll 5-NOY-1992 10bull54 - 9208

HEIGHT lml

I~ MAXI

I L 1~middotamp SZ~- ~A I 13c

I J) ( r I X~0iWJ~0gtJ gt L gt gt I 1c I 120

-- N

~ deg

I I I I I I e C I ~=-41 122 -94 -92 -90 -88 -86 -84 -82 -80

-

Figure 7 contours of maximum hindcast significant wave height

_

~

~tl ~

NOltNttIJ 17lt ltNS S iJJ)y

Maximum Significant Wave Height - Hurricane Andrew 1992

30 ~

26 ~ I ~-~~~- == -shy -~ 35-_

24deg1--shy --------~- middot-middot-------middotmiddotmiddot----middot------middot-middot~middot- - - ~-----1---middot

22middot ~~~~~~~-~ -94 -92 -90 -88 -84deg-86 -82 -somiddot

c

-

Platform Selection bullbullbull1111111111111

0 BasisCriteria for Selection middot

0 Review of Phase I Platforms middot 0 Alternative Platforms

0 PMB Recommendations

0 Participants Comments

gt

L

l I

BasisCriteria for Selection bullbullbull1111111111111

0 Benefit for Calibration - Classification (UnexpectedExpected)

- Failure Mode (StructuresFoundations)

- Damaged StructuresKnown response

- Better Representation of Full ~opulation

0 Availability of Data - Existing Models

- Inspection Data

- Damage Data

PREDICTED BEHAVIOR

-

OBSERVED

BEHAVIOR

-SURVIVED DAMAGED FAILED

SURVIVED

ST134W WD90A

-MC311 -

MC397 -

--

-

-

ST151K ST130Q

-

-

-

DAMAGED - -

-

-

--

-

-

ST52 SS139 ST177B -

ST161A

-_

-

FAILED

-

-

-

ST151H ST130A ST72

Operator - Platfonns - Platfonns Survived Damagedamp

- Falled -

bull

-

-

Jle eBDIGIPllDbl -Amocoshy 30 1 Chevron -

143 12 Exxon 72 Mobll 22 1 Phllllps 7 bull

-Shell 39 Unocal 25 1

Total Partlclnts Platfonns 338 15

Other Operators bull- 310 13

-

Total Platfonns - -

648 28

fbml

Platforms Considered 6 7

tie of Total Platfonns 1 25

OVERVIEW OF PLATFORMS SURVIVEDDAMAGEDFAILED

-

shy

v

Platforms Evaluated in Phase I

-

-Platfonn Water Year Number Analysis Avallabillty of Availability Platform

- Name Depth rl of Model DetaDed rl ClassUlcatlon

Installation legs Inspection Soll Report Based on Information Phase I Calibration - -ft

- -- Suaill flalfnrm ~BSi -STISIK 137 1963 8 CAP Yes Yes Un eel Sarvlval

-STIJOO 170 1964 4 CAPmiddotshy Yes ST134data u ST134W 137 1981 4 CAP Yes Yesmiddot E

-

edSarvlval -

middotsurvival

WD90A 184 - 1964 8 CAP Yes E -

- Sarvlvill

MC3ll 343 1978 8 KARMA

MC397 468 1991 4 KARMA -

bull ~reglmmiddot- -

Yes Sure Survival

bull Yes - Sare Sanlval-

-T23STS2 63 1969 4 CAP

T2SSS139 62 1969 4 CAP -

Yes

Yes

Yes Ex

Yes Une

bull SurvivalLikely to DamalI

middot bull SanlvalLlkely to DamalI

ST161A 118 1964 8 USFOS

Eail11a pound1al(bullllm Clss ST177B 142 --1965 8 CAP

STISIH 137 1964 - 8 CAP C

STIJOA 140 19S8 8 CAP - -

Yes

Yes

ollamed

Yes

- Yes E

-

at about 1 mUe

Yes

ST134data

bull SafvlvalLlkelv to Dam- bull- -

E middotrahre

Emeried Failure -

Likely FaUare

T21ST72 61 1969 4 CAP Collamed No Likely FaUure

Platforms Identified for Evaluating Their Likelihood of Falling Under UnexpeCted Survival Category

CllEllATOa

AllB

llLOCll

Sl1lvcnJllE

NAME

YEAR

INSTAUJCD

WATER

IJBPJB

middotmiddotDISTANCE

ftOMSllOllE

(MIUIS)

Ill

(M)

AMOCO BllW

STllI

WDI075

A

bull D

4

n uo 17J

44

32

bull 20

10

bull CllEVllOH STtllO c I 1bull 21 11

STl130 D 0 161 21 11 STllJ4 I 0 137 D - 11

STIUI G I 137 32 11

STIUI I 0 121 32 11

STIUI

J 0 140 32 11

STIUI It 4 137 32 11

STllSI PROl)I 0 137 32 11

STllSI PROD-2 0 137 32 11

STll7 A 0

140 35 11

WDl117 c 65 214 34 10

DXON STshy E 51 14 bull STl165 A t3 bull 34 WD0073

WD0073

WD0073

A _ c

4

62

4

IA IA

17J

22

22

22

bullbullbull WD003-shy E 65 161 Z7

A 0 IN 31 MOBD Bl12I

ssn A bull

50

50

31

7

bull 7

MIJllPllY SS114 B 7 50 7 7

PlllUIPS SSl14t A 55 33 7

SAMEDAN ST17J A 4 117 7 bull 7

SJIELL WD0103 A 65 223 27 WDllOJ bull 65 221 27 WDll04 c 65 221 27 WDOI04 D 67 2 27

UNOCAL

SSl20I

SSl20I

bull SSl20I

SS0209

SS0209

SS0209

SS0209

SS 0253

SS 0253

E

I

PtJllE

A

bull D

G A

c

0

4

4

7J

61

0

0

69

103

97

97

9S

100

95

100

165

175

34

34

34

34

34

34

34

54

54

bullbullbullbullbullbullbullbullbull

Platforms Recommended for Phase II

llalr y_ w- Aot Av- _ I 1 Doplla Model nm bull bull _ bull I I H11 s18-1 Irht lbwl~ bullbullbull(-illtdlel OM lllMtD

II ---

-STISIIC 137 1963 I CAP Yeo Yeo I Yeo

STtshy -170 1964 4 CAP Yeo ST134dlU I Yeo

STl34W 137 Yeo Yeo -1981 4 CAP D Yeo

WD90A 114 1964 I CAP Yeo - D

MC311 30 1971 I KARMA Yeo -sre-

MC397 middot 461 1991 4 ICARMA Yeo Sireshy - - r- -

__ T13ST5l 63 1969 4 CAP Yeo Yeobull middot- ~n- D Yeo _ Tl5SSl39 62 1969 4 CAP Yeo Yeo I Yeo

_ STIT711 142 1965 I CAP Yeo I Yeo middot-1shySTl61A Ill 1964 I USFOS Yeo Yeo - D bulloshy

- r-

STISIH 137 1964 I CAP y D y -_ STl31A 140 1951 I CAP Yeo STl34dlU I Yeo

T21Sll 61 1969 4 CAP No rar-middot I Yeo

~-rmiddot-middot SS209A 95 1967 16 CAP Yeo D -SSU4R

11 UI A 41 -in D SS 2741AIM CAP bull

_ _ -shy - -shy- -shy

CALIBRATION

bull PHASEIAPPROACH

bull CALIBRATION ISSUES

bull PHASE II WORK SCOPE

bull SOFlWARE AVAILABILITY

bull EFFECT OF ADDITIONAL PLATFORMS

bull WEIGHTING APPROACH

bull EXAMPLE APPLICATION middot

(l=Ofce Mode] Wind

~

fcurrent l -middot

~SW ~leldl4

ultimate resistance force I first

design load Illgt

i _

deflection

-

bull

4

Historical Hurrlcane Dariiage and Survival of Platforms

Components Required for Calibration (Petrauskas 1992)

PRIOR Distribution

PLATFORM EXPERIENCE

BAYESIAN Survivals and Failures

UPDATING

Waves Winds and Currents bull

POSTERIOR Distribution

- A_middotJA ~ Survival Data Failure Data

Result Result

Load Resistance

middotfl

-

Bayesian Updaing middotGeneral Description (Petrauskas 1992)

L Imiddot

WHERE TO INTRODUCE BIAS FACTOR (B

On Resistance (Rmue - B (RPREDICreD I

middotOn Loads (Smue middot - B (SPRE01cTEo

On Safety Margin (RSmue - B (RSPREDICTED

CALIBRATION METHODOLOGY

gt

r

FIELD OBSERVATIONS

Observed Behavior

survivals Damages

Failures

r

bull

CAPACITY ANALYSIS

Predicted Load and Resistance Predicted Failure Modes

Survival Damage Failure cases

RELIABILITY ANALYSIS

- Probability of Failure ~

Ukellhood Failure

BAYESIAN UPDATING

Prior Distnbution of B Combined Likelihood Function

Posterior Dlstnbution of B

BIAS FACTOR

Mean Value cov bull

Represents Modeling Uncertainties

Prior Distribution of B Likelihood Functionmiddot Observing Success case STISIK and Posterior Distribution of B

2 00

uo

160

140

120

e 100 L $ uoIshy

0 60 ~ ~ 040

020

0 00

0 00 0 50

_ I 00

b

I 50 2 00

-shy lk [bl STISIK)

--shy Pnor-f(b)

--shy Pos1enor r(bl ISIK)

Prior or B Mean bull 10 COV bull 030

Posterlor of Bbull Meanbull 136COV 0 16

2 50

Likelihood Function and Posterior Distribution Example Platform STISIK

- -

~

- Prior and Posterior Distributions or B Failure cases (ST177B STlSlH ST130A T21) Damage cases (T23 T25 ST161A) Success cases (STlSlK ST130Q ST134W WD90A MC311 MC397)

400

350

300

2 so 0-

I

bull bull

bull bull gt

- bullbull ~ bull bull I bull

_ bull bull bullbull I bull bull bullII V -

bull bull

I bull ~ bull

bullbull bullI bull bullI bull bull bull

middot -~

bull__ ~ middotI middotmiddot

L 2 00

e ~ I SO

-I 00

- 0 so

000 gt

0 00 020 0 40 060 0 80 I 00 120 140 160 110 200

b -

-

shy

Prior-r

- - - - Posterior fLlll failure cases

bull bull bull bull bull bull bull Posterior flall damage cases

- bull bull - bull middot Postenor flall success cases

Posterior flall success fulure damage cases

-

Posterior of B Mean= 119 COV 0101

Posterior Distribution or Blas Factor (8) bull Different Categories and All Cases Combined

shy

l

PHASE I - KEY CONCLUSIONS

bull GLOBAL CORRECTION FACTOR

bull 15 TO 20 CONSERVATISM MEAN)

bull UNEXPECTED SURVIVALS HAVE GREATEST IMPACT

bull DAMAGED CASES MORE INFORMATIVE

bull CALIBRATION ISSUES PHASE II

bull

EFFECT OF IMPROVED CAPACITY ANALYSIS RECIPE

bull APPLICABILITY TO MULTIPLE FAILURE MODES

bull NO FOUNDATION DAMAGE CASE INCLUDED

bull IMPROVED REPRESENTATION OF FULL ANDREW DISTRIBUTION

-bull NUMBER OF PLATFORMS CONSIDERED

I 1 I

I t ) bull ~

PHASE II WORK SCOPE (CALIBRATION)

bull

bull ESTABLISH MULTIPLE BIAS FACTORS

Use Nine Base Case Platforms

middot - Capacity Analysis for Revised recipe

Foundation Blas Factor from Foundation JIP

bull DEVELOP WEIGHTING APPROACH

bull UPDATE MULTIPLE BIAS FACTORS

bullEffect of Platforms Not In the Nine Base Cases

bull EXAMPLE APPLICATION OF BIAS FACTORS I

bullbull

PHASE I PHASE II

CAPACITYbull ANALYSISRu bull

---------------------1----middot

Jc1 - iJbullraquoJ dbull

FttlS frt[f~l ratdhJJ

-

LIKELIHOOD FUNCTION

DEVELOPMENT

lk Jb Jfall1N) bull Jb)

111 Ill Isurvtvoq bull 1 - bl

-middot -------------------shy ~---------------------middot-middot

r111ir8 Cb1 111 Cbl BAYESIAN

UPDATING

~cove

R1R2 bull middot---------------------~----

Jblbull P(g cOJ

1 bull JbRI (MBSI I bull 1

~middot

llcJllJfalluro) bull P(gcOJ Ill 111 b21fallun) bull P ((g1 CO) U 1112 c OD

middot--------------------shy ----------------------shy bull

r111111r8 Cbl 111 Ill i-1nro1

deg ~1b2)laquottfrr (b1b21-lnfo) re Ill bull 1-11middot b2) bull lodegamp

middot

i1 covamp i2 cov~i

-

shy

--

Likelihood Function vs b All Success Case STISI K STl30Q STl34W WD90A MC311 MC397

I 0000

0 9000

OIOOO

0 7000

~ 0amp000

ta osoOO

~ 0 4000

0 3000

0 2000

0 1000

bull 0 0000

bull

I

middot

middot bull I

Ibull

-sibull ~ -middot-middot-middot-middot -middot-

----7----+-----~ I

Ibull bull bull I

rbull

--- Jk ISIK[blsuccess)

- bull - bull - lk 130Q[blsuccess) bull

lk 134W[blsuccess)

Jk-WD90A(blsuccc~I

- - - lk MC397[blsucccss)

--ltgt-- lk middotMC31 l[blsucccss)

--- Jk [bl 6 out of6succcsHascs)

05 I 5 2 250

b

Likelihood Functionsmiddot Success Cases - Phbullsc L

shy

bullbull bull bull bull bull bull -------------------~

bull bull bullbull bullbull bullbull bullbull bullbull bullbull bull

bull bullbull

bullbullbullbull

bullbull

bullbullbullbullbullbullbull

bullbullbullbullbullbull

bullbullbullbullbullbullbull

z 0

t z ) LL

c 0 0c i ~ i

J

~ sect ~--------------~ e l5

I

I

SOFTWARE

bull PF PROGRAM

bull CALREL

bull PF +CALREL

bull RELACS

bull

gt i

l

SENSITIVITY STUDIES

POSTERIOR DISTRIBUTION OF B

bull EFFECT OF CHANGE IN DISTRIBUTION OF PRIOR OF B

bull EFFECT OF REMOVAL OF ONE PLATFORM FROM EACH CATEGORY

bull EFFECT OF REMOVAL OF THREE DAMAGE CASES

bull EFFECT OF ADDITIONAL 13 SIMILAR PLATFORMS

-I

Prior and Posterior Distributions of B Variation In COV of PriorofB fromOl to04 -- -

4500

-4000

-

- -

3500 --middot - bull bull - bull bull Prior-r ICOV =02

3000

- - - - Prior-r ICOV =03 -

bull -~ e 2soo -

bull bull bull bull bull bull bull Pnor-r ICOV =o 4t frJ ~ 2000 Posterior rlCOV =0 2

bull bull Postenor rlCOV =0 3 bull bull 1500 bull middot bull - shy -middot- Posterior rlCOV =04

middot~ I 000 - bull

-middotmiddot ~ ~middotmiddot M -~

- middot 0 500

wbullbullbull _ I middot~ middotmiddotmiddotmiddotmiddot middotmiddotmiddotmiddotmiddot - -~ ~ __ middot_ _ bull -shy0 000

0 000 0200 0 400 0 600 0 800 IOOO 1200 1400 1600 1800 2000 -

- - b

-

=shy

Posterior Distribution of Bias Factor (B) bull Effect of Variation In COV of Prior of B

bull

bullPrior and Posterior Distributions of B Evaluating Effect of Removal of One Platform from Each Category from the Base Case of 13 Platforms

400

~ ~ 350

j IA i

3 00 bull I bull I ---Prior-r I bull l

250 1 I 1

I bull bull ---Posterior r lall 13 casesshy- ~E 2 00 bull bull bull bull bull bull bull Posterior r without STISIK

- ~ 1 bull - bull I - bull bull - bull bull Posterior r Iwithout T2I I 50 I

J-tt~i - bull - bull - Posterior r Iwithout T2S

tOO [-

I I ~ ~ I 0 50

0 00

___ r J i

l~T

~--

0 00 0 20 0 40 0 60 080 I 00 120 140 160 110 2 00

b

zshy

Posterior Distribution of Bias Factor (8) bull Effect of Removal of middotPlatform from Any Category

bullPrior and Posterior Distributions or B Evaluating Effect or Removal or 3 Damage Cases from the Base Case or 13 Platrorms

4

35

3

25

e L 2 ~ bulle shy

I 15

bull I

05

0

0 02 04 06 I 2 I 4 16 IS 2

I I

I

I

I Ibull

J -tmiddot L--

bullI

~

I

fl

II ~

I I v I

middot bull bull

bull L

~ ll- I-~lt

OS I

b

I

bull

1middot bullbullbullbull

~ __

bull

- - - - Prior-r

- - - - Postenor flall failure cases

Postenor flall success cases

--- Posterior flall 13 cases

---Posterior fI success failure cases only

IPosterior or B Mean =t11 COV =0121

-

_I

Posterir Distribution or Bias Factor (B) bull Effect or Removal or 3-Damage Cases

-

bullPrior and Posterior Distributions or B Evaluating Effect or Additional 13 Similar Observations as

Used in Andrew JIP The POsterior or B from Andrew-JIP Considered as New Prior or B

- 500

450

400

350

bull bull

bull bull bull bull J bull

bull bull f ~ I

I ~ Ibull bull

bullbull bull

l bull I -bullbull bull bull

I

I bull bull

bull I I~~bull bull bull

bull ~

I bull bull

~ ~_ ~~~

- - New Prior r bull

Posterior 114 new failure cases ~ 300- bull bull bull bull bull bull bull Posterior f 13 new damage casesl 150 ~

- bull bull - bull bull Posterior 116 new success cases i 2 00

Postenor 1113 new success failure -150 damage cases

I 00

0 50 -

0 00 0 00 0 20 0 40 0 60 0 80 I 00 120 I 40 160 110 2 00

Posterior ol B Mean= 120 COV= 0071 b

shy

shy

_rshy

Posterior Distribution of Bias Factor (B) bull Effect of Additional 13-Similar Observations

1

1 I

WEIGHTING APPROACH bull

bull ASSESSMENT OF PLATFORMS DATA

Select Platforms Most Mappable to the Analyzed Platforms

Unlikely SurvivalDamage Cases Identification

Based on Cut-Off Criteria Agemiddot Hs etc

bull INFORMATION REQUIRED FOR COMPARISON

bull CATEGORIZATION OF ADDITIONALmiddot PLATFORMS

bull UPDATING OF CALIBRATION RESULTS

--

OVERVIEW OF PLATFORMS SURVIVEDDAMAGED -~

Operator -

- Platforms

Survived CS)

Platforms

Damaged CD)

Platforms Used In

Phase IJIP -

JI~ ~Bill~IRBDll -

Amoco -

Chevron -

Exxon

Mob II

Phllllps

Shell

Unocal -

30

143

72 22

7

39 25shy -

1

12

bull

1 bull

1

1(S)+1 (D)

3 (S) +3(D)

2 (S)

bull shy

bull

bull

bull

Total Partlclnants Platforms 338 - 15 10C6S+4D) -

Other Operators

310

13

3 (D)

Total Platforms 648 28 13C6S+7D) -

Summary of Andrew JIP bull Phase I Platforms

Plairshy w v- N-ber Lea-le 1 shy car Hind_ o-w N-bull Deplh fl fl IJpe la bull n Yeadon Jadtet lbelledtf liefalit -- _ -OD

IL H - -shy _ICllshy IL IL - - rQClto ~ _

K STUIK 137 1963 8 365 608S4 Cellardeck

-middot STl-shy 170 1964 4 -~ K 39 62269 Cellardeck - --ST134W 137 1911 4 43 608S4 Smvlnlshy- _ - -

WD90A 184 1964 8 K 267Sbull34 83 SOS31 SUboftllar deck SmvMI

MC311 343 1978 8 x S7 62 - MC397 468 1991 4 x shy S33bull6SO 65911 -shySmeSmvlvol -- gt -

r-- T23ST52 63 1969 4 ton- shyKshy 52594 S03 -T25SSl39 62 1969 4 Kshy SOS94 S063 - middot ton-shy__ -

bullbull a ton--shyST161A 118 1964 8 K - 347S 51495 Cellarclecti

r~

ST1771 142 1965 8 K 39 60167 CeDordeck failme

STISIH 137 1964 8 K 395 608S4 CeDardeck Failme

K ST130A 140 1958 a SS 6095 I lblv Pai1uno - T21ST72 61 1969 4 Kshy S4S94 4974 I lhlv Pai1uno -

- -

JE

INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON

bull WATER DEPTH

bull bull YEAR OF INSTALLATIONDESIGN

bull PLATFORM ORIENTATION

bull PLATFORM LATITUDE AND LONGITUDE

bull KEY SKETCHES OF PLATFORM STRUCTURAL FRAMING

Number of Legs Plies Leg Batter Number of Conductors bull Deck Elevations (Including sub-cellar deck If any) Jacket Framing Elevations Brace type In Vertical Frames Sizes of Primary Members (leg pile vertical frames braces primary horizontals) Joint Details (can gap etc) LegPiie Annulus Grouting Status

bull middotPOST ANDREW INSPECTION REPORT Significant Wave Height Hs (ft)

bull INFORMATION ON DAMAGES PRIOR TO ANDREW

bull RESULTS FROM IN-HOUSE STUDIES IF AVAILABLE

bull ~RIGINAL DESIGN CRITERIA INFORMATION IF AVAILABLE

bull SOIL SHEAR STRENGTH PROFILE IF AVAILABLE

Platforms Identified for Evaluating Therr Likelihood of Falling Under Unexpected Survival Category

OPlRATOR

middotADA

BLOClt

llmucnJllE

NAME

YEAll

INSTAUJD

WA1ER

DBPTll

DISTANCE

JlllOM SHOllE

(MILU)

Bl

(M)

AMOCO 111113

STl11

WDI075

A

bull D

St

17

120

1n

44

20

10 I

CllllVllQfl STlllO c 61 110 21 11 STIUO D Q 1110 21 11

STl134 p Q 137 29 11

ST0151 G 61 137 11

ST0151 I 3 121 11

ST0151 J Q 140 11

STl151 It 137 11

ST0151 PJl001 Q UT 11

ST0151 PROD-2 Q UT 11

ST017 A 3 140 35 11

WDIUT c 65 214 34 10

llXXON STI055 II 51 6T 14 I

STl15 A t3 34

WDI073

WD0073

Amiddotshy Q

1111 1111

22

22

I

I

WDI073 c 1n 22 I

- WDOCl93-shy II

65 1110 2T A 3 194 31 MOBU 111121

SS IOl7

A B

501

50

50

31

T

I

T 6

MtlllPllY SS114 B T 50 T T

PlllllIPS SS0149 A 5 55 33 T

SAMllDAN STln A 107 T T

SBllLL WDl103 A 65 223 2T WDl103 B 65 221 2T WD0104 c 65 221 2T WD0104 D 6T 20 2T

UNOCAL SS020I

SS 020I SSl20I

SS 0209

SS0209

II p

PURll

A

B

3

n 61

103

97

97

95

100

34

34

34

34 34

bullbull bullbullbull

- SS0209 D u 95 34 I

- SS 1209 SSl253

SS0253

G A

c

6T

2

100

165

1T5

34

54

54

bullbull I

I I

I I

SOUTH MARCH

~LAND

N

SOUTH PaTOI ENE ISWltD AREA

MMS Defined Limits ror Offshore Structures Requiring Inspections Following Andrew

t---Jbulln3 T25bull I bullTtl

bull1$1Htl1Abull bull1SIK1779bull

~

middot-DaTA AREA~T

CRANO rsLE

AREA bull1300

bullll4W

=shy

bullPath of Hurricane Andrew with Platforms Used In Calibration

Likelihood Function vs b All Success Case STISI K STIJOQ ST134W WD90A MCJll MCJ97

I 0000

0 9000

0 8000

07000

-Bo6000

tiiii 0 5000 -

04000

0 3000

0 2000

01000

0 0000

0 05 25

ri1rt middot-middot-middot-middot-middot-

I 5 2

b

---11 middot151K(blsu~ccsI

- bull - bull - fk 130Q(bfU~lCSS)

lie bull134W(blsuccc]

lk-WD90A[blsultCS]

- - - lk middotMC397(bfUtlCS]

--o-- lk-MCJll(bbuctcsJ

--- ll (bl 6 out of 6 Mlttcs case]

_

( Likelihood Functions middotSuccess Cases - p - 1 c I

CATEGORIZATION OF ADDITIONAL PLATFORMS

I I I

bull CATEGORIES

Unexpected Survivals

Expected Survivals

bull APPROACH

Identify Corresponding Platform(s) Evaluated In the JIP

Determine Max Wave Height at Platform Location

Evaluate Whether Wave Hit the Deck

middot - Evaluate Likelihood of R and S to Differ from the Base Cases

Review Joint Information

Assess Special Features

bull IDENTIFY NUMBER OF PLATFORMS IN EACH CATEGORY

I

UPDATING OF CALIBRATION RESULTS

bull INCLUDE EFFECT OF ADDITIONAL PLATFORMS ON bullBbull

Consider Increased 1Uncertalntles In R and S

Develop bullNeW Ukellhood Functions for Different Categories

middotLet bullxbull Number of Ukely Unexpected Survival Cases and y Number of Ukely Expected Survival Cases

Determine Combined Likelihood Function lk (b Iadditional platforms) middot

bull REVISED LIKELIHOOD FUNCTION

lk(b I all cases) = lk(b I 9 platforms) middot x lk(b Ix platforms) x lk(b I y platforms)

bull ESTABLISH REVISED BIAS FACTOR B

f8 (b) u f8 (b) lk (b I all cases)

II

t t ~

shy

APPLICATIONS OF BIAS FACTORS

ii APPLICABLE TO PARTICULAR LOAD AND RESISTANCE RECIPE

bull USE BY THE API COMMITTEES AND MMS

Criteria Revisions

Recommended Design Guidelines Revisions bull

bullbull REASSESSMENT STUDIES AND DECISION MAKING

middot

bull RISK ASSESSMENT STUDIES BY THE OPERATORS

Individual Platforms

Fleet of Platforms

middot IMR Program Implementation

bullbull NOT TO BE USED IN DESIGN

I I fl J

EXAMPLE USE OF BIAS FACTORS

bull PHASE I EXAMPLE APPLICATION

bull EVALUATION OF ANNUAL PROBABILITY OF FAILURE

Establish Annual Maximum Seastate Data

Determine Coefficients to Define Base Shear

Determine Ultimate Capacity of Platform

Determine Probablllty of Failure Versus bullbbull

-bull DETERMINE UPDATED PROBABILITY OF FAILURE

Pf - P [S gt R]

Pf I P [~ gt Rb I b] f 8 (b) db

Pr - J P [S gt Rb I b] f bull 8 (b) db

bull TOTAL PROBABILITIES OF FAILURE

Total Pf - 1 bull Il [1 middotPf] - l Pf bull

Total Pf - 1 bull Il [1middot Pf] - l Pfmiddot

Total Pr = 1 - n [1middot PfbullJ = l Pfshy

I I I

EXAMPLE APPLICATION bull PHASE I

PLATFORM ST 151 K

bull GULF OF MEXICO

bull WATER DEPTH = 137 FT

bull ANNUAL MAXIMUM HS MEDIAN 175 ft COV 031

bull ASSUMED STORM DURATION 3 hoursmiddot

bull ASSUMED NUMBER OF WAVES 800

bull MEDIAN HS FOR STORM HOURS 1 2 AND 3

bull MEDIAN CURRENT VALUE bull

bull DETERMINE ULTIMATE CAPACITY FOR THREE DIRECTIONS

bull DETERMINE ANNUAL FAILURE PROBABILITIES

Broadsidemiddot Diagonal End-on

0037 0025 0007

0071 0044 0018

0022 0016 0004

DETERMINE TOTAL ANNUAL PROBABILITY OF FAILURE

Pf - 007

Pf - 013

Pr - 004

bull

EXAMPLE USE OF BIAS FACTORS - PHASE II middot

bull DETERMINE PREDICTIVE PROBABILITY OF FAILURE

Case - 1

Pf _ S S Pf (b b2) fa1a2 (b b2) db db2

g bull -1 (b1 R I S) - 1

92 - (b2 R2 I S) - 1

Case -2

Pf s ~ [bJ f11 (b) db

Pf [bJ p [g lt 0)

g - bull 1 (b1 R1 I S) - 1

1

bull1) l

I

BENEFITS OF ANDREW JIP bull PHASE II

lt

bull IMPROVED LOAD AND RESISTANCE RECIPE

bull IMPROVED CALIBRATION PROCEDURE

bull MULTIPLE BIAS FACTORS

bull PROCEDURE TO INCLUDE EFFECT OF OTHER PLATFORMS

bull PROCEDURE TO USE MULTIPLE BIAS FACTORS

  • Structure Bookmarks
    • ANDREW PHASE II JOINT1 INDUSTRY PROJECT
    • PROGRESS MEETING NUMBER 1
    • ANDREW PHASE II JOINT INDUSTRY PROJECT PROGRESS MEETING NUMBER 1 NOVEMBER 16 AGENDA
    • ANDREW PHASE II UST OF PARTICIPANTS
    • ANDREW PHASE I OBJECTIVES AND RESULTS
    • ANDREW PHASE II PROJECT OBJECTIVES
    • SUBSEQUENT PROJECTS
    • ANDREW PHASE II PROJECT TEAM
    • ANDREW PHASE II UPDATE MODELING middotAND ANALYSIS RECIPE
    • -ANDREW PHASE II CONDUCTOR RESISTANCE
    • ANDREW PHASE Il WA VE FORCE PROFILE
    • ANDREW PHASE II WA VE FORCE PROFILEmiddot
    • ANDREW PHASE II BRACE MODELING
    • ANDREW PHASE II FOUNDATIONS
    • ANDREW PHASE II middot middot JOINT MODELING
    • ANDREW PHASE Il JOINT MODELING OPTIONS DISCUSSION
    • ANDREW PHASE II JOINT MODEJJNG OPTIONS ILLUSTRATIONS
    • ANDREW PHASE II JOINT MODELING PHASE II METHOD
    • ANDREW PHASE II ANDREW STORM CONDmONS UPDATED
    • Platform Selection bullbullbull1111111111111
    • BasisCriteria for Selection bullbullbull1111111111111
    • CALIBRATION
    • WHERE TO INTRODUCE BIAS FACTOR (B
    • PHASE I -KEY CONCLUSIONS
    • bull CALIBRATION ISSUES PHASE II
    • PHASE II WORK SCOPE (CALIBRATION)
    • SOFTWARE
    • SENSITIVITY STUDIES POSTERIOR DISTRIBUTION OF B
    • WEIGHTING APPROACH bull
    • INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON
    • CATEGORIZATION OF ADDITIONAL PLATFORMS
    • UPDATING OF CALIBRATION RESULTS
    • APPLICATIONS OF BIAS FACTORS
    • EXAMPLE USE OF BIAS FACTORS
    • EXAMPLE APPLICATION bull PHASE I PLATFORM ST 151 K
    • EXAMPLE USE OF BIAS FACTORS -PHASE II middot
    • BENEFITS OF ANDREW JIP bull PHASE II
Page 11:  · . ' ANDREW PHASE II PROJECT OBJECTIVES . ' • ·UPDATE MODELING AND ANALYSIS RECIPE· JOINTS ' ' ·BRACES WAVE PROFILE FOUNDATIONS . ' ' CONDUCTORS· • ANALVZE PLATFORMS USING

~-----------____________________

Project STlSlK Model dir_2 version 7 Mon Nov 14 124200 1994

177 s

Ill

74

37

0- - e -zo - ~o

- Jo

-100

-10

-ISO

z

UV 1-+x

Project STlSlX Model Conductor Version 1 Mon Nov 14 140620 1994

bullmiddot

I

I

I I

I

I I

I I I I I

1 ==

-L

-L - L

-

- L

-

-

L

-

Uniform Displacement step 6 Inelastic Events Legend

Elastic

Strut Buckling Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield Beam Clmn FUlly Plastic Fracture

--

i~~~~~~~~~~~---~--~~-J___--JJ~_middot~~~=-

Project ST151K Model Conductor Version1 Mon Nov 14 142735 1994

---

~

I

I

middot

I

I

I

- shy- shy

- shy

- shy

4x z

CIV Uniform Displacement step 33 Inelastic Events Legend

bull Elasti~ Strut Buckling Strut Residual Strut Reloading Plastic StrutNLTruss Beam Clmn Initial Yield Beam Clmn-Fully Plastic Fracture

gt bull

Project STlSlK middot Model Conductor Version 2 Mon Nov 14 142025 1994

-----------------

bullf

I

I

I

II I I

I I I I

I ==

-

Variable Displacement step 9 Inelastic Events Legend

Elastic Strut Residual

bull

Strut Buckling Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield Beam Clmn Fully Plastic Fracture

--

Project ST151K Model Conductor Version 2 Mon Nov 14 142145 1994

I I I I

I I

I

I

I - - -

-Ishy-

-

Variable Displacement step 35 InelasticEvents Legend

Elastic Strut Buckling Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield Beam Clmn Fully Plastic Fracture

Project ST151K Model Conductormiddot

Version 1

Clgt - Cut Plane Force Fx Mon Nov 14 140051 1994

- -_ vii I fOIZM

0 _

~

N ID 0 bull 0

x 0

middotrt ~ () bull u 0 ~ 0 Ir

N bull

0

0 1 2 3 4 5 X Deck Disp [Ft)

I I 1 1 WbullTI- bullbull

lltcxi lr-C

011lt~~~~~~~---~~~~~~~~L~~~~~~~--~~~~~~~~i~~~~~~~--

Uniform Node Displacement

shy

Ji

r

Comparison of Pushover Curves for ST 151K Variable Load Pattern and Monotonic Load Pattern (Andrew)

Soll Improved 100

6000

5000 -i 4000

f

llS

3000 Ul

J 2000

10 c11 c1lgtolts 1000

0 v

bull ~

f

_ ----- --- ---shy

bull --Variable Lold

- - Andrew Load

I I I I I I

4 5 62 30 1

Deck Dlsplacement (ft)

bullbull r 1 II

~

ANDREW PHASE Il

WA VE FORCE PROFILE

bull DECK WA VE FORCE ~

PHASE I USED SECTION 17 (1056H degCd)

PHASE Il USE REVISED SECTION 17 (1000H Cd)

bull PHASE I USED SINGLE FORCE PROFILE FOR PUSHOVER

- USED WAVE HEIGHT CLOSE TO ULTIMATE CAPACITY

bull PHASE J

Il WILL STUDY THE

USE OF VARIABLE

PROFILE ~

- IF FORCE LEVEL INCREASES (OR DECREASES) BY A FACTOR EQUIVALENT TO A 2 FOOT CHANGE IN WA VE HEIGHT THE INCREMENTAL FO~CE PA1IERN IS CHANGED ACCORDINGLY (CAP FEATURE)

- INITIAL EXAMPLE

f~

a

middotn

a

I N -a a a amiddot

+

~ a

l

a w gt 0c 0 gt a I shyc -CJ wc w gtc(

== c( I shyZ w E W a 0 z- 8

ii sect

~ ~ bullT 0

I if

x ~

bullbullbull

II ~------------________

ANDREW PHASE II

WA VE FORCE PROFILEmiddot

CAP AUTOMATIC APPLICATION FEATURE

STAT

fR1 0 lt f lt1

STAT

STAT

f(R3-R2) middot f lt1

Project ST151K Model Wave_Loading Version 1 rue Nov 8 164107 1994 PT

E~ nss FT

SL 14amp FT

a 111 FT

EL 14 li

E-1 37 FT

El o FT

1middot1middot ______________ _ ________ __ _ __

0

200 400 600 800 1000 12000 Force at Elevatloii (kip)

200

a ll_

Wave Force at Major Elevations for Discrete Wave Heights

180

bullHmiddot40ft

Hmiddot495ft

H bull 51 ft

Hmiddot55ft_

H-59ft I

Andrew (H bull 60 86 ft) II

i I

_11---__

- middot- Hmiddot30ft

-

- bull -

middot bull bull bull middot

---Hmiddot53ft

bullbullbullbullHmiddot57ft

-----H-63ft

Elevation Above Mudllne 100

(ft)

180

140

120

80

60

40

20

Loading at Major Elevations

---H-49511

-deg-Hmiddot608611

0 200 400 600 800 1000 1200

Force It Elevatlon (kip)

--

Base Shear vs Wave Height ST151K

6000

5000 1---------Ult1mateCapacitybull 4890 kip (100 Improved Soll)-----------~--

4000a a lii ~ 3000

81 ampI

2000

1000

0 +-~~~--i~~~~~~~~~-+-~~~--lf--~~~~~~~~-+-~~~--l

30 35 40 45 50 55 60 65

Wave Height (ft)

bull middot

~

bull

II

2

g

~

It)

0

cCD E 0 E a c E euro~ CDshygtItgtmiddoto ~ti c -a Gic CD gt ~

-= c -Cll

I z

bullO 0i i ~ g 0

() CI -bull(q-ll) iuawow llu1wniiaAO

11

shy

116

114

112

g 110

bullc = 108

~ ~ 106 -sect 104

102

100

98

96

-

6555 60

Wave Height vs Center of Action ST151K

30 35 40 45 50

Wave Height (ft)

t

Comparison of Pushover Curves for ST 151K Variable Load Pattern and Monotonic Load Pattern (Andrew)

Soll Improved 100

--- --------- --- ----6000

5000

- 4000 8 - I

1 3000 Ul

81 m 2000

_J -shy

1000

---Variable Load

- - Andrew Load

0 f-~~~~~r--~~~~-t-~~~~~-t-~~~~~+-~~~~----t~~~~~-1

0 1 - 2 3 4 5 6

Deck Displacement (ft)

-middotshy

1 t I I ~

ANDREW PHASE II

BRACE MODELING

bull MARSHALL STRUTS USED FORPHASE I

bull PHASE II WILL ASSESS DIFFERENCE IN USING BUCKLING BEAMS AND MARSHALL STRUTS

middot bull COMPARE FOR 2D FRAME

bull COMPARE FOR PLATFORM 177 WHERE LOWER

BAY EXHIBITS LARGE BRACE MOMENTS

Project andrew Model beamframe Version 1 Mon Nov 14 145447 1994

~

middot Frame Test buckling beam-column model

-

Project andrew Model strutframe Version 1 Mon Nov 14 145754 1994

bull

Frame Test strut model

middot-middot _____________--_

_middotmiddotr_~ --ll--~-middot__

Buckling Beam bull

Strut (K=O 65)

Lateral Displacement

ANDREW PHASE II

FOUNDATIONS

bull ~LICIT NONLINEAR API FOUNDATION USED INiPHASE I

bull FOUNDATIONS

JIP SELECTED MINI-VANE DATA FOR

STRENGTH PROFILE

bull COMMENTS ON BENCHMARK COMPARISONS

- SIGNIFICANT VARIABILITY DUE TO FOUNDATION MODELING METHODS

II Ii

z

ctV 4x SPlO caissonfy-42s soil-api(lf-097step-114)

Inelastic Events Legend Elastic Strut Buckling

Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield

--------shyBeam Clmn Fully Plastic Fracture

~--~~~~~~~~~--~----~__i~~---~~~~__--~

Pro)ect apipile Model South Pelto 10 Version Wed May 18072235 1994 - -

1bullt

+

t

f i bullbullbull

bullbull bullbull

I I

5

~x SPlO Caissonfy=42s soil=api(lf=lOOstep=137)

Inelastic Events Legend

Elastic Strut Bucklilg

Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield

earn Clmn Full~ Plas1c Fracture

ANDREW PHASE II middot middot

JOINT MODELING

middotbull API RIGID-PLASTIC USED IN PHASE I

a ELASTIC-BRITTLE USED IN PHASE IB

amiddot MSL WILL PROVIDE NONLINEAR FORCE-DEFORMATION

AND MOMENT-DEFORMATION RELATIONS FOR PHASE II

a MODELING OPTIONS FOR PHASE II -

l

- UNCO~LEDmiddotONE DIMENSIONAL MEMBER END JOINT MODELING

J Ii

ANDREW PHASE Il

JOINT MODELING OPTIONS DISCUSSION

bull JD COUPLED INTERACTION SURFACE

- IDEAL

- DATA NOT AVAILABLE TO CALIBRATE

bull 2D COUPLED INTERACTION SURFACE

- SIGNIFICANT MODELING IMPROVEMENT

- DATAAVAILABLE

- IMPLEMENTATION PRACTICAL BUT NEEDS DEVELOPMENT AND CALIBRATION

bull 2D UNCOUPLED INTERACTION SURFACE

- SIGNIFICANT MODELING IMPROVEMENT

- DATA AVAILABLE

- IMPLEMENTATION PRACTICAL AND AVAILABLE

ANDREW PHASE II

JOINT MODEJJNG OPTIONS ILLUSTRATIONS

P

e

ANDREW PHASE II

JOINT MODELING PHASE II METHOD

e

I I I

ANDREW PHASE II

ANDREW STORM CONDmONS UPDATED

bull -PHASE I USED OCEANWEATHER HINDCAST-(Nov 1992)

bull middotmiddotPHASE II WILL USE THE REVISED HINDCAST (Nov 1994)

bull lt c lt I lt

2-2

l---4-~~~+-~~-+-~~~1--~~-+-~~--l--~T--+-~~--i24

f-tJVtMfiz~ 1Cfl (r STvb(

Plollol bullbull 5middotNOYmiddot9214bull1J Irbullbull Illbull [ANDREVl~NDREVCUSTbullMAXll 5-NOY-1992 10bull54 - 9208

HEIGHT lml

I~ MAXI

I L 1~middotamp SZ~- ~A I 13c

I J) ( r I X~0iWJ~0gtJ gt L gt gt I 1c I 120

-- N

~ deg

I I I I I I e C I ~=-41 122 -94 -92 -90 -88 -86 -84 -82 -80

-

Figure 7 contours of maximum hindcast significant wave height

_

~

~tl ~

NOltNttIJ 17lt ltNS S iJJ)y

Maximum Significant Wave Height - Hurricane Andrew 1992

30 ~

26 ~ I ~-~~~- == -shy -~ 35-_

24deg1--shy --------~- middot-middot-------middotmiddotmiddot----middot------middot-middot~middot- - - ~-----1---middot

22middot ~~~~~~~-~ -94 -92 -90 -88 -84deg-86 -82 -somiddot

c

-

Platform Selection bullbullbull1111111111111

0 BasisCriteria for Selection middot

0 Review of Phase I Platforms middot 0 Alternative Platforms

0 PMB Recommendations

0 Participants Comments

gt

L

l I

BasisCriteria for Selection bullbullbull1111111111111

0 Benefit for Calibration - Classification (UnexpectedExpected)

- Failure Mode (StructuresFoundations)

- Damaged StructuresKnown response

- Better Representation of Full ~opulation

0 Availability of Data - Existing Models

- Inspection Data

- Damage Data

PREDICTED BEHAVIOR

-

OBSERVED

BEHAVIOR

-SURVIVED DAMAGED FAILED

SURVIVED

ST134W WD90A

-MC311 -

MC397 -

--

-

-

ST151K ST130Q

-

-

-

DAMAGED - -

-

-

--

-

-

ST52 SS139 ST177B -

ST161A

-_

-

FAILED

-

-

-

ST151H ST130A ST72

Operator - Platfonns - Platfonns Survived Damagedamp

- Falled -

bull

-

-

Jle eBDIGIPllDbl -Amocoshy 30 1 Chevron -

143 12 Exxon 72 Mobll 22 1 Phllllps 7 bull

-Shell 39 Unocal 25 1

Total Partlclnts Platfonns 338 15

Other Operators bull- 310 13

-

Total Platfonns - -

648 28

fbml

Platforms Considered 6 7

tie of Total Platfonns 1 25

OVERVIEW OF PLATFORMS SURVIVEDDAMAGEDFAILED

-

shy

v

Platforms Evaluated in Phase I

-

-Platfonn Water Year Number Analysis Avallabillty of Availability Platform

- Name Depth rl of Model DetaDed rl ClassUlcatlon

Installation legs Inspection Soll Report Based on Information Phase I Calibration - -ft

- -- Suaill flalfnrm ~BSi -STISIK 137 1963 8 CAP Yes Yes Un eel Sarvlval

-STIJOO 170 1964 4 CAPmiddotshy Yes ST134data u ST134W 137 1981 4 CAP Yes Yesmiddot E

-

edSarvlval -

middotsurvival

WD90A 184 - 1964 8 CAP Yes E -

- Sarvlvill

MC3ll 343 1978 8 KARMA

MC397 468 1991 4 KARMA -

bull ~reglmmiddot- -

Yes Sure Survival

bull Yes - Sare Sanlval-

-T23STS2 63 1969 4 CAP

T2SSS139 62 1969 4 CAP -

Yes

Yes

Yes Ex

Yes Une

bull SurvivalLikely to DamalI

middot bull SanlvalLlkely to DamalI

ST161A 118 1964 8 USFOS

Eail11a pound1al(bullllm Clss ST177B 142 --1965 8 CAP

STISIH 137 1964 - 8 CAP C

STIJOA 140 19S8 8 CAP - -

Yes

Yes

ollamed

Yes

- Yes E

-

at about 1 mUe

Yes

ST134data

bull SafvlvalLlkelv to Dam- bull- -

E middotrahre

Emeried Failure -

Likely FaUare

T21ST72 61 1969 4 CAP Collamed No Likely FaUure

Platforms Identified for Evaluating Their Likelihood of Falling Under UnexpeCted Survival Category

CllEllATOa

AllB

llLOCll

Sl1lvcnJllE

NAME

YEAR

INSTAUJCD

WATER

IJBPJB

middotmiddotDISTANCE

ftOMSllOllE

(MIUIS)

Ill

(M)

AMOCO BllW

STllI

WDI075

A

bull D

4

n uo 17J

44

32

bull 20

10

bull CllEVllOH STtllO c I 1bull 21 11

STl130 D 0 161 21 11 STllJ4 I 0 137 D - 11

STIUI G I 137 32 11

STIUI I 0 121 32 11

STIUI

J 0 140 32 11

STIUI It 4 137 32 11

STllSI PROl)I 0 137 32 11

STllSI PROD-2 0 137 32 11

STll7 A 0

140 35 11

WDl117 c 65 214 34 10

DXON STshy E 51 14 bull STl165 A t3 bull 34 WD0073

WD0073

WD0073

A _ c

4

62

4

IA IA

17J

22

22

22

bullbullbull WD003-shy E 65 161 Z7

A 0 IN 31 MOBD Bl12I

ssn A bull

50

50

31

7

bull 7

MIJllPllY SS114 B 7 50 7 7

PlllUIPS SSl14t A 55 33 7

SAMEDAN ST17J A 4 117 7 bull 7

SJIELL WD0103 A 65 223 27 WDllOJ bull 65 221 27 WDll04 c 65 221 27 WDOI04 D 67 2 27

UNOCAL

SSl20I

SSl20I

bull SSl20I

SS0209

SS0209

SS0209

SS0209

SS 0253

SS 0253

E

I

PtJllE

A

bull D

G A

c

0

4

4

7J

61

0

0

69

103

97

97

9S

100

95

100

165

175

34

34

34

34

34

34

34

54

54

bullbullbullbullbullbullbullbullbull

Platforms Recommended for Phase II

llalr y_ w- Aot Av- _ I 1 Doplla Model nm bull bull _ bull I I H11 s18-1 Irht lbwl~ bullbullbull(-illtdlel OM lllMtD

II ---

-STISIIC 137 1963 I CAP Yeo Yeo I Yeo

STtshy -170 1964 4 CAP Yeo ST134dlU I Yeo

STl34W 137 Yeo Yeo -1981 4 CAP D Yeo

WD90A 114 1964 I CAP Yeo - D

MC311 30 1971 I KARMA Yeo -sre-

MC397 middot 461 1991 4 ICARMA Yeo Sireshy - - r- -

__ T13ST5l 63 1969 4 CAP Yeo Yeobull middot- ~n- D Yeo _ Tl5SSl39 62 1969 4 CAP Yeo Yeo I Yeo

_ STIT711 142 1965 I CAP Yeo I Yeo middot-1shySTl61A Ill 1964 I USFOS Yeo Yeo - D bulloshy

- r-

STISIH 137 1964 I CAP y D y -_ STl31A 140 1951 I CAP Yeo STl34dlU I Yeo

T21Sll 61 1969 4 CAP No rar-middot I Yeo

~-rmiddot-middot SS209A 95 1967 16 CAP Yeo D -SSU4R

11 UI A 41 -in D SS 2741AIM CAP bull

_ _ -shy - -shy- -shy

CALIBRATION

bull PHASEIAPPROACH

bull CALIBRATION ISSUES

bull PHASE II WORK SCOPE

bull SOFlWARE AVAILABILITY

bull EFFECT OF ADDITIONAL PLATFORMS

bull WEIGHTING APPROACH

bull EXAMPLE APPLICATION middot

(l=Ofce Mode] Wind

~

fcurrent l -middot

~SW ~leldl4

ultimate resistance force I first

design load Illgt

i _

deflection

-

bull

4

Historical Hurrlcane Dariiage and Survival of Platforms

Components Required for Calibration (Petrauskas 1992)

PRIOR Distribution

PLATFORM EXPERIENCE

BAYESIAN Survivals and Failures

UPDATING

Waves Winds and Currents bull

POSTERIOR Distribution

- A_middotJA ~ Survival Data Failure Data

Result Result

Load Resistance

middotfl

-

Bayesian Updaing middotGeneral Description (Petrauskas 1992)

L Imiddot

WHERE TO INTRODUCE BIAS FACTOR (B

On Resistance (Rmue - B (RPREDICreD I

middotOn Loads (Smue middot - B (SPRE01cTEo

On Safety Margin (RSmue - B (RSPREDICTED

CALIBRATION METHODOLOGY

gt

r

FIELD OBSERVATIONS

Observed Behavior

survivals Damages

Failures

r

bull

CAPACITY ANALYSIS

Predicted Load and Resistance Predicted Failure Modes

Survival Damage Failure cases

RELIABILITY ANALYSIS

- Probability of Failure ~

Ukellhood Failure

BAYESIAN UPDATING

Prior Distnbution of B Combined Likelihood Function

Posterior Dlstnbution of B

BIAS FACTOR

Mean Value cov bull

Represents Modeling Uncertainties

Prior Distribution of B Likelihood Functionmiddot Observing Success case STISIK and Posterior Distribution of B

2 00

uo

160

140

120

e 100 L $ uoIshy

0 60 ~ ~ 040

020

0 00

0 00 0 50

_ I 00

b

I 50 2 00

-shy lk [bl STISIK)

--shy Pnor-f(b)

--shy Pos1enor r(bl ISIK)

Prior or B Mean bull 10 COV bull 030

Posterlor of Bbull Meanbull 136COV 0 16

2 50

Likelihood Function and Posterior Distribution Example Platform STISIK

- -

~

- Prior and Posterior Distributions or B Failure cases (ST177B STlSlH ST130A T21) Damage cases (T23 T25 ST161A) Success cases (STlSlK ST130Q ST134W WD90A MC311 MC397)

400

350

300

2 so 0-

I

bull bull

bull bull gt

- bullbull ~ bull bull I bull

_ bull bull bullbull I bull bull bullII V -

bull bull

I bull ~ bull

bullbull bullI bull bullI bull bull bull

middot -~

bull__ ~ middotI middotmiddot

L 2 00

e ~ I SO

-I 00

- 0 so

000 gt

0 00 020 0 40 060 0 80 I 00 120 140 160 110 200

b -

-

shy

Prior-r

- - - - Posterior fLlll failure cases

bull bull bull bull bull bull bull Posterior flall damage cases

- bull bull - bull middot Postenor flall success cases

Posterior flall success fulure damage cases

-

Posterior of B Mean= 119 COV 0101

Posterior Distribution or Blas Factor (8) bull Different Categories and All Cases Combined

shy

l

PHASE I - KEY CONCLUSIONS

bull GLOBAL CORRECTION FACTOR

bull 15 TO 20 CONSERVATISM MEAN)

bull UNEXPECTED SURVIVALS HAVE GREATEST IMPACT

bull DAMAGED CASES MORE INFORMATIVE

bull CALIBRATION ISSUES PHASE II

bull

EFFECT OF IMPROVED CAPACITY ANALYSIS RECIPE

bull APPLICABILITY TO MULTIPLE FAILURE MODES

bull NO FOUNDATION DAMAGE CASE INCLUDED

bull IMPROVED REPRESENTATION OF FULL ANDREW DISTRIBUTION

-bull NUMBER OF PLATFORMS CONSIDERED

I 1 I

I t ) bull ~

PHASE II WORK SCOPE (CALIBRATION)

bull

bull ESTABLISH MULTIPLE BIAS FACTORS

Use Nine Base Case Platforms

middot - Capacity Analysis for Revised recipe

Foundation Blas Factor from Foundation JIP

bull DEVELOP WEIGHTING APPROACH

bull UPDATE MULTIPLE BIAS FACTORS

bullEffect of Platforms Not In the Nine Base Cases

bull EXAMPLE APPLICATION OF BIAS FACTORS I

bullbull

PHASE I PHASE II

CAPACITYbull ANALYSISRu bull

---------------------1----middot

Jc1 - iJbullraquoJ dbull

FttlS frt[f~l ratdhJJ

-

LIKELIHOOD FUNCTION

DEVELOPMENT

lk Jb Jfall1N) bull Jb)

111 Ill Isurvtvoq bull 1 - bl

-middot -------------------shy ~---------------------middot-middot

r111ir8 Cb1 111 Cbl BAYESIAN

UPDATING

~cove

R1R2 bull middot---------------------~----

Jblbull P(g cOJ

1 bull JbRI (MBSI I bull 1

~middot

llcJllJfalluro) bull P(gcOJ Ill 111 b21fallun) bull P ((g1 CO) U 1112 c OD

middot--------------------shy ----------------------shy bull

r111111r8 Cbl 111 Ill i-1nro1

deg ~1b2)laquottfrr (b1b21-lnfo) re Ill bull 1-11middot b2) bull lodegamp

middot

i1 covamp i2 cov~i

-

shy

--

Likelihood Function vs b All Success Case STISI K STl30Q STl34W WD90A MC311 MC397

I 0000

0 9000

OIOOO

0 7000

~ 0amp000

ta osoOO

~ 0 4000

0 3000

0 2000

0 1000

bull 0 0000

bull

I

middot

middot bull I

Ibull

-sibull ~ -middot-middot-middot-middot -middot-

----7----+-----~ I

Ibull bull bull I

rbull

--- Jk ISIK[blsuccess)

- bull - bull - lk 130Q[blsuccess) bull

lk 134W[blsuccess)

Jk-WD90A(blsuccc~I

- - - lk MC397[blsucccss)

--ltgt-- lk middotMC31 l[blsucccss)

--- Jk [bl 6 out of6succcsHascs)

05 I 5 2 250

b

Likelihood Functionsmiddot Success Cases - Phbullsc L

shy

bullbull bull bull bull bull bull -------------------~

bull bull bullbull bullbull bullbull bullbull bullbull bullbull bull

bull bullbull

bullbullbullbull

bullbull

bullbullbullbullbullbullbull

bullbullbullbullbullbull

bullbullbullbullbullbullbull

z 0

t z ) LL

c 0 0c i ~ i

J

~ sect ~--------------~ e l5

I

I

SOFTWARE

bull PF PROGRAM

bull CALREL

bull PF +CALREL

bull RELACS

bull

gt i

l

SENSITIVITY STUDIES

POSTERIOR DISTRIBUTION OF B

bull EFFECT OF CHANGE IN DISTRIBUTION OF PRIOR OF B

bull EFFECT OF REMOVAL OF ONE PLATFORM FROM EACH CATEGORY

bull EFFECT OF REMOVAL OF THREE DAMAGE CASES

bull EFFECT OF ADDITIONAL 13 SIMILAR PLATFORMS

-I

Prior and Posterior Distributions of B Variation In COV of PriorofB fromOl to04 -- -

4500

-4000

-

- -

3500 --middot - bull bull - bull bull Prior-r ICOV =02

3000

- - - - Prior-r ICOV =03 -

bull -~ e 2soo -

bull bull bull bull bull bull bull Pnor-r ICOV =o 4t frJ ~ 2000 Posterior rlCOV =0 2

bull bull Postenor rlCOV =0 3 bull bull 1500 bull middot bull - shy -middot- Posterior rlCOV =04

middot~ I 000 - bull

-middotmiddot ~ ~middotmiddot M -~

- middot 0 500

wbullbullbull _ I middot~ middotmiddotmiddotmiddotmiddot middotmiddotmiddotmiddotmiddot - -~ ~ __ middot_ _ bull -shy0 000

0 000 0200 0 400 0 600 0 800 IOOO 1200 1400 1600 1800 2000 -

- - b

-

=shy

Posterior Distribution of Bias Factor (B) bull Effect of Variation In COV of Prior of B

bull

bullPrior and Posterior Distributions of B Evaluating Effect of Removal of One Platform from Each Category from the Base Case of 13 Platforms

400

~ ~ 350

j IA i

3 00 bull I bull I ---Prior-r I bull l

250 1 I 1

I bull bull ---Posterior r lall 13 casesshy- ~E 2 00 bull bull bull bull bull bull bull Posterior r without STISIK

- ~ 1 bull - bull I - bull bull - bull bull Posterior r Iwithout T2I I 50 I

J-tt~i - bull - bull - Posterior r Iwithout T2S

tOO [-

I I ~ ~ I 0 50

0 00

___ r J i

l~T

~--

0 00 0 20 0 40 0 60 080 I 00 120 140 160 110 2 00

b

zshy

Posterior Distribution of Bias Factor (8) bull Effect of Removal of middotPlatform from Any Category

bullPrior and Posterior Distributions or B Evaluating Effect or Removal or 3 Damage Cases from the Base Case or 13 Platrorms

4

35

3

25

e L 2 ~ bulle shy

I 15

bull I

05

0

0 02 04 06 I 2 I 4 16 IS 2

I I

I

I

I Ibull

J -tmiddot L--

bullI

~

I

fl

II ~

I I v I

middot bull bull

bull L

~ ll- I-~lt

OS I

b

I

bull

1middot bullbullbullbull

~ __

bull

- - - - Prior-r

- - - - Postenor flall failure cases

Postenor flall success cases

--- Posterior flall 13 cases

---Posterior fI success failure cases only

IPosterior or B Mean =t11 COV =0121

-

_I

Posterir Distribution or Bias Factor (B) bull Effect or Removal or 3-Damage Cases

-

bullPrior and Posterior Distributions or B Evaluating Effect or Additional 13 Similar Observations as

Used in Andrew JIP The POsterior or B from Andrew-JIP Considered as New Prior or B

- 500

450

400

350

bull bull

bull bull bull bull J bull

bull bull f ~ I

I ~ Ibull bull

bullbull bull

l bull I -bullbull bull bull

I

I bull bull

bull I I~~bull bull bull

bull ~

I bull bull

~ ~_ ~~~

- - New Prior r bull

Posterior 114 new failure cases ~ 300- bull bull bull bull bull bull bull Posterior f 13 new damage casesl 150 ~

- bull bull - bull bull Posterior 116 new success cases i 2 00

Postenor 1113 new success failure -150 damage cases

I 00

0 50 -

0 00 0 00 0 20 0 40 0 60 0 80 I 00 120 I 40 160 110 2 00

Posterior ol B Mean= 120 COV= 0071 b

shy

shy

_rshy

Posterior Distribution of Bias Factor (B) bull Effect of Additional 13-Similar Observations

1

1 I

WEIGHTING APPROACH bull

bull ASSESSMENT OF PLATFORMS DATA

Select Platforms Most Mappable to the Analyzed Platforms

Unlikely SurvivalDamage Cases Identification

Based on Cut-Off Criteria Agemiddot Hs etc

bull INFORMATION REQUIRED FOR COMPARISON

bull CATEGORIZATION OF ADDITIONALmiddot PLATFORMS

bull UPDATING OF CALIBRATION RESULTS

--

OVERVIEW OF PLATFORMS SURVIVEDDAMAGED -~

Operator -

- Platforms

Survived CS)

Platforms

Damaged CD)

Platforms Used In

Phase IJIP -

JI~ ~Bill~IRBDll -

Amoco -

Chevron -

Exxon

Mob II

Phllllps

Shell

Unocal -

30

143

72 22

7

39 25shy -

1

12

bull

1 bull

1

1(S)+1 (D)

3 (S) +3(D)

2 (S)

bull shy

bull

bull

bull

Total Partlclnants Platforms 338 - 15 10C6S+4D) -

Other Operators

310

13

3 (D)

Total Platforms 648 28 13C6S+7D) -

Summary of Andrew JIP bull Phase I Platforms

Plairshy w v- N-ber Lea-le 1 shy car Hind_ o-w N-bull Deplh fl fl IJpe la bull n Yeadon Jadtet lbelledtf liefalit -- _ -OD

IL H - -shy _ICllshy IL IL - - rQClto ~ _

K STUIK 137 1963 8 365 608S4 Cellardeck

-middot STl-shy 170 1964 4 -~ K 39 62269 Cellardeck - --ST134W 137 1911 4 43 608S4 Smvlnlshy- _ - -

WD90A 184 1964 8 K 267Sbull34 83 SOS31 SUboftllar deck SmvMI

MC311 343 1978 8 x S7 62 - MC397 468 1991 4 x shy S33bull6SO 65911 -shySmeSmvlvol -- gt -

r-- T23ST52 63 1969 4 ton- shyKshy 52594 S03 -T25SSl39 62 1969 4 Kshy SOS94 S063 - middot ton-shy__ -

bullbull a ton--shyST161A 118 1964 8 K - 347S 51495 Cellarclecti

r~

ST1771 142 1965 8 K 39 60167 CeDordeck failme

STISIH 137 1964 8 K 395 608S4 CeDardeck Failme

K ST130A 140 1958 a SS 6095 I lblv Pai1uno - T21ST72 61 1969 4 Kshy S4S94 4974 I lhlv Pai1uno -

- -

JE

INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON

bull WATER DEPTH

bull bull YEAR OF INSTALLATIONDESIGN

bull PLATFORM ORIENTATION

bull PLATFORM LATITUDE AND LONGITUDE

bull KEY SKETCHES OF PLATFORM STRUCTURAL FRAMING

Number of Legs Plies Leg Batter Number of Conductors bull Deck Elevations (Including sub-cellar deck If any) Jacket Framing Elevations Brace type In Vertical Frames Sizes of Primary Members (leg pile vertical frames braces primary horizontals) Joint Details (can gap etc) LegPiie Annulus Grouting Status

bull middotPOST ANDREW INSPECTION REPORT Significant Wave Height Hs (ft)

bull INFORMATION ON DAMAGES PRIOR TO ANDREW

bull RESULTS FROM IN-HOUSE STUDIES IF AVAILABLE

bull ~RIGINAL DESIGN CRITERIA INFORMATION IF AVAILABLE

bull SOIL SHEAR STRENGTH PROFILE IF AVAILABLE

Platforms Identified for Evaluating Therr Likelihood of Falling Under Unexpected Survival Category

OPlRATOR

middotADA

BLOClt

llmucnJllE

NAME

YEAll

INSTAUJD

WA1ER

DBPTll

DISTANCE

JlllOM SHOllE

(MILU)

Bl

(M)

AMOCO 111113

STl11

WDI075

A

bull D

St

17

120

1n

44

20

10 I

CllllVllQfl STlllO c 61 110 21 11 STIUO D Q 1110 21 11

STl134 p Q 137 29 11

ST0151 G 61 137 11

ST0151 I 3 121 11

ST0151 J Q 140 11

STl151 It 137 11

ST0151 PJl001 Q UT 11

ST0151 PROD-2 Q UT 11

ST017 A 3 140 35 11

WDIUT c 65 214 34 10

llXXON STI055 II 51 6T 14 I

STl15 A t3 34

WDI073

WD0073

Amiddotshy Q

1111 1111

22

22

I

I

WDI073 c 1n 22 I

- WDOCl93-shy II

65 1110 2T A 3 194 31 MOBU 111121

SS IOl7

A B

501

50

50

31

T

I

T 6

MtlllPllY SS114 B T 50 T T

PlllllIPS SS0149 A 5 55 33 T

SAMllDAN STln A 107 T T

SBllLL WDl103 A 65 223 2T WDl103 B 65 221 2T WD0104 c 65 221 2T WD0104 D 6T 20 2T

UNOCAL SS020I

SS 020I SSl20I

SS 0209

SS0209

II p

PURll

A

B

3

n 61

103

97

97

95

100

34

34

34

34 34

bullbull bullbullbull

- SS0209 D u 95 34 I

- SS 1209 SSl253

SS0253

G A

c

6T

2

100

165

1T5

34

54

54

bullbull I

I I

I I

SOUTH MARCH

~LAND

N

SOUTH PaTOI ENE ISWltD AREA

MMS Defined Limits ror Offshore Structures Requiring Inspections Following Andrew

t---Jbulln3 T25bull I bullTtl

bull1$1Htl1Abull bull1SIK1779bull

~

middot-DaTA AREA~T

CRANO rsLE

AREA bull1300

bullll4W

=shy

bullPath of Hurricane Andrew with Platforms Used In Calibration

Likelihood Function vs b All Success Case STISI K STIJOQ ST134W WD90A MCJll MCJ97

I 0000

0 9000

0 8000

07000

-Bo6000

tiiii 0 5000 -

04000

0 3000

0 2000

01000

0 0000

0 05 25

ri1rt middot-middot-middot-middot-middot-

I 5 2

b

---11 middot151K(blsu~ccsI

- bull - bull - fk 130Q(bfU~lCSS)

lie bull134W(blsuccc]

lk-WD90A[blsultCS]

- - - lk middotMC397(bfUtlCS]

--o-- lk-MCJll(bbuctcsJ

--- ll (bl 6 out of 6 Mlttcs case]

_

( Likelihood Functions middotSuccess Cases - p - 1 c I

CATEGORIZATION OF ADDITIONAL PLATFORMS

I I I

bull CATEGORIES

Unexpected Survivals

Expected Survivals

bull APPROACH

Identify Corresponding Platform(s) Evaluated In the JIP

Determine Max Wave Height at Platform Location

Evaluate Whether Wave Hit the Deck

middot - Evaluate Likelihood of R and S to Differ from the Base Cases

Review Joint Information

Assess Special Features

bull IDENTIFY NUMBER OF PLATFORMS IN EACH CATEGORY

I

UPDATING OF CALIBRATION RESULTS

bull INCLUDE EFFECT OF ADDITIONAL PLATFORMS ON bullBbull

Consider Increased 1Uncertalntles In R and S

Develop bullNeW Ukellhood Functions for Different Categories

middotLet bullxbull Number of Ukely Unexpected Survival Cases and y Number of Ukely Expected Survival Cases

Determine Combined Likelihood Function lk (b Iadditional platforms) middot

bull REVISED LIKELIHOOD FUNCTION

lk(b I all cases) = lk(b I 9 platforms) middot x lk(b Ix platforms) x lk(b I y platforms)

bull ESTABLISH REVISED BIAS FACTOR B

f8 (b) u f8 (b) lk (b I all cases)

II

t t ~

shy

APPLICATIONS OF BIAS FACTORS

ii APPLICABLE TO PARTICULAR LOAD AND RESISTANCE RECIPE

bull USE BY THE API COMMITTEES AND MMS

Criteria Revisions

Recommended Design Guidelines Revisions bull

bullbull REASSESSMENT STUDIES AND DECISION MAKING

middot

bull RISK ASSESSMENT STUDIES BY THE OPERATORS

Individual Platforms

Fleet of Platforms

middot IMR Program Implementation

bullbull NOT TO BE USED IN DESIGN

I I fl J

EXAMPLE USE OF BIAS FACTORS

bull PHASE I EXAMPLE APPLICATION

bull EVALUATION OF ANNUAL PROBABILITY OF FAILURE

Establish Annual Maximum Seastate Data

Determine Coefficients to Define Base Shear

Determine Ultimate Capacity of Platform

Determine Probablllty of Failure Versus bullbbull

-bull DETERMINE UPDATED PROBABILITY OF FAILURE

Pf - P [S gt R]

Pf I P [~ gt Rb I b] f 8 (b) db

Pr - J P [S gt Rb I b] f bull 8 (b) db

bull TOTAL PROBABILITIES OF FAILURE

Total Pf - 1 bull Il [1 middotPf] - l Pf bull

Total Pf - 1 bull Il [1middot Pf] - l Pfmiddot

Total Pr = 1 - n [1middot PfbullJ = l Pfshy

I I I

EXAMPLE APPLICATION bull PHASE I

PLATFORM ST 151 K

bull GULF OF MEXICO

bull WATER DEPTH = 137 FT

bull ANNUAL MAXIMUM HS MEDIAN 175 ft COV 031

bull ASSUMED STORM DURATION 3 hoursmiddot

bull ASSUMED NUMBER OF WAVES 800

bull MEDIAN HS FOR STORM HOURS 1 2 AND 3

bull MEDIAN CURRENT VALUE bull

bull DETERMINE ULTIMATE CAPACITY FOR THREE DIRECTIONS

bull DETERMINE ANNUAL FAILURE PROBABILITIES

Broadsidemiddot Diagonal End-on

0037 0025 0007

0071 0044 0018

0022 0016 0004

DETERMINE TOTAL ANNUAL PROBABILITY OF FAILURE

Pf - 007

Pf - 013

Pr - 004

bull

EXAMPLE USE OF BIAS FACTORS - PHASE II middot

bull DETERMINE PREDICTIVE PROBABILITY OF FAILURE

Case - 1

Pf _ S S Pf (b b2) fa1a2 (b b2) db db2

g bull -1 (b1 R I S) - 1

92 - (b2 R2 I S) - 1

Case -2

Pf s ~ [bJ f11 (b) db

Pf [bJ p [g lt 0)

g - bull 1 (b1 R1 I S) - 1

1

bull1) l

I

BENEFITS OF ANDREW JIP bull PHASE II

lt

bull IMPROVED LOAD AND RESISTANCE RECIPE

bull IMPROVED CALIBRATION PROCEDURE

bull MULTIPLE BIAS FACTORS

bull PROCEDURE TO INCLUDE EFFECT OF OTHER PLATFORMS

bull PROCEDURE TO USE MULTIPLE BIAS FACTORS

  • Structure Bookmarks
    • ANDREW PHASE II JOINT1 INDUSTRY PROJECT
    • PROGRESS MEETING NUMBER 1
    • ANDREW PHASE II JOINT INDUSTRY PROJECT PROGRESS MEETING NUMBER 1 NOVEMBER 16 AGENDA
    • ANDREW PHASE II UST OF PARTICIPANTS
    • ANDREW PHASE I OBJECTIVES AND RESULTS
    • ANDREW PHASE II PROJECT OBJECTIVES
    • SUBSEQUENT PROJECTS
    • ANDREW PHASE II PROJECT TEAM
    • ANDREW PHASE II UPDATE MODELING middotAND ANALYSIS RECIPE
    • -ANDREW PHASE II CONDUCTOR RESISTANCE
    • ANDREW PHASE Il WA VE FORCE PROFILE
    • ANDREW PHASE II WA VE FORCE PROFILEmiddot
    • ANDREW PHASE II BRACE MODELING
    • ANDREW PHASE II FOUNDATIONS
    • ANDREW PHASE II middot middot JOINT MODELING
    • ANDREW PHASE Il JOINT MODELING OPTIONS DISCUSSION
    • ANDREW PHASE II JOINT MODEJJNG OPTIONS ILLUSTRATIONS
    • ANDREW PHASE II JOINT MODELING PHASE II METHOD
    • ANDREW PHASE II ANDREW STORM CONDmONS UPDATED
    • Platform Selection bullbullbull1111111111111
    • BasisCriteria for Selection bullbullbull1111111111111
    • CALIBRATION
    • WHERE TO INTRODUCE BIAS FACTOR (B
    • PHASE I -KEY CONCLUSIONS
    • bull CALIBRATION ISSUES PHASE II
    • PHASE II WORK SCOPE (CALIBRATION)
    • SOFTWARE
    • SENSITIVITY STUDIES POSTERIOR DISTRIBUTION OF B
    • WEIGHTING APPROACH bull
    • INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON
    • CATEGORIZATION OF ADDITIONAL PLATFORMS
    • UPDATING OF CALIBRATION RESULTS
    • APPLICATIONS OF BIAS FACTORS
    • EXAMPLE USE OF BIAS FACTORS
    • EXAMPLE APPLICATION bull PHASE I PLATFORM ST 151 K
    • EXAMPLE USE OF BIAS FACTORS -PHASE II middot
    • BENEFITS OF ANDREW JIP bull PHASE II
Page 12:  · . ' ANDREW PHASE II PROJECT OBJECTIVES . ' • ·UPDATE MODELING AND ANALYSIS RECIPE· JOINTS ' ' ·BRACES WAVE PROFILE FOUNDATIONS . ' ' CONDUCTORS· • ANALVZE PLATFORMS USING

Project STlSlX Model Conductor Version 1 Mon Nov 14 140620 1994

bullmiddot

I

I

I I

I

I I

I I I I I

1 ==

-L

-L - L

-

- L

-

-

L

-

Uniform Displacement step 6 Inelastic Events Legend

Elastic

Strut Buckling Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield Beam Clmn FUlly Plastic Fracture

--

i~~~~~~~~~~~---~--~~-J___--JJ~_middot~~~=-

Project ST151K Model Conductor Version1 Mon Nov 14 142735 1994

---

~

I

I

middot

I

I

I

- shy- shy

- shy

- shy

4x z

CIV Uniform Displacement step 33 Inelastic Events Legend

bull Elasti~ Strut Buckling Strut Residual Strut Reloading Plastic StrutNLTruss Beam Clmn Initial Yield Beam Clmn-Fully Plastic Fracture

gt bull

Project STlSlK middot Model Conductor Version 2 Mon Nov 14 142025 1994

-----------------

bullf

I

I

I

II I I

I I I I

I ==

-

Variable Displacement step 9 Inelastic Events Legend

Elastic Strut Residual

bull

Strut Buckling Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield Beam Clmn Fully Plastic Fracture

--

Project ST151K Model Conductor Version 2 Mon Nov 14 142145 1994

I I I I

I I

I

I

I - - -

-Ishy-

-

Variable Displacement step 35 InelasticEvents Legend

Elastic Strut Buckling Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield Beam Clmn Fully Plastic Fracture

Project ST151K Model Conductormiddot

Version 1

Clgt - Cut Plane Force Fx Mon Nov 14 140051 1994

- -_ vii I fOIZM

0 _

~

N ID 0 bull 0

x 0

middotrt ~ () bull u 0 ~ 0 Ir

N bull

0

0 1 2 3 4 5 X Deck Disp [Ft)

I I 1 1 WbullTI- bullbull

lltcxi lr-C

011lt~~~~~~~---~~~~~~~~L~~~~~~~--~~~~~~~~i~~~~~~~--

Uniform Node Displacement

shy

Ji

r

Comparison of Pushover Curves for ST 151K Variable Load Pattern and Monotonic Load Pattern (Andrew)

Soll Improved 100

6000

5000 -i 4000

f

llS

3000 Ul

J 2000

10 c11 c1lgtolts 1000

0 v

bull ~

f

_ ----- --- ---shy

bull --Variable Lold

- - Andrew Load

I I I I I I

4 5 62 30 1

Deck Dlsplacement (ft)

bullbull r 1 II

~

ANDREW PHASE Il

WA VE FORCE PROFILE

bull DECK WA VE FORCE ~

PHASE I USED SECTION 17 (1056H degCd)

PHASE Il USE REVISED SECTION 17 (1000H Cd)

bull PHASE I USED SINGLE FORCE PROFILE FOR PUSHOVER

- USED WAVE HEIGHT CLOSE TO ULTIMATE CAPACITY

bull PHASE J

Il WILL STUDY THE

USE OF VARIABLE

PROFILE ~

- IF FORCE LEVEL INCREASES (OR DECREASES) BY A FACTOR EQUIVALENT TO A 2 FOOT CHANGE IN WA VE HEIGHT THE INCREMENTAL FO~CE PA1IERN IS CHANGED ACCORDINGLY (CAP FEATURE)

- INITIAL EXAMPLE

f~

a

middotn

a

I N -a a a amiddot

+

~ a

l

a w gt 0c 0 gt a I shyc -CJ wc w gtc(

== c( I shyZ w E W a 0 z- 8

ii sect

~ ~ bullT 0

I if

x ~

bullbullbull

II ~------------________

ANDREW PHASE II

WA VE FORCE PROFILEmiddot

CAP AUTOMATIC APPLICATION FEATURE

STAT

fR1 0 lt f lt1

STAT

STAT

f(R3-R2) middot f lt1

Project ST151K Model Wave_Loading Version 1 rue Nov 8 164107 1994 PT

E~ nss FT

SL 14amp FT

a 111 FT

EL 14 li

E-1 37 FT

El o FT

1middot1middot ______________ _ ________ __ _ __

0

200 400 600 800 1000 12000 Force at Elevatloii (kip)

200

a ll_

Wave Force at Major Elevations for Discrete Wave Heights

180

bullHmiddot40ft

Hmiddot495ft

H bull 51 ft

Hmiddot55ft_

H-59ft I

Andrew (H bull 60 86 ft) II

i I

_11---__

- middot- Hmiddot30ft

-

- bull -

middot bull bull bull middot

---Hmiddot53ft

bullbullbullbullHmiddot57ft

-----H-63ft

Elevation Above Mudllne 100

(ft)

180

140

120

80

60

40

20

Loading at Major Elevations

---H-49511

-deg-Hmiddot608611

0 200 400 600 800 1000 1200

Force It Elevatlon (kip)

--

Base Shear vs Wave Height ST151K

6000

5000 1---------Ult1mateCapacitybull 4890 kip (100 Improved Soll)-----------~--

4000a a lii ~ 3000

81 ampI

2000

1000

0 +-~~~--i~~~~~~~~~-+-~~~--lf--~~~~~~~~-+-~~~--l

30 35 40 45 50 55 60 65

Wave Height (ft)

bull middot

~

bull

II

2

g

~

It)

0

cCD E 0 E a c E euro~ CDshygtItgtmiddoto ~ti c -a Gic CD gt ~

-= c -Cll

I z

bullO 0i i ~ g 0

() CI -bull(q-ll) iuawow llu1wniiaAO

11

shy

116

114

112

g 110

bullc = 108

~ ~ 106 -sect 104

102

100

98

96

-

6555 60

Wave Height vs Center of Action ST151K

30 35 40 45 50

Wave Height (ft)

t

Comparison of Pushover Curves for ST 151K Variable Load Pattern and Monotonic Load Pattern (Andrew)

Soll Improved 100

--- --------- --- ----6000

5000

- 4000 8 - I

1 3000 Ul

81 m 2000

_J -shy

1000

---Variable Load

- - Andrew Load

0 f-~~~~~r--~~~~-t-~~~~~-t-~~~~~+-~~~~----t~~~~~-1

0 1 - 2 3 4 5 6

Deck Displacement (ft)

-middotshy

1 t I I ~

ANDREW PHASE II

BRACE MODELING

bull MARSHALL STRUTS USED FORPHASE I

bull PHASE II WILL ASSESS DIFFERENCE IN USING BUCKLING BEAMS AND MARSHALL STRUTS

middot bull COMPARE FOR 2D FRAME

bull COMPARE FOR PLATFORM 177 WHERE LOWER

BAY EXHIBITS LARGE BRACE MOMENTS

Project andrew Model beamframe Version 1 Mon Nov 14 145447 1994

~

middot Frame Test buckling beam-column model

-

Project andrew Model strutframe Version 1 Mon Nov 14 145754 1994

bull

Frame Test strut model

middot-middot _____________--_

_middotmiddotr_~ --ll--~-middot__

Buckling Beam bull

Strut (K=O 65)

Lateral Displacement

ANDREW PHASE II

FOUNDATIONS

bull ~LICIT NONLINEAR API FOUNDATION USED INiPHASE I

bull FOUNDATIONS

JIP SELECTED MINI-VANE DATA FOR

STRENGTH PROFILE

bull COMMENTS ON BENCHMARK COMPARISONS

- SIGNIFICANT VARIABILITY DUE TO FOUNDATION MODELING METHODS

II Ii

z

ctV 4x SPlO caissonfy-42s soil-api(lf-097step-114)

Inelastic Events Legend Elastic Strut Buckling

Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield

--------shyBeam Clmn Fully Plastic Fracture

~--~~~~~~~~~--~----~__i~~---~~~~__--~

Pro)ect apipile Model South Pelto 10 Version Wed May 18072235 1994 - -

1bullt

+

t

f i bullbullbull

bullbull bullbull

I I

5

~x SPlO Caissonfy=42s soil=api(lf=lOOstep=137)

Inelastic Events Legend

Elastic Strut Bucklilg

Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield

earn Clmn Full~ Plas1c Fracture

ANDREW PHASE II middot middot

JOINT MODELING

middotbull API RIGID-PLASTIC USED IN PHASE I

a ELASTIC-BRITTLE USED IN PHASE IB

amiddot MSL WILL PROVIDE NONLINEAR FORCE-DEFORMATION

AND MOMENT-DEFORMATION RELATIONS FOR PHASE II

a MODELING OPTIONS FOR PHASE II -

l

- UNCO~LEDmiddotONE DIMENSIONAL MEMBER END JOINT MODELING

J Ii

ANDREW PHASE Il

JOINT MODELING OPTIONS DISCUSSION

bull JD COUPLED INTERACTION SURFACE

- IDEAL

- DATA NOT AVAILABLE TO CALIBRATE

bull 2D COUPLED INTERACTION SURFACE

- SIGNIFICANT MODELING IMPROVEMENT

- DATAAVAILABLE

- IMPLEMENTATION PRACTICAL BUT NEEDS DEVELOPMENT AND CALIBRATION

bull 2D UNCOUPLED INTERACTION SURFACE

- SIGNIFICANT MODELING IMPROVEMENT

- DATA AVAILABLE

- IMPLEMENTATION PRACTICAL AND AVAILABLE

ANDREW PHASE II

JOINT MODEJJNG OPTIONS ILLUSTRATIONS

P

e

ANDREW PHASE II

JOINT MODELING PHASE II METHOD

e

I I I

ANDREW PHASE II

ANDREW STORM CONDmONS UPDATED

bull -PHASE I USED OCEANWEATHER HINDCAST-(Nov 1992)

bull middotmiddotPHASE II WILL USE THE REVISED HINDCAST (Nov 1994)

bull lt c lt I lt

2-2

l---4-~~~+-~~-+-~~~1--~~-+-~~--l--~T--+-~~--i24

f-tJVtMfiz~ 1Cfl (r STvb(

Plollol bullbull 5middotNOYmiddot9214bull1J Irbullbull Illbull [ANDREVl~NDREVCUSTbullMAXll 5-NOY-1992 10bull54 - 9208

HEIGHT lml

I~ MAXI

I L 1~middotamp SZ~- ~A I 13c

I J) ( r I X~0iWJ~0gtJ gt L gt gt I 1c I 120

-- N

~ deg

I I I I I I e C I ~=-41 122 -94 -92 -90 -88 -86 -84 -82 -80

-

Figure 7 contours of maximum hindcast significant wave height

_

~

~tl ~

NOltNttIJ 17lt ltNS S iJJ)y

Maximum Significant Wave Height - Hurricane Andrew 1992

30 ~

26 ~ I ~-~~~- == -shy -~ 35-_

24deg1--shy --------~- middot-middot-------middotmiddotmiddot----middot------middot-middot~middot- - - ~-----1---middot

22middot ~~~~~~~-~ -94 -92 -90 -88 -84deg-86 -82 -somiddot

c

-

Platform Selection bullbullbull1111111111111

0 BasisCriteria for Selection middot

0 Review of Phase I Platforms middot 0 Alternative Platforms

0 PMB Recommendations

0 Participants Comments

gt

L

l I

BasisCriteria for Selection bullbullbull1111111111111

0 Benefit for Calibration - Classification (UnexpectedExpected)

- Failure Mode (StructuresFoundations)

- Damaged StructuresKnown response

- Better Representation of Full ~opulation

0 Availability of Data - Existing Models

- Inspection Data

- Damage Data

PREDICTED BEHAVIOR

-

OBSERVED

BEHAVIOR

-SURVIVED DAMAGED FAILED

SURVIVED

ST134W WD90A

-MC311 -

MC397 -

--

-

-

ST151K ST130Q

-

-

-

DAMAGED - -

-

-

--

-

-

ST52 SS139 ST177B -

ST161A

-_

-

FAILED

-

-

-

ST151H ST130A ST72

Operator - Platfonns - Platfonns Survived Damagedamp

- Falled -

bull

-

-

Jle eBDIGIPllDbl -Amocoshy 30 1 Chevron -

143 12 Exxon 72 Mobll 22 1 Phllllps 7 bull

-Shell 39 Unocal 25 1

Total Partlclnts Platfonns 338 15

Other Operators bull- 310 13

-

Total Platfonns - -

648 28

fbml

Platforms Considered 6 7

tie of Total Platfonns 1 25

OVERVIEW OF PLATFORMS SURVIVEDDAMAGEDFAILED

-

shy

v

Platforms Evaluated in Phase I

-

-Platfonn Water Year Number Analysis Avallabillty of Availability Platform

- Name Depth rl of Model DetaDed rl ClassUlcatlon

Installation legs Inspection Soll Report Based on Information Phase I Calibration - -ft

- -- Suaill flalfnrm ~BSi -STISIK 137 1963 8 CAP Yes Yes Un eel Sarvlval

-STIJOO 170 1964 4 CAPmiddotshy Yes ST134data u ST134W 137 1981 4 CAP Yes Yesmiddot E

-

edSarvlval -

middotsurvival

WD90A 184 - 1964 8 CAP Yes E -

- Sarvlvill

MC3ll 343 1978 8 KARMA

MC397 468 1991 4 KARMA -

bull ~reglmmiddot- -

Yes Sure Survival

bull Yes - Sare Sanlval-

-T23STS2 63 1969 4 CAP

T2SSS139 62 1969 4 CAP -

Yes

Yes

Yes Ex

Yes Une

bull SurvivalLikely to DamalI

middot bull SanlvalLlkely to DamalI

ST161A 118 1964 8 USFOS

Eail11a pound1al(bullllm Clss ST177B 142 --1965 8 CAP

STISIH 137 1964 - 8 CAP C

STIJOA 140 19S8 8 CAP - -

Yes

Yes

ollamed

Yes

- Yes E

-

at about 1 mUe

Yes

ST134data

bull SafvlvalLlkelv to Dam- bull- -

E middotrahre

Emeried Failure -

Likely FaUare

T21ST72 61 1969 4 CAP Collamed No Likely FaUure

Platforms Identified for Evaluating Their Likelihood of Falling Under UnexpeCted Survival Category

CllEllATOa

AllB

llLOCll

Sl1lvcnJllE

NAME

YEAR

INSTAUJCD

WATER

IJBPJB

middotmiddotDISTANCE

ftOMSllOllE

(MIUIS)

Ill

(M)

AMOCO BllW

STllI

WDI075

A

bull D

4

n uo 17J

44

32

bull 20

10

bull CllEVllOH STtllO c I 1bull 21 11

STl130 D 0 161 21 11 STllJ4 I 0 137 D - 11

STIUI G I 137 32 11

STIUI I 0 121 32 11

STIUI

J 0 140 32 11

STIUI It 4 137 32 11

STllSI PROl)I 0 137 32 11

STllSI PROD-2 0 137 32 11

STll7 A 0

140 35 11

WDl117 c 65 214 34 10

DXON STshy E 51 14 bull STl165 A t3 bull 34 WD0073

WD0073

WD0073

A _ c

4

62

4

IA IA

17J

22

22

22

bullbullbull WD003-shy E 65 161 Z7

A 0 IN 31 MOBD Bl12I

ssn A bull

50

50

31

7

bull 7

MIJllPllY SS114 B 7 50 7 7

PlllUIPS SSl14t A 55 33 7

SAMEDAN ST17J A 4 117 7 bull 7

SJIELL WD0103 A 65 223 27 WDllOJ bull 65 221 27 WDll04 c 65 221 27 WDOI04 D 67 2 27

UNOCAL

SSl20I

SSl20I

bull SSl20I

SS0209

SS0209

SS0209

SS0209

SS 0253

SS 0253

E

I

PtJllE

A

bull D

G A

c

0

4

4

7J

61

0

0

69

103

97

97

9S

100

95

100

165

175

34

34

34

34

34

34

34

54

54

bullbullbullbullbullbullbullbullbull

Platforms Recommended for Phase II

llalr y_ w- Aot Av- _ I 1 Doplla Model nm bull bull _ bull I I H11 s18-1 Irht lbwl~ bullbullbull(-illtdlel OM lllMtD

II ---

-STISIIC 137 1963 I CAP Yeo Yeo I Yeo

STtshy -170 1964 4 CAP Yeo ST134dlU I Yeo

STl34W 137 Yeo Yeo -1981 4 CAP D Yeo

WD90A 114 1964 I CAP Yeo - D

MC311 30 1971 I KARMA Yeo -sre-

MC397 middot 461 1991 4 ICARMA Yeo Sireshy - - r- -

__ T13ST5l 63 1969 4 CAP Yeo Yeobull middot- ~n- D Yeo _ Tl5SSl39 62 1969 4 CAP Yeo Yeo I Yeo

_ STIT711 142 1965 I CAP Yeo I Yeo middot-1shySTl61A Ill 1964 I USFOS Yeo Yeo - D bulloshy

- r-

STISIH 137 1964 I CAP y D y -_ STl31A 140 1951 I CAP Yeo STl34dlU I Yeo

T21Sll 61 1969 4 CAP No rar-middot I Yeo

~-rmiddot-middot SS209A 95 1967 16 CAP Yeo D -SSU4R

11 UI A 41 -in D SS 2741AIM CAP bull

_ _ -shy - -shy- -shy

CALIBRATION

bull PHASEIAPPROACH

bull CALIBRATION ISSUES

bull PHASE II WORK SCOPE

bull SOFlWARE AVAILABILITY

bull EFFECT OF ADDITIONAL PLATFORMS

bull WEIGHTING APPROACH

bull EXAMPLE APPLICATION middot

(l=Ofce Mode] Wind

~

fcurrent l -middot

~SW ~leldl4

ultimate resistance force I first

design load Illgt

i _

deflection

-

bull

4

Historical Hurrlcane Dariiage and Survival of Platforms

Components Required for Calibration (Petrauskas 1992)

PRIOR Distribution

PLATFORM EXPERIENCE

BAYESIAN Survivals and Failures

UPDATING

Waves Winds and Currents bull

POSTERIOR Distribution

- A_middotJA ~ Survival Data Failure Data

Result Result

Load Resistance

middotfl

-

Bayesian Updaing middotGeneral Description (Petrauskas 1992)

L Imiddot

WHERE TO INTRODUCE BIAS FACTOR (B

On Resistance (Rmue - B (RPREDICreD I

middotOn Loads (Smue middot - B (SPRE01cTEo

On Safety Margin (RSmue - B (RSPREDICTED

CALIBRATION METHODOLOGY

gt

r

FIELD OBSERVATIONS

Observed Behavior

survivals Damages

Failures

r

bull

CAPACITY ANALYSIS

Predicted Load and Resistance Predicted Failure Modes

Survival Damage Failure cases

RELIABILITY ANALYSIS

- Probability of Failure ~

Ukellhood Failure

BAYESIAN UPDATING

Prior Distnbution of B Combined Likelihood Function

Posterior Dlstnbution of B

BIAS FACTOR

Mean Value cov bull

Represents Modeling Uncertainties

Prior Distribution of B Likelihood Functionmiddot Observing Success case STISIK and Posterior Distribution of B

2 00

uo

160

140

120

e 100 L $ uoIshy

0 60 ~ ~ 040

020

0 00

0 00 0 50

_ I 00

b

I 50 2 00

-shy lk [bl STISIK)

--shy Pnor-f(b)

--shy Pos1enor r(bl ISIK)

Prior or B Mean bull 10 COV bull 030

Posterlor of Bbull Meanbull 136COV 0 16

2 50

Likelihood Function and Posterior Distribution Example Platform STISIK

- -

~

- Prior and Posterior Distributions or B Failure cases (ST177B STlSlH ST130A T21) Damage cases (T23 T25 ST161A) Success cases (STlSlK ST130Q ST134W WD90A MC311 MC397)

400

350

300

2 so 0-

I

bull bull

bull bull gt

- bullbull ~ bull bull I bull

_ bull bull bullbull I bull bull bullII V -

bull bull

I bull ~ bull

bullbull bullI bull bullI bull bull bull

middot -~

bull__ ~ middotI middotmiddot

L 2 00

e ~ I SO

-I 00

- 0 so

000 gt

0 00 020 0 40 060 0 80 I 00 120 140 160 110 200

b -

-

shy

Prior-r

- - - - Posterior fLlll failure cases

bull bull bull bull bull bull bull Posterior flall damage cases

- bull bull - bull middot Postenor flall success cases

Posterior flall success fulure damage cases

-

Posterior of B Mean= 119 COV 0101

Posterior Distribution or Blas Factor (8) bull Different Categories and All Cases Combined

shy

l

PHASE I - KEY CONCLUSIONS

bull GLOBAL CORRECTION FACTOR

bull 15 TO 20 CONSERVATISM MEAN)

bull UNEXPECTED SURVIVALS HAVE GREATEST IMPACT

bull DAMAGED CASES MORE INFORMATIVE

bull CALIBRATION ISSUES PHASE II

bull

EFFECT OF IMPROVED CAPACITY ANALYSIS RECIPE

bull APPLICABILITY TO MULTIPLE FAILURE MODES

bull NO FOUNDATION DAMAGE CASE INCLUDED

bull IMPROVED REPRESENTATION OF FULL ANDREW DISTRIBUTION

-bull NUMBER OF PLATFORMS CONSIDERED

I 1 I

I t ) bull ~

PHASE II WORK SCOPE (CALIBRATION)

bull

bull ESTABLISH MULTIPLE BIAS FACTORS

Use Nine Base Case Platforms

middot - Capacity Analysis for Revised recipe

Foundation Blas Factor from Foundation JIP

bull DEVELOP WEIGHTING APPROACH

bull UPDATE MULTIPLE BIAS FACTORS

bullEffect of Platforms Not In the Nine Base Cases

bull EXAMPLE APPLICATION OF BIAS FACTORS I

bullbull

PHASE I PHASE II

CAPACITYbull ANALYSISRu bull

---------------------1----middot

Jc1 - iJbullraquoJ dbull

FttlS frt[f~l ratdhJJ

-

LIKELIHOOD FUNCTION

DEVELOPMENT

lk Jb Jfall1N) bull Jb)

111 Ill Isurvtvoq bull 1 - bl

-middot -------------------shy ~---------------------middot-middot

r111ir8 Cb1 111 Cbl BAYESIAN

UPDATING

~cove

R1R2 bull middot---------------------~----

Jblbull P(g cOJ

1 bull JbRI (MBSI I bull 1

~middot

llcJllJfalluro) bull P(gcOJ Ill 111 b21fallun) bull P ((g1 CO) U 1112 c OD

middot--------------------shy ----------------------shy bull

r111111r8 Cbl 111 Ill i-1nro1

deg ~1b2)laquottfrr (b1b21-lnfo) re Ill bull 1-11middot b2) bull lodegamp

middot

i1 covamp i2 cov~i

-

shy

--

Likelihood Function vs b All Success Case STISI K STl30Q STl34W WD90A MC311 MC397

I 0000

0 9000

OIOOO

0 7000

~ 0amp000

ta osoOO

~ 0 4000

0 3000

0 2000

0 1000

bull 0 0000

bull

I

middot

middot bull I

Ibull

-sibull ~ -middot-middot-middot-middot -middot-

----7----+-----~ I

Ibull bull bull I

rbull

--- Jk ISIK[blsuccess)

- bull - bull - lk 130Q[blsuccess) bull

lk 134W[blsuccess)

Jk-WD90A(blsuccc~I

- - - lk MC397[blsucccss)

--ltgt-- lk middotMC31 l[blsucccss)

--- Jk [bl 6 out of6succcsHascs)

05 I 5 2 250

b

Likelihood Functionsmiddot Success Cases - Phbullsc L

shy

bullbull bull bull bull bull bull -------------------~

bull bull bullbull bullbull bullbull bullbull bullbull bullbull bull

bull bullbull

bullbullbullbull

bullbull

bullbullbullbullbullbullbull

bullbullbullbullbullbull

bullbullbullbullbullbullbull

z 0

t z ) LL

c 0 0c i ~ i

J

~ sect ~--------------~ e l5

I

I

SOFTWARE

bull PF PROGRAM

bull CALREL

bull PF +CALREL

bull RELACS

bull

gt i

l

SENSITIVITY STUDIES

POSTERIOR DISTRIBUTION OF B

bull EFFECT OF CHANGE IN DISTRIBUTION OF PRIOR OF B

bull EFFECT OF REMOVAL OF ONE PLATFORM FROM EACH CATEGORY

bull EFFECT OF REMOVAL OF THREE DAMAGE CASES

bull EFFECT OF ADDITIONAL 13 SIMILAR PLATFORMS

-I

Prior and Posterior Distributions of B Variation In COV of PriorofB fromOl to04 -- -

4500

-4000

-

- -

3500 --middot - bull bull - bull bull Prior-r ICOV =02

3000

- - - - Prior-r ICOV =03 -

bull -~ e 2soo -

bull bull bull bull bull bull bull Pnor-r ICOV =o 4t frJ ~ 2000 Posterior rlCOV =0 2

bull bull Postenor rlCOV =0 3 bull bull 1500 bull middot bull - shy -middot- Posterior rlCOV =04

middot~ I 000 - bull

-middotmiddot ~ ~middotmiddot M -~

- middot 0 500

wbullbullbull _ I middot~ middotmiddotmiddotmiddotmiddot middotmiddotmiddotmiddotmiddot - -~ ~ __ middot_ _ bull -shy0 000

0 000 0200 0 400 0 600 0 800 IOOO 1200 1400 1600 1800 2000 -

- - b

-

=shy

Posterior Distribution of Bias Factor (B) bull Effect of Variation In COV of Prior of B

bull

bullPrior and Posterior Distributions of B Evaluating Effect of Removal of One Platform from Each Category from the Base Case of 13 Platforms

400

~ ~ 350

j IA i

3 00 bull I bull I ---Prior-r I bull l

250 1 I 1

I bull bull ---Posterior r lall 13 casesshy- ~E 2 00 bull bull bull bull bull bull bull Posterior r without STISIK

- ~ 1 bull - bull I - bull bull - bull bull Posterior r Iwithout T2I I 50 I

J-tt~i - bull - bull - Posterior r Iwithout T2S

tOO [-

I I ~ ~ I 0 50

0 00

___ r J i

l~T

~--

0 00 0 20 0 40 0 60 080 I 00 120 140 160 110 2 00

b

zshy

Posterior Distribution of Bias Factor (8) bull Effect of Removal of middotPlatform from Any Category

bullPrior and Posterior Distributions or B Evaluating Effect or Removal or 3 Damage Cases from the Base Case or 13 Platrorms

4

35

3

25

e L 2 ~ bulle shy

I 15

bull I

05

0

0 02 04 06 I 2 I 4 16 IS 2

I I

I

I

I Ibull

J -tmiddot L--

bullI

~

I

fl

II ~

I I v I

middot bull bull

bull L

~ ll- I-~lt

OS I

b

I

bull

1middot bullbullbullbull

~ __

bull

- - - - Prior-r

- - - - Postenor flall failure cases

Postenor flall success cases

--- Posterior flall 13 cases

---Posterior fI success failure cases only

IPosterior or B Mean =t11 COV =0121

-

_I

Posterir Distribution or Bias Factor (B) bull Effect or Removal or 3-Damage Cases

-

bullPrior and Posterior Distributions or B Evaluating Effect or Additional 13 Similar Observations as

Used in Andrew JIP The POsterior or B from Andrew-JIP Considered as New Prior or B

- 500

450

400

350

bull bull

bull bull bull bull J bull

bull bull f ~ I

I ~ Ibull bull

bullbull bull

l bull I -bullbull bull bull

I

I bull bull

bull I I~~bull bull bull

bull ~

I bull bull

~ ~_ ~~~

- - New Prior r bull

Posterior 114 new failure cases ~ 300- bull bull bull bull bull bull bull Posterior f 13 new damage casesl 150 ~

- bull bull - bull bull Posterior 116 new success cases i 2 00

Postenor 1113 new success failure -150 damage cases

I 00

0 50 -

0 00 0 00 0 20 0 40 0 60 0 80 I 00 120 I 40 160 110 2 00

Posterior ol B Mean= 120 COV= 0071 b

shy

shy

_rshy

Posterior Distribution of Bias Factor (B) bull Effect of Additional 13-Similar Observations

1

1 I

WEIGHTING APPROACH bull

bull ASSESSMENT OF PLATFORMS DATA

Select Platforms Most Mappable to the Analyzed Platforms

Unlikely SurvivalDamage Cases Identification

Based on Cut-Off Criteria Agemiddot Hs etc

bull INFORMATION REQUIRED FOR COMPARISON

bull CATEGORIZATION OF ADDITIONALmiddot PLATFORMS

bull UPDATING OF CALIBRATION RESULTS

--

OVERVIEW OF PLATFORMS SURVIVEDDAMAGED -~

Operator -

- Platforms

Survived CS)

Platforms

Damaged CD)

Platforms Used In

Phase IJIP -

JI~ ~Bill~IRBDll -

Amoco -

Chevron -

Exxon

Mob II

Phllllps

Shell

Unocal -

30

143

72 22

7

39 25shy -

1

12

bull

1 bull

1

1(S)+1 (D)

3 (S) +3(D)

2 (S)

bull shy

bull

bull

bull

Total Partlclnants Platforms 338 - 15 10C6S+4D) -

Other Operators

310

13

3 (D)

Total Platforms 648 28 13C6S+7D) -

Summary of Andrew JIP bull Phase I Platforms

Plairshy w v- N-ber Lea-le 1 shy car Hind_ o-w N-bull Deplh fl fl IJpe la bull n Yeadon Jadtet lbelledtf liefalit -- _ -OD

IL H - -shy _ICllshy IL IL - - rQClto ~ _

K STUIK 137 1963 8 365 608S4 Cellardeck

-middot STl-shy 170 1964 4 -~ K 39 62269 Cellardeck - --ST134W 137 1911 4 43 608S4 Smvlnlshy- _ - -

WD90A 184 1964 8 K 267Sbull34 83 SOS31 SUboftllar deck SmvMI

MC311 343 1978 8 x S7 62 - MC397 468 1991 4 x shy S33bull6SO 65911 -shySmeSmvlvol -- gt -

r-- T23ST52 63 1969 4 ton- shyKshy 52594 S03 -T25SSl39 62 1969 4 Kshy SOS94 S063 - middot ton-shy__ -

bullbull a ton--shyST161A 118 1964 8 K - 347S 51495 Cellarclecti

r~

ST1771 142 1965 8 K 39 60167 CeDordeck failme

STISIH 137 1964 8 K 395 608S4 CeDardeck Failme

K ST130A 140 1958 a SS 6095 I lblv Pai1uno - T21ST72 61 1969 4 Kshy S4S94 4974 I lhlv Pai1uno -

- -

JE

INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON

bull WATER DEPTH

bull bull YEAR OF INSTALLATIONDESIGN

bull PLATFORM ORIENTATION

bull PLATFORM LATITUDE AND LONGITUDE

bull KEY SKETCHES OF PLATFORM STRUCTURAL FRAMING

Number of Legs Plies Leg Batter Number of Conductors bull Deck Elevations (Including sub-cellar deck If any) Jacket Framing Elevations Brace type In Vertical Frames Sizes of Primary Members (leg pile vertical frames braces primary horizontals) Joint Details (can gap etc) LegPiie Annulus Grouting Status

bull middotPOST ANDREW INSPECTION REPORT Significant Wave Height Hs (ft)

bull INFORMATION ON DAMAGES PRIOR TO ANDREW

bull RESULTS FROM IN-HOUSE STUDIES IF AVAILABLE

bull ~RIGINAL DESIGN CRITERIA INFORMATION IF AVAILABLE

bull SOIL SHEAR STRENGTH PROFILE IF AVAILABLE

Platforms Identified for Evaluating Therr Likelihood of Falling Under Unexpected Survival Category

OPlRATOR

middotADA

BLOClt

llmucnJllE

NAME

YEAll

INSTAUJD

WA1ER

DBPTll

DISTANCE

JlllOM SHOllE

(MILU)

Bl

(M)

AMOCO 111113

STl11

WDI075

A

bull D

St

17

120

1n

44

20

10 I

CllllVllQfl STlllO c 61 110 21 11 STIUO D Q 1110 21 11

STl134 p Q 137 29 11

ST0151 G 61 137 11

ST0151 I 3 121 11

ST0151 J Q 140 11

STl151 It 137 11

ST0151 PJl001 Q UT 11

ST0151 PROD-2 Q UT 11

ST017 A 3 140 35 11

WDIUT c 65 214 34 10

llXXON STI055 II 51 6T 14 I

STl15 A t3 34

WDI073

WD0073

Amiddotshy Q

1111 1111

22

22

I

I

WDI073 c 1n 22 I

- WDOCl93-shy II

65 1110 2T A 3 194 31 MOBU 111121

SS IOl7

A B

501

50

50

31

T

I

T 6

MtlllPllY SS114 B T 50 T T

PlllllIPS SS0149 A 5 55 33 T

SAMllDAN STln A 107 T T

SBllLL WDl103 A 65 223 2T WDl103 B 65 221 2T WD0104 c 65 221 2T WD0104 D 6T 20 2T

UNOCAL SS020I

SS 020I SSl20I

SS 0209

SS0209

II p

PURll

A

B

3

n 61

103

97

97

95

100

34

34

34

34 34

bullbull bullbullbull

- SS0209 D u 95 34 I

- SS 1209 SSl253

SS0253

G A

c

6T

2

100

165

1T5

34

54

54

bullbull I

I I

I I

SOUTH MARCH

~LAND

N

SOUTH PaTOI ENE ISWltD AREA

MMS Defined Limits ror Offshore Structures Requiring Inspections Following Andrew

t---Jbulln3 T25bull I bullTtl

bull1$1Htl1Abull bull1SIK1779bull

~

middot-DaTA AREA~T

CRANO rsLE

AREA bull1300

bullll4W

=shy

bullPath of Hurricane Andrew with Platforms Used In Calibration

Likelihood Function vs b All Success Case STISI K STIJOQ ST134W WD90A MCJll MCJ97

I 0000

0 9000

0 8000

07000

-Bo6000

tiiii 0 5000 -

04000

0 3000

0 2000

01000

0 0000

0 05 25

ri1rt middot-middot-middot-middot-middot-

I 5 2

b

---11 middot151K(blsu~ccsI

- bull - bull - fk 130Q(bfU~lCSS)

lie bull134W(blsuccc]

lk-WD90A[blsultCS]

- - - lk middotMC397(bfUtlCS]

--o-- lk-MCJll(bbuctcsJ

--- ll (bl 6 out of 6 Mlttcs case]

_

( Likelihood Functions middotSuccess Cases - p - 1 c I

CATEGORIZATION OF ADDITIONAL PLATFORMS

I I I

bull CATEGORIES

Unexpected Survivals

Expected Survivals

bull APPROACH

Identify Corresponding Platform(s) Evaluated In the JIP

Determine Max Wave Height at Platform Location

Evaluate Whether Wave Hit the Deck

middot - Evaluate Likelihood of R and S to Differ from the Base Cases

Review Joint Information

Assess Special Features

bull IDENTIFY NUMBER OF PLATFORMS IN EACH CATEGORY

I

UPDATING OF CALIBRATION RESULTS

bull INCLUDE EFFECT OF ADDITIONAL PLATFORMS ON bullBbull

Consider Increased 1Uncertalntles In R and S

Develop bullNeW Ukellhood Functions for Different Categories

middotLet bullxbull Number of Ukely Unexpected Survival Cases and y Number of Ukely Expected Survival Cases

Determine Combined Likelihood Function lk (b Iadditional platforms) middot

bull REVISED LIKELIHOOD FUNCTION

lk(b I all cases) = lk(b I 9 platforms) middot x lk(b Ix platforms) x lk(b I y platforms)

bull ESTABLISH REVISED BIAS FACTOR B

f8 (b) u f8 (b) lk (b I all cases)

II

t t ~

shy

APPLICATIONS OF BIAS FACTORS

ii APPLICABLE TO PARTICULAR LOAD AND RESISTANCE RECIPE

bull USE BY THE API COMMITTEES AND MMS

Criteria Revisions

Recommended Design Guidelines Revisions bull

bullbull REASSESSMENT STUDIES AND DECISION MAKING

middot

bull RISK ASSESSMENT STUDIES BY THE OPERATORS

Individual Platforms

Fleet of Platforms

middot IMR Program Implementation

bullbull NOT TO BE USED IN DESIGN

I I fl J

EXAMPLE USE OF BIAS FACTORS

bull PHASE I EXAMPLE APPLICATION

bull EVALUATION OF ANNUAL PROBABILITY OF FAILURE

Establish Annual Maximum Seastate Data

Determine Coefficients to Define Base Shear

Determine Ultimate Capacity of Platform

Determine Probablllty of Failure Versus bullbbull

-bull DETERMINE UPDATED PROBABILITY OF FAILURE

Pf - P [S gt R]

Pf I P [~ gt Rb I b] f 8 (b) db

Pr - J P [S gt Rb I b] f bull 8 (b) db

bull TOTAL PROBABILITIES OF FAILURE

Total Pf - 1 bull Il [1 middotPf] - l Pf bull

Total Pf - 1 bull Il [1middot Pf] - l Pfmiddot

Total Pr = 1 - n [1middot PfbullJ = l Pfshy

I I I

EXAMPLE APPLICATION bull PHASE I

PLATFORM ST 151 K

bull GULF OF MEXICO

bull WATER DEPTH = 137 FT

bull ANNUAL MAXIMUM HS MEDIAN 175 ft COV 031

bull ASSUMED STORM DURATION 3 hoursmiddot

bull ASSUMED NUMBER OF WAVES 800

bull MEDIAN HS FOR STORM HOURS 1 2 AND 3

bull MEDIAN CURRENT VALUE bull

bull DETERMINE ULTIMATE CAPACITY FOR THREE DIRECTIONS

bull DETERMINE ANNUAL FAILURE PROBABILITIES

Broadsidemiddot Diagonal End-on

0037 0025 0007

0071 0044 0018

0022 0016 0004

DETERMINE TOTAL ANNUAL PROBABILITY OF FAILURE

Pf - 007

Pf - 013

Pr - 004

bull

EXAMPLE USE OF BIAS FACTORS - PHASE II middot

bull DETERMINE PREDICTIVE PROBABILITY OF FAILURE

Case - 1

Pf _ S S Pf (b b2) fa1a2 (b b2) db db2

g bull -1 (b1 R I S) - 1

92 - (b2 R2 I S) - 1

Case -2

Pf s ~ [bJ f11 (b) db

Pf [bJ p [g lt 0)

g - bull 1 (b1 R1 I S) - 1

1

bull1) l

I

BENEFITS OF ANDREW JIP bull PHASE II

lt

bull IMPROVED LOAD AND RESISTANCE RECIPE

bull IMPROVED CALIBRATION PROCEDURE

bull MULTIPLE BIAS FACTORS

bull PROCEDURE TO INCLUDE EFFECT OF OTHER PLATFORMS

bull PROCEDURE TO USE MULTIPLE BIAS FACTORS

  • Structure Bookmarks
    • ANDREW PHASE II JOINT1 INDUSTRY PROJECT
    • PROGRESS MEETING NUMBER 1
    • ANDREW PHASE II JOINT INDUSTRY PROJECT PROGRESS MEETING NUMBER 1 NOVEMBER 16 AGENDA
    • ANDREW PHASE II UST OF PARTICIPANTS
    • ANDREW PHASE I OBJECTIVES AND RESULTS
    • ANDREW PHASE II PROJECT OBJECTIVES
    • SUBSEQUENT PROJECTS
    • ANDREW PHASE II PROJECT TEAM
    • ANDREW PHASE II UPDATE MODELING middotAND ANALYSIS RECIPE
    • -ANDREW PHASE II CONDUCTOR RESISTANCE
    • ANDREW PHASE Il WA VE FORCE PROFILE
    • ANDREW PHASE II WA VE FORCE PROFILEmiddot
    • ANDREW PHASE II BRACE MODELING
    • ANDREW PHASE II FOUNDATIONS
    • ANDREW PHASE II middot middot JOINT MODELING
    • ANDREW PHASE Il JOINT MODELING OPTIONS DISCUSSION
    • ANDREW PHASE II JOINT MODEJJNG OPTIONS ILLUSTRATIONS
    • ANDREW PHASE II JOINT MODELING PHASE II METHOD
    • ANDREW PHASE II ANDREW STORM CONDmONS UPDATED
    • Platform Selection bullbullbull1111111111111
    • BasisCriteria for Selection bullbullbull1111111111111
    • CALIBRATION
    • WHERE TO INTRODUCE BIAS FACTOR (B
    • PHASE I -KEY CONCLUSIONS
    • bull CALIBRATION ISSUES PHASE II
    • PHASE II WORK SCOPE (CALIBRATION)
    • SOFTWARE
    • SENSITIVITY STUDIES POSTERIOR DISTRIBUTION OF B
    • WEIGHTING APPROACH bull
    • INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON
    • CATEGORIZATION OF ADDITIONAL PLATFORMS
    • UPDATING OF CALIBRATION RESULTS
    • APPLICATIONS OF BIAS FACTORS
    • EXAMPLE USE OF BIAS FACTORS
    • EXAMPLE APPLICATION bull PHASE I PLATFORM ST 151 K
    • EXAMPLE USE OF BIAS FACTORS -PHASE II middot
    • BENEFITS OF ANDREW JIP bull PHASE II
Page 13:  · . ' ANDREW PHASE II PROJECT OBJECTIVES . ' • ·UPDATE MODELING AND ANALYSIS RECIPE· JOINTS ' ' ·BRACES WAVE PROFILE FOUNDATIONS . ' ' CONDUCTORS· • ANALVZE PLATFORMS USING

--

i~~~~~~~~~~~---~--~~-J___--JJ~_middot~~~=-

Project ST151K Model Conductor Version1 Mon Nov 14 142735 1994

---

~

I

I

middot

I

I

I

- shy- shy

- shy

- shy

4x z

CIV Uniform Displacement step 33 Inelastic Events Legend

bull Elasti~ Strut Buckling Strut Residual Strut Reloading Plastic StrutNLTruss Beam Clmn Initial Yield Beam Clmn-Fully Plastic Fracture

gt bull

Project STlSlK middot Model Conductor Version 2 Mon Nov 14 142025 1994

-----------------

bullf

I

I

I

II I I

I I I I

I ==

-

Variable Displacement step 9 Inelastic Events Legend

Elastic Strut Residual

bull

Strut Buckling Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield Beam Clmn Fully Plastic Fracture

--

Project ST151K Model Conductor Version 2 Mon Nov 14 142145 1994

I I I I

I I

I

I

I - - -

-Ishy-

-

Variable Displacement step 35 InelasticEvents Legend

Elastic Strut Buckling Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield Beam Clmn Fully Plastic Fracture

Project ST151K Model Conductormiddot

Version 1

Clgt - Cut Plane Force Fx Mon Nov 14 140051 1994

- -_ vii I fOIZM

0 _

~

N ID 0 bull 0

x 0

middotrt ~ () bull u 0 ~ 0 Ir

N bull

0

0 1 2 3 4 5 X Deck Disp [Ft)

I I 1 1 WbullTI- bullbull

lltcxi lr-C

011lt~~~~~~~---~~~~~~~~L~~~~~~~--~~~~~~~~i~~~~~~~--

Uniform Node Displacement

shy

Ji

r

Comparison of Pushover Curves for ST 151K Variable Load Pattern and Monotonic Load Pattern (Andrew)

Soll Improved 100

6000

5000 -i 4000

f

llS

3000 Ul

J 2000

10 c11 c1lgtolts 1000

0 v

bull ~

f

_ ----- --- ---shy

bull --Variable Lold

- - Andrew Load

I I I I I I

4 5 62 30 1

Deck Dlsplacement (ft)

bullbull r 1 II

~

ANDREW PHASE Il

WA VE FORCE PROFILE

bull DECK WA VE FORCE ~

PHASE I USED SECTION 17 (1056H degCd)

PHASE Il USE REVISED SECTION 17 (1000H Cd)

bull PHASE I USED SINGLE FORCE PROFILE FOR PUSHOVER

- USED WAVE HEIGHT CLOSE TO ULTIMATE CAPACITY

bull PHASE J

Il WILL STUDY THE

USE OF VARIABLE

PROFILE ~

- IF FORCE LEVEL INCREASES (OR DECREASES) BY A FACTOR EQUIVALENT TO A 2 FOOT CHANGE IN WA VE HEIGHT THE INCREMENTAL FO~CE PA1IERN IS CHANGED ACCORDINGLY (CAP FEATURE)

- INITIAL EXAMPLE

f~

a

middotn

a

I N -a a a amiddot

+

~ a

l

a w gt 0c 0 gt a I shyc -CJ wc w gtc(

== c( I shyZ w E W a 0 z- 8

ii sect

~ ~ bullT 0

I if

x ~

bullbullbull

II ~------------________

ANDREW PHASE II

WA VE FORCE PROFILEmiddot

CAP AUTOMATIC APPLICATION FEATURE

STAT

fR1 0 lt f lt1

STAT

STAT

f(R3-R2) middot f lt1

Project ST151K Model Wave_Loading Version 1 rue Nov 8 164107 1994 PT

E~ nss FT

SL 14amp FT

a 111 FT

EL 14 li

E-1 37 FT

El o FT

1middot1middot ______________ _ ________ __ _ __

0

200 400 600 800 1000 12000 Force at Elevatloii (kip)

200

a ll_

Wave Force at Major Elevations for Discrete Wave Heights

180

bullHmiddot40ft

Hmiddot495ft

H bull 51 ft

Hmiddot55ft_

H-59ft I

Andrew (H bull 60 86 ft) II

i I

_11---__

- middot- Hmiddot30ft

-

- bull -

middot bull bull bull middot

---Hmiddot53ft

bullbullbullbullHmiddot57ft

-----H-63ft

Elevation Above Mudllne 100

(ft)

180

140

120

80

60

40

20

Loading at Major Elevations

---H-49511

-deg-Hmiddot608611

0 200 400 600 800 1000 1200

Force It Elevatlon (kip)

--

Base Shear vs Wave Height ST151K

6000

5000 1---------Ult1mateCapacitybull 4890 kip (100 Improved Soll)-----------~--

4000a a lii ~ 3000

81 ampI

2000

1000

0 +-~~~--i~~~~~~~~~-+-~~~--lf--~~~~~~~~-+-~~~--l

30 35 40 45 50 55 60 65

Wave Height (ft)

bull middot

~

bull

II

2

g

~

It)

0

cCD E 0 E a c E euro~ CDshygtItgtmiddoto ~ti c -a Gic CD gt ~

-= c -Cll

I z

bullO 0i i ~ g 0

() CI -bull(q-ll) iuawow llu1wniiaAO

11

shy

116

114

112

g 110

bullc = 108

~ ~ 106 -sect 104

102

100

98

96

-

6555 60

Wave Height vs Center of Action ST151K

30 35 40 45 50

Wave Height (ft)

t

Comparison of Pushover Curves for ST 151K Variable Load Pattern and Monotonic Load Pattern (Andrew)

Soll Improved 100

--- --------- --- ----6000

5000

- 4000 8 - I

1 3000 Ul

81 m 2000

_J -shy

1000

---Variable Load

- - Andrew Load

0 f-~~~~~r--~~~~-t-~~~~~-t-~~~~~+-~~~~----t~~~~~-1

0 1 - 2 3 4 5 6

Deck Displacement (ft)

-middotshy

1 t I I ~

ANDREW PHASE II

BRACE MODELING

bull MARSHALL STRUTS USED FORPHASE I

bull PHASE II WILL ASSESS DIFFERENCE IN USING BUCKLING BEAMS AND MARSHALL STRUTS

middot bull COMPARE FOR 2D FRAME

bull COMPARE FOR PLATFORM 177 WHERE LOWER

BAY EXHIBITS LARGE BRACE MOMENTS

Project andrew Model beamframe Version 1 Mon Nov 14 145447 1994

~

middot Frame Test buckling beam-column model

-

Project andrew Model strutframe Version 1 Mon Nov 14 145754 1994

bull

Frame Test strut model

middot-middot _____________--_

_middotmiddotr_~ --ll--~-middot__

Buckling Beam bull

Strut (K=O 65)

Lateral Displacement

ANDREW PHASE II

FOUNDATIONS

bull ~LICIT NONLINEAR API FOUNDATION USED INiPHASE I

bull FOUNDATIONS

JIP SELECTED MINI-VANE DATA FOR

STRENGTH PROFILE

bull COMMENTS ON BENCHMARK COMPARISONS

- SIGNIFICANT VARIABILITY DUE TO FOUNDATION MODELING METHODS

II Ii

z

ctV 4x SPlO caissonfy-42s soil-api(lf-097step-114)

Inelastic Events Legend Elastic Strut Buckling

Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield

--------shyBeam Clmn Fully Plastic Fracture

~--~~~~~~~~~--~----~__i~~---~~~~__--~

Pro)ect apipile Model South Pelto 10 Version Wed May 18072235 1994 - -

1bullt

+

t

f i bullbullbull

bullbull bullbull

I I

5

~x SPlO Caissonfy=42s soil=api(lf=lOOstep=137)

Inelastic Events Legend

Elastic Strut Bucklilg

Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield

earn Clmn Full~ Plas1c Fracture

ANDREW PHASE II middot middot

JOINT MODELING

middotbull API RIGID-PLASTIC USED IN PHASE I

a ELASTIC-BRITTLE USED IN PHASE IB

amiddot MSL WILL PROVIDE NONLINEAR FORCE-DEFORMATION

AND MOMENT-DEFORMATION RELATIONS FOR PHASE II

a MODELING OPTIONS FOR PHASE II -

l

- UNCO~LEDmiddotONE DIMENSIONAL MEMBER END JOINT MODELING

J Ii

ANDREW PHASE Il

JOINT MODELING OPTIONS DISCUSSION

bull JD COUPLED INTERACTION SURFACE

- IDEAL

- DATA NOT AVAILABLE TO CALIBRATE

bull 2D COUPLED INTERACTION SURFACE

- SIGNIFICANT MODELING IMPROVEMENT

- DATAAVAILABLE

- IMPLEMENTATION PRACTICAL BUT NEEDS DEVELOPMENT AND CALIBRATION

bull 2D UNCOUPLED INTERACTION SURFACE

- SIGNIFICANT MODELING IMPROVEMENT

- DATA AVAILABLE

- IMPLEMENTATION PRACTICAL AND AVAILABLE

ANDREW PHASE II

JOINT MODEJJNG OPTIONS ILLUSTRATIONS

P

e

ANDREW PHASE II

JOINT MODELING PHASE II METHOD

e

I I I

ANDREW PHASE II

ANDREW STORM CONDmONS UPDATED

bull -PHASE I USED OCEANWEATHER HINDCAST-(Nov 1992)

bull middotmiddotPHASE II WILL USE THE REVISED HINDCAST (Nov 1994)

bull lt c lt I lt

2-2

l---4-~~~+-~~-+-~~~1--~~-+-~~--l--~T--+-~~--i24

f-tJVtMfiz~ 1Cfl (r STvb(

Plollol bullbull 5middotNOYmiddot9214bull1J Irbullbull Illbull [ANDREVl~NDREVCUSTbullMAXll 5-NOY-1992 10bull54 - 9208

HEIGHT lml

I~ MAXI

I L 1~middotamp SZ~- ~A I 13c

I J) ( r I X~0iWJ~0gtJ gt L gt gt I 1c I 120

-- N

~ deg

I I I I I I e C I ~=-41 122 -94 -92 -90 -88 -86 -84 -82 -80

-

Figure 7 contours of maximum hindcast significant wave height

_

~

~tl ~

NOltNttIJ 17lt ltNS S iJJ)y

Maximum Significant Wave Height - Hurricane Andrew 1992

30 ~

26 ~ I ~-~~~- == -shy -~ 35-_

24deg1--shy --------~- middot-middot-------middotmiddotmiddot----middot------middot-middot~middot- - - ~-----1---middot

22middot ~~~~~~~-~ -94 -92 -90 -88 -84deg-86 -82 -somiddot

c

-

Platform Selection bullbullbull1111111111111

0 BasisCriteria for Selection middot

0 Review of Phase I Platforms middot 0 Alternative Platforms

0 PMB Recommendations

0 Participants Comments

gt

L

l I

BasisCriteria for Selection bullbullbull1111111111111

0 Benefit for Calibration - Classification (UnexpectedExpected)

- Failure Mode (StructuresFoundations)

- Damaged StructuresKnown response

- Better Representation of Full ~opulation

0 Availability of Data - Existing Models

- Inspection Data

- Damage Data

PREDICTED BEHAVIOR

-

OBSERVED

BEHAVIOR

-SURVIVED DAMAGED FAILED

SURVIVED

ST134W WD90A

-MC311 -

MC397 -

--

-

-

ST151K ST130Q

-

-

-

DAMAGED - -

-

-

--

-

-

ST52 SS139 ST177B -

ST161A

-_

-

FAILED

-

-

-

ST151H ST130A ST72

Operator - Platfonns - Platfonns Survived Damagedamp

- Falled -

bull

-

-

Jle eBDIGIPllDbl -Amocoshy 30 1 Chevron -

143 12 Exxon 72 Mobll 22 1 Phllllps 7 bull

-Shell 39 Unocal 25 1

Total Partlclnts Platfonns 338 15

Other Operators bull- 310 13

-

Total Platfonns - -

648 28

fbml

Platforms Considered 6 7

tie of Total Platfonns 1 25

OVERVIEW OF PLATFORMS SURVIVEDDAMAGEDFAILED

-

shy

v

Platforms Evaluated in Phase I

-

-Platfonn Water Year Number Analysis Avallabillty of Availability Platform

- Name Depth rl of Model DetaDed rl ClassUlcatlon

Installation legs Inspection Soll Report Based on Information Phase I Calibration - -ft

- -- Suaill flalfnrm ~BSi -STISIK 137 1963 8 CAP Yes Yes Un eel Sarvlval

-STIJOO 170 1964 4 CAPmiddotshy Yes ST134data u ST134W 137 1981 4 CAP Yes Yesmiddot E

-

edSarvlval -

middotsurvival

WD90A 184 - 1964 8 CAP Yes E -

- Sarvlvill

MC3ll 343 1978 8 KARMA

MC397 468 1991 4 KARMA -

bull ~reglmmiddot- -

Yes Sure Survival

bull Yes - Sare Sanlval-

-T23STS2 63 1969 4 CAP

T2SSS139 62 1969 4 CAP -

Yes

Yes

Yes Ex

Yes Une

bull SurvivalLikely to DamalI

middot bull SanlvalLlkely to DamalI

ST161A 118 1964 8 USFOS

Eail11a pound1al(bullllm Clss ST177B 142 --1965 8 CAP

STISIH 137 1964 - 8 CAP C

STIJOA 140 19S8 8 CAP - -

Yes

Yes

ollamed

Yes

- Yes E

-

at about 1 mUe

Yes

ST134data

bull SafvlvalLlkelv to Dam- bull- -

E middotrahre

Emeried Failure -

Likely FaUare

T21ST72 61 1969 4 CAP Collamed No Likely FaUure

Platforms Identified for Evaluating Their Likelihood of Falling Under UnexpeCted Survival Category

CllEllATOa

AllB

llLOCll

Sl1lvcnJllE

NAME

YEAR

INSTAUJCD

WATER

IJBPJB

middotmiddotDISTANCE

ftOMSllOllE

(MIUIS)

Ill

(M)

AMOCO BllW

STllI

WDI075

A

bull D

4

n uo 17J

44

32

bull 20

10

bull CllEVllOH STtllO c I 1bull 21 11

STl130 D 0 161 21 11 STllJ4 I 0 137 D - 11

STIUI G I 137 32 11

STIUI I 0 121 32 11

STIUI

J 0 140 32 11

STIUI It 4 137 32 11

STllSI PROl)I 0 137 32 11

STllSI PROD-2 0 137 32 11

STll7 A 0

140 35 11

WDl117 c 65 214 34 10

DXON STshy E 51 14 bull STl165 A t3 bull 34 WD0073

WD0073

WD0073

A _ c

4

62

4

IA IA

17J

22

22

22

bullbullbull WD003-shy E 65 161 Z7

A 0 IN 31 MOBD Bl12I

ssn A bull

50

50

31

7

bull 7

MIJllPllY SS114 B 7 50 7 7

PlllUIPS SSl14t A 55 33 7

SAMEDAN ST17J A 4 117 7 bull 7

SJIELL WD0103 A 65 223 27 WDllOJ bull 65 221 27 WDll04 c 65 221 27 WDOI04 D 67 2 27

UNOCAL

SSl20I

SSl20I

bull SSl20I

SS0209

SS0209

SS0209

SS0209

SS 0253

SS 0253

E

I

PtJllE

A

bull D

G A

c

0

4

4

7J

61

0

0

69

103

97

97

9S

100

95

100

165

175

34

34

34

34

34

34

34

54

54

bullbullbullbullbullbullbullbullbull

Platforms Recommended for Phase II

llalr y_ w- Aot Av- _ I 1 Doplla Model nm bull bull _ bull I I H11 s18-1 Irht lbwl~ bullbullbull(-illtdlel OM lllMtD

II ---

-STISIIC 137 1963 I CAP Yeo Yeo I Yeo

STtshy -170 1964 4 CAP Yeo ST134dlU I Yeo

STl34W 137 Yeo Yeo -1981 4 CAP D Yeo

WD90A 114 1964 I CAP Yeo - D

MC311 30 1971 I KARMA Yeo -sre-

MC397 middot 461 1991 4 ICARMA Yeo Sireshy - - r- -

__ T13ST5l 63 1969 4 CAP Yeo Yeobull middot- ~n- D Yeo _ Tl5SSl39 62 1969 4 CAP Yeo Yeo I Yeo

_ STIT711 142 1965 I CAP Yeo I Yeo middot-1shySTl61A Ill 1964 I USFOS Yeo Yeo - D bulloshy

- r-

STISIH 137 1964 I CAP y D y -_ STl31A 140 1951 I CAP Yeo STl34dlU I Yeo

T21Sll 61 1969 4 CAP No rar-middot I Yeo

~-rmiddot-middot SS209A 95 1967 16 CAP Yeo D -SSU4R

11 UI A 41 -in D SS 2741AIM CAP bull

_ _ -shy - -shy- -shy

CALIBRATION

bull PHASEIAPPROACH

bull CALIBRATION ISSUES

bull PHASE II WORK SCOPE

bull SOFlWARE AVAILABILITY

bull EFFECT OF ADDITIONAL PLATFORMS

bull WEIGHTING APPROACH

bull EXAMPLE APPLICATION middot

(l=Ofce Mode] Wind

~

fcurrent l -middot

~SW ~leldl4

ultimate resistance force I first

design load Illgt

i _

deflection

-

bull

4

Historical Hurrlcane Dariiage and Survival of Platforms

Components Required for Calibration (Petrauskas 1992)

PRIOR Distribution

PLATFORM EXPERIENCE

BAYESIAN Survivals and Failures

UPDATING

Waves Winds and Currents bull

POSTERIOR Distribution

- A_middotJA ~ Survival Data Failure Data

Result Result

Load Resistance

middotfl

-

Bayesian Updaing middotGeneral Description (Petrauskas 1992)

L Imiddot

WHERE TO INTRODUCE BIAS FACTOR (B

On Resistance (Rmue - B (RPREDICreD I

middotOn Loads (Smue middot - B (SPRE01cTEo

On Safety Margin (RSmue - B (RSPREDICTED

CALIBRATION METHODOLOGY

gt

r

FIELD OBSERVATIONS

Observed Behavior

survivals Damages

Failures

r

bull

CAPACITY ANALYSIS

Predicted Load and Resistance Predicted Failure Modes

Survival Damage Failure cases

RELIABILITY ANALYSIS

- Probability of Failure ~

Ukellhood Failure

BAYESIAN UPDATING

Prior Distnbution of B Combined Likelihood Function

Posterior Dlstnbution of B

BIAS FACTOR

Mean Value cov bull

Represents Modeling Uncertainties

Prior Distribution of B Likelihood Functionmiddot Observing Success case STISIK and Posterior Distribution of B

2 00

uo

160

140

120

e 100 L $ uoIshy

0 60 ~ ~ 040

020

0 00

0 00 0 50

_ I 00

b

I 50 2 00

-shy lk [bl STISIK)

--shy Pnor-f(b)

--shy Pos1enor r(bl ISIK)

Prior or B Mean bull 10 COV bull 030

Posterlor of Bbull Meanbull 136COV 0 16

2 50

Likelihood Function and Posterior Distribution Example Platform STISIK

- -

~

- Prior and Posterior Distributions or B Failure cases (ST177B STlSlH ST130A T21) Damage cases (T23 T25 ST161A) Success cases (STlSlK ST130Q ST134W WD90A MC311 MC397)

400

350

300

2 so 0-

I

bull bull

bull bull gt

- bullbull ~ bull bull I bull

_ bull bull bullbull I bull bull bullII V -

bull bull

I bull ~ bull

bullbull bullI bull bullI bull bull bull

middot -~

bull__ ~ middotI middotmiddot

L 2 00

e ~ I SO

-I 00

- 0 so

000 gt

0 00 020 0 40 060 0 80 I 00 120 140 160 110 200

b -

-

shy

Prior-r

- - - - Posterior fLlll failure cases

bull bull bull bull bull bull bull Posterior flall damage cases

- bull bull - bull middot Postenor flall success cases

Posterior flall success fulure damage cases

-

Posterior of B Mean= 119 COV 0101

Posterior Distribution or Blas Factor (8) bull Different Categories and All Cases Combined

shy

l

PHASE I - KEY CONCLUSIONS

bull GLOBAL CORRECTION FACTOR

bull 15 TO 20 CONSERVATISM MEAN)

bull UNEXPECTED SURVIVALS HAVE GREATEST IMPACT

bull DAMAGED CASES MORE INFORMATIVE

bull CALIBRATION ISSUES PHASE II

bull

EFFECT OF IMPROVED CAPACITY ANALYSIS RECIPE

bull APPLICABILITY TO MULTIPLE FAILURE MODES

bull NO FOUNDATION DAMAGE CASE INCLUDED

bull IMPROVED REPRESENTATION OF FULL ANDREW DISTRIBUTION

-bull NUMBER OF PLATFORMS CONSIDERED

I 1 I

I t ) bull ~

PHASE II WORK SCOPE (CALIBRATION)

bull

bull ESTABLISH MULTIPLE BIAS FACTORS

Use Nine Base Case Platforms

middot - Capacity Analysis for Revised recipe

Foundation Blas Factor from Foundation JIP

bull DEVELOP WEIGHTING APPROACH

bull UPDATE MULTIPLE BIAS FACTORS

bullEffect of Platforms Not In the Nine Base Cases

bull EXAMPLE APPLICATION OF BIAS FACTORS I

bullbull

PHASE I PHASE II

CAPACITYbull ANALYSISRu bull

---------------------1----middot

Jc1 - iJbullraquoJ dbull

FttlS frt[f~l ratdhJJ

-

LIKELIHOOD FUNCTION

DEVELOPMENT

lk Jb Jfall1N) bull Jb)

111 Ill Isurvtvoq bull 1 - bl

-middot -------------------shy ~---------------------middot-middot

r111ir8 Cb1 111 Cbl BAYESIAN

UPDATING

~cove

R1R2 bull middot---------------------~----

Jblbull P(g cOJ

1 bull JbRI (MBSI I bull 1

~middot

llcJllJfalluro) bull P(gcOJ Ill 111 b21fallun) bull P ((g1 CO) U 1112 c OD

middot--------------------shy ----------------------shy bull

r111111r8 Cbl 111 Ill i-1nro1

deg ~1b2)laquottfrr (b1b21-lnfo) re Ill bull 1-11middot b2) bull lodegamp

middot

i1 covamp i2 cov~i

-

shy

--

Likelihood Function vs b All Success Case STISI K STl30Q STl34W WD90A MC311 MC397

I 0000

0 9000

OIOOO

0 7000

~ 0amp000

ta osoOO

~ 0 4000

0 3000

0 2000

0 1000

bull 0 0000

bull

I

middot

middot bull I

Ibull

-sibull ~ -middot-middot-middot-middot -middot-

----7----+-----~ I

Ibull bull bull I

rbull

--- Jk ISIK[blsuccess)

- bull - bull - lk 130Q[blsuccess) bull

lk 134W[blsuccess)

Jk-WD90A(blsuccc~I

- - - lk MC397[blsucccss)

--ltgt-- lk middotMC31 l[blsucccss)

--- Jk [bl 6 out of6succcsHascs)

05 I 5 2 250

b

Likelihood Functionsmiddot Success Cases - Phbullsc L

shy

bullbull bull bull bull bull bull -------------------~

bull bull bullbull bullbull bullbull bullbull bullbull bullbull bull

bull bullbull

bullbullbullbull

bullbull

bullbullbullbullbullbullbull

bullbullbullbullbullbull

bullbullbullbullbullbullbull

z 0

t z ) LL

c 0 0c i ~ i

J

~ sect ~--------------~ e l5

I

I

SOFTWARE

bull PF PROGRAM

bull CALREL

bull PF +CALREL

bull RELACS

bull

gt i

l

SENSITIVITY STUDIES

POSTERIOR DISTRIBUTION OF B

bull EFFECT OF CHANGE IN DISTRIBUTION OF PRIOR OF B

bull EFFECT OF REMOVAL OF ONE PLATFORM FROM EACH CATEGORY

bull EFFECT OF REMOVAL OF THREE DAMAGE CASES

bull EFFECT OF ADDITIONAL 13 SIMILAR PLATFORMS

-I

Prior and Posterior Distributions of B Variation In COV of PriorofB fromOl to04 -- -

4500

-4000

-

- -

3500 --middot - bull bull - bull bull Prior-r ICOV =02

3000

- - - - Prior-r ICOV =03 -

bull -~ e 2soo -

bull bull bull bull bull bull bull Pnor-r ICOV =o 4t frJ ~ 2000 Posterior rlCOV =0 2

bull bull Postenor rlCOV =0 3 bull bull 1500 bull middot bull - shy -middot- Posterior rlCOV =04

middot~ I 000 - bull

-middotmiddot ~ ~middotmiddot M -~

- middot 0 500

wbullbullbull _ I middot~ middotmiddotmiddotmiddotmiddot middotmiddotmiddotmiddotmiddot - -~ ~ __ middot_ _ bull -shy0 000

0 000 0200 0 400 0 600 0 800 IOOO 1200 1400 1600 1800 2000 -

- - b

-

=shy

Posterior Distribution of Bias Factor (B) bull Effect of Variation In COV of Prior of B

bull

bullPrior and Posterior Distributions of B Evaluating Effect of Removal of One Platform from Each Category from the Base Case of 13 Platforms

400

~ ~ 350

j IA i

3 00 bull I bull I ---Prior-r I bull l

250 1 I 1

I bull bull ---Posterior r lall 13 casesshy- ~E 2 00 bull bull bull bull bull bull bull Posterior r without STISIK

- ~ 1 bull - bull I - bull bull - bull bull Posterior r Iwithout T2I I 50 I

J-tt~i - bull - bull - Posterior r Iwithout T2S

tOO [-

I I ~ ~ I 0 50

0 00

___ r J i

l~T

~--

0 00 0 20 0 40 0 60 080 I 00 120 140 160 110 2 00

b

zshy

Posterior Distribution of Bias Factor (8) bull Effect of Removal of middotPlatform from Any Category

bullPrior and Posterior Distributions or B Evaluating Effect or Removal or 3 Damage Cases from the Base Case or 13 Platrorms

4

35

3

25

e L 2 ~ bulle shy

I 15

bull I

05

0

0 02 04 06 I 2 I 4 16 IS 2

I I

I

I

I Ibull

J -tmiddot L--

bullI

~

I

fl

II ~

I I v I

middot bull bull

bull L

~ ll- I-~lt

OS I

b

I

bull

1middot bullbullbullbull

~ __

bull

- - - - Prior-r

- - - - Postenor flall failure cases

Postenor flall success cases

--- Posterior flall 13 cases

---Posterior fI success failure cases only

IPosterior or B Mean =t11 COV =0121

-

_I

Posterir Distribution or Bias Factor (B) bull Effect or Removal or 3-Damage Cases

-

bullPrior and Posterior Distributions or B Evaluating Effect or Additional 13 Similar Observations as

Used in Andrew JIP The POsterior or B from Andrew-JIP Considered as New Prior or B

- 500

450

400

350

bull bull

bull bull bull bull J bull

bull bull f ~ I

I ~ Ibull bull

bullbull bull

l bull I -bullbull bull bull

I

I bull bull

bull I I~~bull bull bull

bull ~

I bull bull

~ ~_ ~~~

- - New Prior r bull

Posterior 114 new failure cases ~ 300- bull bull bull bull bull bull bull Posterior f 13 new damage casesl 150 ~

- bull bull - bull bull Posterior 116 new success cases i 2 00

Postenor 1113 new success failure -150 damage cases

I 00

0 50 -

0 00 0 00 0 20 0 40 0 60 0 80 I 00 120 I 40 160 110 2 00

Posterior ol B Mean= 120 COV= 0071 b

shy

shy

_rshy

Posterior Distribution of Bias Factor (B) bull Effect of Additional 13-Similar Observations

1

1 I

WEIGHTING APPROACH bull

bull ASSESSMENT OF PLATFORMS DATA

Select Platforms Most Mappable to the Analyzed Platforms

Unlikely SurvivalDamage Cases Identification

Based on Cut-Off Criteria Agemiddot Hs etc

bull INFORMATION REQUIRED FOR COMPARISON

bull CATEGORIZATION OF ADDITIONALmiddot PLATFORMS

bull UPDATING OF CALIBRATION RESULTS

--

OVERVIEW OF PLATFORMS SURVIVEDDAMAGED -~

Operator -

- Platforms

Survived CS)

Platforms

Damaged CD)

Platforms Used In

Phase IJIP -

JI~ ~Bill~IRBDll -

Amoco -

Chevron -

Exxon

Mob II

Phllllps

Shell

Unocal -

30

143

72 22

7

39 25shy -

1

12

bull

1 bull

1

1(S)+1 (D)

3 (S) +3(D)

2 (S)

bull shy

bull

bull

bull

Total Partlclnants Platforms 338 - 15 10C6S+4D) -

Other Operators

310

13

3 (D)

Total Platforms 648 28 13C6S+7D) -

Summary of Andrew JIP bull Phase I Platforms

Plairshy w v- N-ber Lea-le 1 shy car Hind_ o-w N-bull Deplh fl fl IJpe la bull n Yeadon Jadtet lbelledtf liefalit -- _ -OD

IL H - -shy _ICllshy IL IL - - rQClto ~ _

K STUIK 137 1963 8 365 608S4 Cellardeck

-middot STl-shy 170 1964 4 -~ K 39 62269 Cellardeck - --ST134W 137 1911 4 43 608S4 Smvlnlshy- _ - -

WD90A 184 1964 8 K 267Sbull34 83 SOS31 SUboftllar deck SmvMI

MC311 343 1978 8 x S7 62 - MC397 468 1991 4 x shy S33bull6SO 65911 -shySmeSmvlvol -- gt -

r-- T23ST52 63 1969 4 ton- shyKshy 52594 S03 -T25SSl39 62 1969 4 Kshy SOS94 S063 - middot ton-shy__ -

bullbull a ton--shyST161A 118 1964 8 K - 347S 51495 Cellarclecti

r~

ST1771 142 1965 8 K 39 60167 CeDordeck failme

STISIH 137 1964 8 K 395 608S4 CeDardeck Failme

K ST130A 140 1958 a SS 6095 I lblv Pai1uno - T21ST72 61 1969 4 Kshy S4S94 4974 I lhlv Pai1uno -

- -

JE

INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON

bull WATER DEPTH

bull bull YEAR OF INSTALLATIONDESIGN

bull PLATFORM ORIENTATION

bull PLATFORM LATITUDE AND LONGITUDE

bull KEY SKETCHES OF PLATFORM STRUCTURAL FRAMING

Number of Legs Plies Leg Batter Number of Conductors bull Deck Elevations (Including sub-cellar deck If any) Jacket Framing Elevations Brace type In Vertical Frames Sizes of Primary Members (leg pile vertical frames braces primary horizontals) Joint Details (can gap etc) LegPiie Annulus Grouting Status

bull middotPOST ANDREW INSPECTION REPORT Significant Wave Height Hs (ft)

bull INFORMATION ON DAMAGES PRIOR TO ANDREW

bull RESULTS FROM IN-HOUSE STUDIES IF AVAILABLE

bull ~RIGINAL DESIGN CRITERIA INFORMATION IF AVAILABLE

bull SOIL SHEAR STRENGTH PROFILE IF AVAILABLE

Platforms Identified for Evaluating Therr Likelihood of Falling Under Unexpected Survival Category

OPlRATOR

middotADA

BLOClt

llmucnJllE

NAME

YEAll

INSTAUJD

WA1ER

DBPTll

DISTANCE

JlllOM SHOllE

(MILU)

Bl

(M)

AMOCO 111113

STl11

WDI075

A

bull D

St

17

120

1n

44

20

10 I

CllllVllQfl STlllO c 61 110 21 11 STIUO D Q 1110 21 11

STl134 p Q 137 29 11

ST0151 G 61 137 11

ST0151 I 3 121 11

ST0151 J Q 140 11

STl151 It 137 11

ST0151 PJl001 Q UT 11

ST0151 PROD-2 Q UT 11

ST017 A 3 140 35 11

WDIUT c 65 214 34 10

llXXON STI055 II 51 6T 14 I

STl15 A t3 34

WDI073

WD0073

Amiddotshy Q

1111 1111

22

22

I

I

WDI073 c 1n 22 I

- WDOCl93-shy II

65 1110 2T A 3 194 31 MOBU 111121

SS IOl7

A B

501

50

50

31

T

I

T 6

MtlllPllY SS114 B T 50 T T

PlllllIPS SS0149 A 5 55 33 T

SAMllDAN STln A 107 T T

SBllLL WDl103 A 65 223 2T WDl103 B 65 221 2T WD0104 c 65 221 2T WD0104 D 6T 20 2T

UNOCAL SS020I

SS 020I SSl20I

SS 0209

SS0209

II p

PURll

A

B

3

n 61

103

97

97

95

100

34

34

34

34 34

bullbull bullbullbull

- SS0209 D u 95 34 I

- SS 1209 SSl253

SS0253

G A

c

6T

2

100

165

1T5

34

54

54

bullbull I

I I

I I

SOUTH MARCH

~LAND

N

SOUTH PaTOI ENE ISWltD AREA

MMS Defined Limits ror Offshore Structures Requiring Inspections Following Andrew

t---Jbulln3 T25bull I bullTtl

bull1$1Htl1Abull bull1SIK1779bull

~

middot-DaTA AREA~T

CRANO rsLE

AREA bull1300

bullll4W

=shy

bullPath of Hurricane Andrew with Platforms Used In Calibration

Likelihood Function vs b All Success Case STISI K STIJOQ ST134W WD90A MCJll MCJ97

I 0000

0 9000

0 8000

07000

-Bo6000

tiiii 0 5000 -

04000

0 3000

0 2000

01000

0 0000

0 05 25

ri1rt middot-middot-middot-middot-middot-

I 5 2

b

---11 middot151K(blsu~ccsI

- bull - bull - fk 130Q(bfU~lCSS)

lie bull134W(blsuccc]

lk-WD90A[blsultCS]

- - - lk middotMC397(bfUtlCS]

--o-- lk-MCJll(bbuctcsJ

--- ll (bl 6 out of 6 Mlttcs case]

_

( Likelihood Functions middotSuccess Cases - p - 1 c I

CATEGORIZATION OF ADDITIONAL PLATFORMS

I I I

bull CATEGORIES

Unexpected Survivals

Expected Survivals

bull APPROACH

Identify Corresponding Platform(s) Evaluated In the JIP

Determine Max Wave Height at Platform Location

Evaluate Whether Wave Hit the Deck

middot - Evaluate Likelihood of R and S to Differ from the Base Cases

Review Joint Information

Assess Special Features

bull IDENTIFY NUMBER OF PLATFORMS IN EACH CATEGORY

I

UPDATING OF CALIBRATION RESULTS

bull INCLUDE EFFECT OF ADDITIONAL PLATFORMS ON bullBbull

Consider Increased 1Uncertalntles In R and S

Develop bullNeW Ukellhood Functions for Different Categories

middotLet bullxbull Number of Ukely Unexpected Survival Cases and y Number of Ukely Expected Survival Cases

Determine Combined Likelihood Function lk (b Iadditional platforms) middot

bull REVISED LIKELIHOOD FUNCTION

lk(b I all cases) = lk(b I 9 platforms) middot x lk(b Ix platforms) x lk(b I y platforms)

bull ESTABLISH REVISED BIAS FACTOR B

f8 (b) u f8 (b) lk (b I all cases)

II

t t ~

shy

APPLICATIONS OF BIAS FACTORS

ii APPLICABLE TO PARTICULAR LOAD AND RESISTANCE RECIPE

bull USE BY THE API COMMITTEES AND MMS

Criteria Revisions

Recommended Design Guidelines Revisions bull

bullbull REASSESSMENT STUDIES AND DECISION MAKING

middot

bull RISK ASSESSMENT STUDIES BY THE OPERATORS

Individual Platforms

Fleet of Platforms

middot IMR Program Implementation

bullbull NOT TO BE USED IN DESIGN

I I fl J

EXAMPLE USE OF BIAS FACTORS

bull PHASE I EXAMPLE APPLICATION

bull EVALUATION OF ANNUAL PROBABILITY OF FAILURE

Establish Annual Maximum Seastate Data

Determine Coefficients to Define Base Shear

Determine Ultimate Capacity of Platform

Determine Probablllty of Failure Versus bullbbull

-bull DETERMINE UPDATED PROBABILITY OF FAILURE

Pf - P [S gt R]

Pf I P [~ gt Rb I b] f 8 (b) db

Pr - J P [S gt Rb I b] f bull 8 (b) db

bull TOTAL PROBABILITIES OF FAILURE

Total Pf - 1 bull Il [1 middotPf] - l Pf bull

Total Pf - 1 bull Il [1middot Pf] - l Pfmiddot

Total Pr = 1 - n [1middot PfbullJ = l Pfshy

I I I

EXAMPLE APPLICATION bull PHASE I

PLATFORM ST 151 K

bull GULF OF MEXICO

bull WATER DEPTH = 137 FT

bull ANNUAL MAXIMUM HS MEDIAN 175 ft COV 031

bull ASSUMED STORM DURATION 3 hoursmiddot

bull ASSUMED NUMBER OF WAVES 800

bull MEDIAN HS FOR STORM HOURS 1 2 AND 3

bull MEDIAN CURRENT VALUE bull

bull DETERMINE ULTIMATE CAPACITY FOR THREE DIRECTIONS

bull DETERMINE ANNUAL FAILURE PROBABILITIES

Broadsidemiddot Diagonal End-on

0037 0025 0007

0071 0044 0018

0022 0016 0004

DETERMINE TOTAL ANNUAL PROBABILITY OF FAILURE

Pf - 007

Pf - 013

Pr - 004

bull

EXAMPLE USE OF BIAS FACTORS - PHASE II middot

bull DETERMINE PREDICTIVE PROBABILITY OF FAILURE

Case - 1

Pf _ S S Pf (b b2) fa1a2 (b b2) db db2

g bull -1 (b1 R I S) - 1

92 - (b2 R2 I S) - 1

Case -2

Pf s ~ [bJ f11 (b) db

Pf [bJ p [g lt 0)

g - bull 1 (b1 R1 I S) - 1

1

bull1) l

I

BENEFITS OF ANDREW JIP bull PHASE II

lt

bull IMPROVED LOAD AND RESISTANCE RECIPE

bull IMPROVED CALIBRATION PROCEDURE

bull MULTIPLE BIAS FACTORS

bull PROCEDURE TO INCLUDE EFFECT OF OTHER PLATFORMS

bull PROCEDURE TO USE MULTIPLE BIAS FACTORS

  • Structure Bookmarks
    • ANDREW PHASE II JOINT1 INDUSTRY PROJECT
    • PROGRESS MEETING NUMBER 1
    • ANDREW PHASE II JOINT INDUSTRY PROJECT PROGRESS MEETING NUMBER 1 NOVEMBER 16 AGENDA
    • ANDREW PHASE II UST OF PARTICIPANTS
    • ANDREW PHASE I OBJECTIVES AND RESULTS
    • ANDREW PHASE II PROJECT OBJECTIVES
    • SUBSEQUENT PROJECTS
    • ANDREW PHASE II PROJECT TEAM
    • ANDREW PHASE II UPDATE MODELING middotAND ANALYSIS RECIPE
    • -ANDREW PHASE II CONDUCTOR RESISTANCE
    • ANDREW PHASE Il WA VE FORCE PROFILE
    • ANDREW PHASE II WA VE FORCE PROFILEmiddot
    • ANDREW PHASE II BRACE MODELING
    • ANDREW PHASE II FOUNDATIONS
    • ANDREW PHASE II middot middot JOINT MODELING
    • ANDREW PHASE Il JOINT MODELING OPTIONS DISCUSSION
    • ANDREW PHASE II JOINT MODEJJNG OPTIONS ILLUSTRATIONS
    • ANDREW PHASE II JOINT MODELING PHASE II METHOD
    • ANDREW PHASE II ANDREW STORM CONDmONS UPDATED
    • Platform Selection bullbullbull1111111111111
    • BasisCriteria for Selection bullbullbull1111111111111
    • CALIBRATION
    • WHERE TO INTRODUCE BIAS FACTOR (B
    • PHASE I -KEY CONCLUSIONS
    • bull CALIBRATION ISSUES PHASE II
    • PHASE II WORK SCOPE (CALIBRATION)
    • SOFTWARE
    • SENSITIVITY STUDIES POSTERIOR DISTRIBUTION OF B
    • WEIGHTING APPROACH bull
    • INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON
    • CATEGORIZATION OF ADDITIONAL PLATFORMS
    • UPDATING OF CALIBRATION RESULTS
    • APPLICATIONS OF BIAS FACTORS
    • EXAMPLE USE OF BIAS FACTORS
    • EXAMPLE APPLICATION bull PHASE I PLATFORM ST 151 K
    • EXAMPLE USE OF BIAS FACTORS -PHASE II middot
    • BENEFITS OF ANDREW JIP bull PHASE II
Page 14:  · . ' ANDREW PHASE II PROJECT OBJECTIVES . ' • ·UPDATE MODELING AND ANALYSIS RECIPE· JOINTS ' ' ·BRACES WAVE PROFILE FOUNDATIONS . ' ' CONDUCTORS· • ANALVZE PLATFORMS USING

Project STlSlK middot Model Conductor Version 2 Mon Nov 14 142025 1994

-----------------

bullf

I

I

I

II I I

I I I I

I ==

-

Variable Displacement step 9 Inelastic Events Legend

Elastic Strut Residual

bull

Strut Buckling Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield Beam Clmn Fully Plastic Fracture

--

Project ST151K Model Conductor Version 2 Mon Nov 14 142145 1994

I I I I

I I

I

I

I - - -

-Ishy-

-

Variable Displacement step 35 InelasticEvents Legend

Elastic Strut Buckling Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield Beam Clmn Fully Plastic Fracture

Project ST151K Model Conductormiddot

Version 1

Clgt - Cut Plane Force Fx Mon Nov 14 140051 1994

- -_ vii I fOIZM

0 _

~

N ID 0 bull 0

x 0

middotrt ~ () bull u 0 ~ 0 Ir

N bull

0

0 1 2 3 4 5 X Deck Disp [Ft)

I I 1 1 WbullTI- bullbull

lltcxi lr-C

011lt~~~~~~~---~~~~~~~~L~~~~~~~--~~~~~~~~i~~~~~~~--

Uniform Node Displacement

shy

Ji

r

Comparison of Pushover Curves for ST 151K Variable Load Pattern and Monotonic Load Pattern (Andrew)

Soll Improved 100

6000

5000 -i 4000

f

llS

3000 Ul

J 2000

10 c11 c1lgtolts 1000

0 v

bull ~

f

_ ----- --- ---shy

bull --Variable Lold

- - Andrew Load

I I I I I I

4 5 62 30 1

Deck Dlsplacement (ft)

bullbull r 1 II

~

ANDREW PHASE Il

WA VE FORCE PROFILE

bull DECK WA VE FORCE ~

PHASE I USED SECTION 17 (1056H degCd)

PHASE Il USE REVISED SECTION 17 (1000H Cd)

bull PHASE I USED SINGLE FORCE PROFILE FOR PUSHOVER

- USED WAVE HEIGHT CLOSE TO ULTIMATE CAPACITY

bull PHASE J

Il WILL STUDY THE

USE OF VARIABLE

PROFILE ~

- IF FORCE LEVEL INCREASES (OR DECREASES) BY A FACTOR EQUIVALENT TO A 2 FOOT CHANGE IN WA VE HEIGHT THE INCREMENTAL FO~CE PA1IERN IS CHANGED ACCORDINGLY (CAP FEATURE)

- INITIAL EXAMPLE

f~

a

middotn

a

I N -a a a amiddot

+

~ a

l

a w gt 0c 0 gt a I shyc -CJ wc w gtc(

== c( I shyZ w E W a 0 z- 8

ii sect

~ ~ bullT 0

I if

x ~

bullbullbull

II ~------------________

ANDREW PHASE II

WA VE FORCE PROFILEmiddot

CAP AUTOMATIC APPLICATION FEATURE

STAT

fR1 0 lt f lt1

STAT

STAT

f(R3-R2) middot f lt1

Project ST151K Model Wave_Loading Version 1 rue Nov 8 164107 1994 PT

E~ nss FT

SL 14amp FT

a 111 FT

EL 14 li

E-1 37 FT

El o FT

1middot1middot ______________ _ ________ __ _ __

0

200 400 600 800 1000 12000 Force at Elevatloii (kip)

200

a ll_

Wave Force at Major Elevations for Discrete Wave Heights

180

bullHmiddot40ft

Hmiddot495ft

H bull 51 ft

Hmiddot55ft_

H-59ft I

Andrew (H bull 60 86 ft) II

i I

_11---__

- middot- Hmiddot30ft

-

- bull -

middot bull bull bull middot

---Hmiddot53ft

bullbullbullbullHmiddot57ft

-----H-63ft

Elevation Above Mudllne 100

(ft)

180

140

120

80

60

40

20

Loading at Major Elevations

---H-49511

-deg-Hmiddot608611

0 200 400 600 800 1000 1200

Force It Elevatlon (kip)

--

Base Shear vs Wave Height ST151K

6000

5000 1---------Ult1mateCapacitybull 4890 kip (100 Improved Soll)-----------~--

4000a a lii ~ 3000

81 ampI

2000

1000

0 +-~~~--i~~~~~~~~~-+-~~~--lf--~~~~~~~~-+-~~~--l

30 35 40 45 50 55 60 65

Wave Height (ft)

bull middot

~

bull

II

2

g

~

It)

0

cCD E 0 E a c E euro~ CDshygtItgtmiddoto ~ti c -a Gic CD gt ~

-= c -Cll

I z

bullO 0i i ~ g 0

() CI -bull(q-ll) iuawow llu1wniiaAO

11

shy

116

114

112

g 110

bullc = 108

~ ~ 106 -sect 104

102

100

98

96

-

6555 60

Wave Height vs Center of Action ST151K

30 35 40 45 50

Wave Height (ft)

t

Comparison of Pushover Curves for ST 151K Variable Load Pattern and Monotonic Load Pattern (Andrew)

Soll Improved 100

--- --------- --- ----6000

5000

- 4000 8 - I

1 3000 Ul

81 m 2000

_J -shy

1000

---Variable Load

- - Andrew Load

0 f-~~~~~r--~~~~-t-~~~~~-t-~~~~~+-~~~~----t~~~~~-1

0 1 - 2 3 4 5 6

Deck Displacement (ft)

-middotshy

1 t I I ~

ANDREW PHASE II

BRACE MODELING

bull MARSHALL STRUTS USED FORPHASE I

bull PHASE II WILL ASSESS DIFFERENCE IN USING BUCKLING BEAMS AND MARSHALL STRUTS

middot bull COMPARE FOR 2D FRAME

bull COMPARE FOR PLATFORM 177 WHERE LOWER

BAY EXHIBITS LARGE BRACE MOMENTS

Project andrew Model beamframe Version 1 Mon Nov 14 145447 1994

~

middot Frame Test buckling beam-column model

-

Project andrew Model strutframe Version 1 Mon Nov 14 145754 1994

bull

Frame Test strut model

middot-middot _____________--_

_middotmiddotr_~ --ll--~-middot__

Buckling Beam bull

Strut (K=O 65)

Lateral Displacement

ANDREW PHASE II

FOUNDATIONS

bull ~LICIT NONLINEAR API FOUNDATION USED INiPHASE I

bull FOUNDATIONS

JIP SELECTED MINI-VANE DATA FOR

STRENGTH PROFILE

bull COMMENTS ON BENCHMARK COMPARISONS

- SIGNIFICANT VARIABILITY DUE TO FOUNDATION MODELING METHODS

II Ii

z

ctV 4x SPlO caissonfy-42s soil-api(lf-097step-114)

Inelastic Events Legend Elastic Strut Buckling

Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield

--------shyBeam Clmn Fully Plastic Fracture

~--~~~~~~~~~--~----~__i~~---~~~~__--~

Pro)ect apipile Model South Pelto 10 Version Wed May 18072235 1994 - -

1bullt

+

t

f i bullbullbull

bullbull bullbull

I I

5

~x SPlO Caissonfy=42s soil=api(lf=lOOstep=137)

Inelastic Events Legend

Elastic Strut Bucklilg

Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield

earn Clmn Full~ Plas1c Fracture

ANDREW PHASE II middot middot

JOINT MODELING

middotbull API RIGID-PLASTIC USED IN PHASE I

a ELASTIC-BRITTLE USED IN PHASE IB

amiddot MSL WILL PROVIDE NONLINEAR FORCE-DEFORMATION

AND MOMENT-DEFORMATION RELATIONS FOR PHASE II

a MODELING OPTIONS FOR PHASE II -

l

- UNCO~LEDmiddotONE DIMENSIONAL MEMBER END JOINT MODELING

J Ii

ANDREW PHASE Il

JOINT MODELING OPTIONS DISCUSSION

bull JD COUPLED INTERACTION SURFACE

- IDEAL

- DATA NOT AVAILABLE TO CALIBRATE

bull 2D COUPLED INTERACTION SURFACE

- SIGNIFICANT MODELING IMPROVEMENT

- DATAAVAILABLE

- IMPLEMENTATION PRACTICAL BUT NEEDS DEVELOPMENT AND CALIBRATION

bull 2D UNCOUPLED INTERACTION SURFACE

- SIGNIFICANT MODELING IMPROVEMENT

- DATA AVAILABLE

- IMPLEMENTATION PRACTICAL AND AVAILABLE

ANDREW PHASE II

JOINT MODEJJNG OPTIONS ILLUSTRATIONS

P

e

ANDREW PHASE II

JOINT MODELING PHASE II METHOD

e

I I I

ANDREW PHASE II

ANDREW STORM CONDmONS UPDATED

bull -PHASE I USED OCEANWEATHER HINDCAST-(Nov 1992)

bull middotmiddotPHASE II WILL USE THE REVISED HINDCAST (Nov 1994)

bull lt c lt I lt

2-2

l---4-~~~+-~~-+-~~~1--~~-+-~~--l--~T--+-~~--i24

f-tJVtMfiz~ 1Cfl (r STvb(

Plollol bullbull 5middotNOYmiddot9214bull1J Irbullbull Illbull [ANDREVl~NDREVCUSTbullMAXll 5-NOY-1992 10bull54 - 9208

HEIGHT lml

I~ MAXI

I L 1~middotamp SZ~- ~A I 13c

I J) ( r I X~0iWJ~0gtJ gt L gt gt I 1c I 120

-- N

~ deg

I I I I I I e C I ~=-41 122 -94 -92 -90 -88 -86 -84 -82 -80

-

Figure 7 contours of maximum hindcast significant wave height

_

~

~tl ~

NOltNttIJ 17lt ltNS S iJJ)y

Maximum Significant Wave Height - Hurricane Andrew 1992

30 ~

26 ~ I ~-~~~- == -shy -~ 35-_

24deg1--shy --------~- middot-middot-------middotmiddotmiddot----middot------middot-middot~middot- - - ~-----1---middot

22middot ~~~~~~~-~ -94 -92 -90 -88 -84deg-86 -82 -somiddot

c

-

Platform Selection bullbullbull1111111111111

0 BasisCriteria for Selection middot

0 Review of Phase I Platforms middot 0 Alternative Platforms

0 PMB Recommendations

0 Participants Comments

gt

L

l I

BasisCriteria for Selection bullbullbull1111111111111

0 Benefit for Calibration - Classification (UnexpectedExpected)

- Failure Mode (StructuresFoundations)

- Damaged StructuresKnown response

- Better Representation of Full ~opulation

0 Availability of Data - Existing Models

- Inspection Data

- Damage Data

PREDICTED BEHAVIOR

-

OBSERVED

BEHAVIOR

-SURVIVED DAMAGED FAILED

SURVIVED

ST134W WD90A

-MC311 -

MC397 -

--

-

-

ST151K ST130Q

-

-

-

DAMAGED - -

-

-

--

-

-

ST52 SS139 ST177B -

ST161A

-_

-

FAILED

-

-

-

ST151H ST130A ST72

Operator - Platfonns - Platfonns Survived Damagedamp

- Falled -

bull

-

-

Jle eBDIGIPllDbl -Amocoshy 30 1 Chevron -

143 12 Exxon 72 Mobll 22 1 Phllllps 7 bull

-Shell 39 Unocal 25 1

Total Partlclnts Platfonns 338 15

Other Operators bull- 310 13

-

Total Platfonns - -

648 28

fbml

Platforms Considered 6 7

tie of Total Platfonns 1 25

OVERVIEW OF PLATFORMS SURVIVEDDAMAGEDFAILED

-

shy

v

Platforms Evaluated in Phase I

-

-Platfonn Water Year Number Analysis Avallabillty of Availability Platform

- Name Depth rl of Model DetaDed rl ClassUlcatlon

Installation legs Inspection Soll Report Based on Information Phase I Calibration - -ft

- -- Suaill flalfnrm ~BSi -STISIK 137 1963 8 CAP Yes Yes Un eel Sarvlval

-STIJOO 170 1964 4 CAPmiddotshy Yes ST134data u ST134W 137 1981 4 CAP Yes Yesmiddot E

-

edSarvlval -

middotsurvival

WD90A 184 - 1964 8 CAP Yes E -

- Sarvlvill

MC3ll 343 1978 8 KARMA

MC397 468 1991 4 KARMA -

bull ~reglmmiddot- -

Yes Sure Survival

bull Yes - Sare Sanlval-

-T23STS2 63 1969 4 CAP

T2SSS139 62 1969 4 CAP -

Yes

Yes

Yes Ex

Yes Une

bull SurvivalLikely to DamalI

middot bull SanlvalLlkely to DamalI

ST161A 118 1964 8 USFOS

Eail11a pound1al(bullllm Clss ST177B 142 --1965 8 CAP

STISIH 137 1964 - 8 CAP C

STIJOA 140 19S8 8 CAP - -

Yes

Yes

ollamed

Yes

- Yes E

-

at about 1 mUe

Yes

ST134data

bull SafvlvalLlkelv to Dam- bull- -

E middotrahre

Emeried Failure -

Likely FaUare

T21ST72 61 1969 4 CAP Collamed No Likely FaUure

Platforms Identified for Evaluating Their Likelihood of Falling Under UnexpeCted Survival Category

CllEllATOa

AllB

llLOCll

Sl1lvcnJllE

NAME

YEAR

INSTAUJCD

WATER

IJBPJB

middotmiddotDISTANCE

ftOMSllOllE

(MIUIS)

Ill

(M)

AMOCO BllW

STllI

WDI075

A

bull D

4

n uo 17J

44

32

bull 20

10

bull CllEVllOH STtllO c I 1bull 21 11

STl130 D 0 161 21 11 STllJ4 I 0 137 D - 11

STIUI G I 137 32 11

STIUI I 0 121 32 11

STIUI

J 0 140 32 11

STIUI It 4 137 32 11

STllSI PROl)I 0 137 32 11

STllSI PROD-2 0 137 32 11

STll7 A 0

140 35 11

WDl117 c 65 214 34 10

DXON STshy E 51 14 bull STl165 A t3 bull 34 WD0073

WD0073

WD0073

A _ c

4

62

4

IA IA

17J

22

22

22

bullbullbull WD003-shy E 65 161 Z7

A 0 IN 31 MOBD Bl12I

ssn A bull

50

50

31

7

bull 7

MIJllPllY SS114 B 7 50 7 7

PlllUIPS SSl14t A 55 33 7

SAMEDAN ST17J A 4 117 7 bull 7

SJIELL WD0103 A 65 223 27 WDllOJ bull 65 221 27 WDll04 c 65 221 27 WDOI04 D 67 2 27

UNOCAL

SSl20I

SSl20I

bull SSl20I

SS0209

SS0209

SS0209

SS0209

SS 0253

SS 0253

E

I

PtJllE

A

bull D

G A

c

0

4

4

7J

61

0

0

69

103

97

97

9S

100

95

100

165

175

34

34

34

34

34

34

34

54

54

bullbullbullbullbullbullbullbullbull

Platforms Recommended for Phase II

llalr y_ w- Aot Av- _ I 1 Doplla Model nm bull bull _ bull I I H11 s18-1 Irht lbwl~ bullbullbull(-illtdlel OM lllMtD

II ---

-STISIIC 137 1963 I CAP Yeo Yeo I Yeo

STtshy -170 1964 4 CAP Yeo ST134dlU I Yeo

STl34W 137 Yeo Yeo -1981 4 CAP D Yeo

WD90A 114 1964 I CAP Yeo - D

MC311 30 1971 I KARMA Yeo -sre-

MC397 middot 461 1991 4 ICARMA Yeo Sireshy - - r- -

__ T13ST5l 63 1969 4 CAP Yeo Yeobull middot- ~n- D Yeo _ Tl5SSl39 62 1969 4 CAP Yeo Yeo I Yeo

_ STIT711 142 1965 I CAP Yeo I Yeo middot-1shySTl61A Ill 1964 I USFOS Yeo Yeo - D bulloshy

- r-

STISIH 137 1964 I CAP y D y -_ STl31A 140 1951 I CAP Yeo STl34dlU I Yeo

T21Sll 61 1969 4 CAP No rar-middot I Yeo

~-rmiddot-middot SS209A 95 1967 16 CAP Yeo D -SSU4R

11 UI A 41 -in D SS 2741AIM CAP bull

_ _ -shy - -shy- -shy

CALIBRATION

bull PHASEIAPPROACH

bull CALIBRATION ISSUES

bull PHASE II WORK SCOPE

bull SOFlWARE AVAILABILITY

bull EFFECT OF ADDITIONAL PLATFORMS

bull WEIGHTING APPROACH

bull EXAMPLE APPLICATION middot

(l=Ofce Mode] Wind

~

fcurrent l -middot

~SW ~leldl4

ultimate resistance force I first

design load Illgt

i _

deflection

-

bull

4

Historical Hurrlcane Dariiage and Survival of Platforms

Components Required for Calibration (Petrauskas 1992)

PRIOR Distribution

PLATFORM EXPERIENCE

BAYESIAN Survivals and Failures

UPDATING

Waves Winds and Currents bull

POSTERIOR Distribution

- A_middotJA ~ Survival Data Failure Data

Result Result

Load Resistance

middotfl

-

Bayesian Updaing middotGeneral Description (Petrauskas 1992)

L Imiddot

WHERE TO INTRODUCE BIAS FACTOR (B

On Resistance (Rmue - B (RPREDICreD I

middotOn Loads (Smue middot - B (SPRE01cTEo

On Safety Margin (RSmue - B (RSPREDICTED

CALIBRATION METHODOLOGY

gt

r

FIELD OBSERVATIONS

Observed Behavior

survivals Damages

Failures

r

bull

CAPACITY ANALYSIS

Predicted Load and Resistance Predicted Failure Modes

Survival Damage Failure cases

RELIABILITY ANALYSIS

- Probability of Failure ~

Ukellhood Failure

BAYESIAN UPDATING

Prior Distnbution of B Combined Likelihood Function

Posterior Dlstnbution of B

BIAS FACTOR

Mean Value cov bull

Represents Modeling Uncertainties

Prior Distribution of B Likelihood Functionmiddot Observing Success case STISIK and Posterior Distribution of B

2 00

uo

160

140

120

e 100 L $ uoIshy

0 60 ~ ~ 040

020

0 00

0 00 0 50

_ I 00

b

I 50 2 00

-shy lk [bl STISIK)

--shy Pnor-f(b)

--shy Pos1enor r(bl ISIK)

Prior or B Mean bull 10 COV bull 030

Posterlor of Bbull Meanbull 136COV 0 16

2 50

Likelihood Function and Posterior Distribution Example Platform STISIK

- -

~

- Prior and Posterior Distributions or B Failure cases (ST177B STlSlH ST130A T21) Damage cases (T23 T25 ST161A) Success cases (STlSlK ST130Q ST134W WD90A MC311 MC397)

400

350

300

2 so 0-

I

bull bull

bull bull gt

- bullbull ~ bull bull I bull

_ bull bull bullbull I bull bull bullII V -

bull bull

I bull ~ bull

bullbull bullI bull bullI bull bull bull

middot -~

bull__ ~ middotI middotmiddot

L 2 00

e ~ I SO

-I 00

- 0 so

000 gt

0 00 020 0 40 060 0 80 I 00 120 140 160 110 200

b -

-

shy

Prior-r

- - - - Posterior fLlll failure cases

bull bull bull bull bull bull bull Posterior flall damage cases

- bull bull - bull middot Postenor flall success cases

Posterior flall success fulure damage cases

-

Posterior of B Mean= 119 COV 0101

Posterior Distribution or Blas Factor (8) bull Different Categories and All Cases Combined

shy

l

PHASE I - KEY CONCLUSIONS

bull GLOBAL CORRECTION FACTOR

bull 15 TO 20 CONSERVATISM MEAN)

bull UNEXPECTED SURVIVALS HAVE GREATEST IMPACT

bull DAMAGED CASES MORE INFORMATIVE

bull CALIBRATION ISSUES PHASE II

bull

EFFECT OF IMPROVED CAPACITY ANALYSIS RECIPE

bull APPLICABILITY TO MULTIPLE FAILURE MODES

bull NO FOUNDATION DAMAGE CASE INCLUDED

bull IMPROVED REPRESENTATION OF FULL ANDREW DISTRIBUTION

-bull NUMBER OF PLATFORMS CONSIDERED

I 1 I

I t ) bull ~

PHASE II WORK SCOPE (CALIBRATION)

bull

bull ESTABLISH MULTIPLE BIAS FACTORS

Use Nine Base Case Platforms

middot - Capacity Analysis for Revised recipe

Foundation Blas Factor from Foundation JIP

bull DEVELOP WEIGHTING APPROACH

bull UPDATE MULTIPLE BIAS FACTORS

bullEffect of Platforms Not In the Nine Base Cases

bull EXAMPLE APPLICATION OF BIAS FACTORS I

bullbull

PHASE I PHASE II

CAPACITYbull ANALYSISRu bull

---------------------1----middot

Jc1 - iJbullraquoJ dbull

FttlS frt[f~l ratdhJJ

-

LIKELIHOOD FUNCTION

DEVELOPMENT

lk Jb Jfall1N) bull Jb)

111 Ill Isurvtvoq bull 1 - bl

-middot -------------------shy ~---------------------middot-middot

r111ir8 Cb1 111 Cbl BAYESIAN

UPDATING

~cove

R1R2 bull middot---------------------~----

Jblbull P(g cOJ

1 bull JbRI (MBSI I bull 1

~middot

llcJllJfalluro) bull P(gcOJ Ill 111 b21fallun) bull P ((g1 CO) U 1112 c OD

middot--------------------shy ----------------------shy bull

r111111r8 Cbl 111 Ill i-1nro1

deg ~1b2)laquottfrr (b1b21-lnfo) re Ill bull 1-11middot b2) bull lodegamp

middot

i1 covamp i2 cov~i

-

shy

--

Likelihood Function vs b All Success Case STISI K STl30Q STl34W WD90A MC311 MC397

I 0000

0 9000

OIOOO

0 7000

~ 0amp000

ta osoOO

~ 0 4000

0 3000

0 2000

0 1000

bull 0 0000

bull

I

middot

middot bull I

Ibull

-sibull ~ -middot-middot-middot-middot -middot-

----7----+-----~ I

Ibull bull bull I

rbull

--- Jk ISIK[blsuccess)

- bull - bull - lk 130Q[blsuccess) bull

lk 134W[blsuccess)

Jk-WD90A(blsuccc~I

- - - lk MC397[blsucccss)

--ltgt-- lk middotMC31 l[blsucccss)

--- Jk [bl 6 out of6succcsHascs)

05 I 5 2 250

b

Likelihood Functionsmiddot Success Cases - Phbullsc L

shy

bullbull bull bull bull bull bull -------------------~

bull bull bullbull bullbull bullbull bullbull bullbull bullbull bull

bull bullbull

bullbullbullbull

bullbull

bullbullbullbullbullbullbull

bullbullbullbullbullbull

bullbullbullbullbullbullbull

z 0

t z ) LL

c 0 0c i ~ i

J

~ sect ~--------------~ e l5

I

I

SOFTWARE

bull PF PROGRAM

bull CALREL

bull PF +CALREL

bull RELACS

bull

gt i

l

SENSITIVITY STUDIES

POSTERIOR DISTRIBUTION OF B

bull EFFECT OF CHANGE IN DISTRIBUTION OF PRIOR OF B

bull EFFECT OF REMOVAL OF ONE PLATFORM FROM EACH CATEGORY

bull EFFECT OF REMOVAL OF THREE DAMAGE CASES

bull EFFECT OF ADDITIONAL 13 SIMILAR PLATFORMS

-I

Prior and Posterior Distributions of B Variation In COV of PriorofB fromOl to04 -- -

4500

-4000

-

- -

3500 --middot - bull bull - bull bull Prior-r ICOV =02

3000

- - - - Prior-r ICOV =03 -

bull -~ e 2soo -

bull bull bull bull bull bull bull Pnor-r ICOV =o 4t frJ ~ 2000 Posterior rlCOV =0 2

bull bull Postenor rlCOV =0 3 bull bull 1500 bull middot bull - shy -middot- Posterior rlCOV =04

middot~ I 000 - bull

-middotmiddot ~ ~middotmiddot M -~

- middot 0 500

wbullbullbull _ I middot~ middotmiddotmiddotmiddotmiddot middotmiddotmiddotmiddotmiddot - -~ ~ __ middot_ _ bull -shy0 000

0 000 0200 0 400 0 600 0 800 IOOO 1200 1400 1600 1800 2000 -

- - b

-

=shy

Posterior Distribution of Bias Factor (B) bull Effect of Variation In COV of Prior of B

bull

bullPrior and Posterior Distributions of B Evaluating Effect of Removal of One Platform from Each Category from the Base Case of 13 Platforms

400

~ ~ 350

j IA i

3 00 bull I bull I ---Prior-r I bull l

250 1 I 1

I bull bull ---Posterior r lall 13 casesshy- ~E 2 00 bull bull bull bull bull bull bull Posterior r without STISIK

- ~ 1 bull - bull I - bull bull - bull bull Posterior r Iwithout T2I I 50 I

J-tt~i - bull - bull - Posterior r Iwithout T2S

tOO [-

I I ~ ~ I 0 50

0 00

___ r J i

l~T

~--

0 00 0 20 0 40 0 60 080 I 00 120 140 160 110 2 00

b

zshy

Posterior Distribution of Bias Factor (8) bull Effect of Removal of middotPlatform from Any Category

bullPrior and Posterior Distributions or B Evaluating Effect or Removal or 3 Damage Cases from the Base Case or 13 Platrorms

4

35

3

25

e L 2 ~ bulle shy

I 15

bull I

05

0

0 02 04 06 I 2 I 4 16 IS 2

I I

I

I

I Ibull

J -tmiddot L--

bullI

~

I

fl

II ~

I I v I

middot bull bull

bull L

~ ll- I-~lt

OS I

b

I

bull

1middot bullbullbullbull

~ __

bull

- - - - Prior-r

- - - - Postenor flall failure cases

Postenor flall success cases

--- Posterior flall 13 cases

---Posterior fI success failure cases only

IPosterior or B Mean =t11 COV =0121

-

_I

Posterir Distribution or Bias Factor (B) bull Effect or Removal or 3-Damage Cases

-

bullPrior and Posterior Distributions or B Evaluating Effect or Additional 13 Similar Observations as

Used in Andrew JIP The POsterior or B from Andrew-JIP Considered as New Prior or B

- 500

450

400

350

bull bull

bull bull bull bull J bull

bull bull f ~ I

I ~ Ibull bull

bullbull bull

l bull I -bullbull bull bull

I

I bull bull

bull I I~~bull bull bull

bull ~

I bull bull

~ ~_ ~~~

- - New Prior r bull

Posterior 114 new failure cases ~ 300- bull bull bull bull bull bull bull Posterior f 13 new damage casesl 150 ~

- bull bull - bull bull Posterior 116 new success cases i 2 00

Postenor 1113 new success failure -150 damage cases

I 00

0 50 -

0 00 0 00 0 20 0 40 0 60 0 80 I 00 120 I 40 160 110 2 00

Posterior ol B Mean= 120 COV= 0071 b

shy

shy

_rshy

Posterior Distribution of Bias Factor (B) bull Effect of Additional 13-Similar Observations

1

1 I

WEIGHTING APPROACH bull

bull ASSESSMENT OF PLATFORMS DATA

Select Platforms Most Mappable to the Analyzed Platforms

Unlikely SurvivalDamage Cases Identification

Based on Cut-Off Criteria Agemiddot Hs etc

bull INFORMATION REQUIRED FOR COMPARISON

bull CATEGORIZATION OF ADDITIONALmiddot PLATFORMS

bull UPDATING OF CALIBRATION RESULTS

--

OVERVIEW OF PLATFORMS SURVIVEDDAMAGED -~

Operator -

- Platforms

Survived CS)

Platforms

Damaged CD)

Platforms Used In

Phase IJIP -

JI~ ~Bill~IRBDll -

Amoco -

Chevron -

Exxon

Mob II

Phllllps

Shell

Unocal -

30

143

72 22

7

39 25shy -

1

12

bull

1 bull

1

1(S)+1 (D)

3 (S) +3(D)

2 (S)

bull shy

bull

bull

bull

Total Partlclnants Platforms 338 - 15 10C6S+4D) -

Other Operators

310

13

3 (D)

Total Platforms 648 28 13C6S+7D) -

Summary of Andrew JIP bull Phase I Platforms

Plairshy w v- N-ber Lea-le 1 shy car Hind_ o-w N-bull Deplh fl fl IJpe la bull n Yeadon Jadtet lbelledtf liefalit -- _ -OD

IL H - -shy _ICllshy IL IL - - rQClto ~ _

K STUIK 137 1963 8 365 608S4 Cellardeck

-middot STl-shy 170 1964 4 -~ K 39 62269 Cellardeck - --ST134W 137 1911 4 43 608S4 Smvlnlshy- _ - -

WD90A 184 1964 8 K 267Sbull34 83 SOS31 SUboftllar deck SmvMI

MC311 343 1978 8 x S7 62 - MC397 468 1991 4 x shy S33bull6SO 65911 -shySmeSmvlvol -- gt -

r-- T23ST52 63 1969 4 ton- shyKshy 52594 S03 -T25SSl39 62 1969 4 Kshy SOS94 S063 - middot ton-shy__ -

bullbull a ton--shyST161A 118 1964 8 K - 347S 51495 Cellarclecti

r~

ST1771 142 1965 8 K 39 60167 CeDordeck failme

STISIH 137 1964 8 K 395 608S4 CeDardeck Failme

K ST130A 140 1958 a SS 6095 I lblv Pai1uno - T21ST72 61 1969 4 Kshy S4S94 4974 I lhlv Pai1uno -

- -

JE

INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON

bull WATER DEPTH

bull bull YEAR OF INSTALLATIONDESIGN

bull PLATFORM ORIENTATION

bull PLATFORM LATITUDE AND LONGITUDE

bull KEY SKETCHES OF PLATFORM STRUCTURAL FRAMING

Number of Legs Plies Leg Batter Number of Conductors bull Deck Elevations (Including sub-cellar deck If any) Jacket Framing Elevations Brace type In Vertical Frames Sizes of Primary Members (leg pile vertical frames braces primary horizontals) Joint Details (can gap etc) LegPiie Annulus Grouting Status

bull middotPOST ANDREW INSPECTION REPORT Significant Wave Height Hs (ft)

bull INFORMATION ON DAMAGES PRIOR TO ANDREW

bull RESULTS FROM IN-HOUSE STUDIES IF AVAILABLE

bull ~RIGINAL DESIGN CRITERIA INFORMATION IF AVAILABLE

bull SOIL SHEAR STRENGTH PROFILE IF AVAILABLE

Platforms Identified for Evaluating Therr Likelihood of Falling Under Unexpected Survival Category

OPlRATOR

middotADA

BLOClt

llmucnJllE

NAME

YEAll

INSTAUJD

WA1ER

DBPTll

DISTANCE

JlllOM SHOllE

(MILU)

Bl

(M)

AMOCO 111113

STl11

WDI075

A

bull D

St

17

120

1n

44

20

10 I

CllllVllQfl STlllO c 61 110 21 11 STIUO D Q 1110 21 11

STl134 p Q 137 29 11

ST0151 G 61 137 11

ST0151 I 3 121 11

ST0151 J Q 140 11

STl151 It 137 11

ST0151 PJl001 Q UT 11

ST0151 PROD-2 Q UT 11

ST017 A 3 140 35 11

WDIUT c 65 214 34 10

llXXON STI055 II 51 6T 14 I

STl15 A t3 34

WDI073

WD0073

Amiddotshy Q

1111 1111

22

22

I

I

WDI073 c 1n 22 I

- WDOCl93-shy II

65 1110 2T A 3 194 31 MOBU 111121

SS IOl7

A B

501

50

50

31

T

I

T 6

MtlllPllY SS114 B T 50 T T

PlllllIPS SS0149 A 5 55 33 T

SAMllDAN STln A 107 T T

SBllLL WDl103 A 65 223 2T WDl103 B 65 221 2T WD0104 c 65 221 2T WD0104 D 6T 20 2T

UNOCAL SS020I

SS 020I SSl20I

SS 0209

SS0209

II p

PURll

A

B

3

n 61

103

97

97

95

100

34

34

34

34 34

bullbull bullbullbull

- SS0209 D u 95 34 I

- SS 1209 SSl253

SS0253

G A

c

6T

2

100

165

1T5

34

54

54

bullbull I

I I

I I

SOUTH MARCH

~LAND

N

SOUTH PaTOI ENE ISWltD AREA

MMS Defined Limits ror Offshore Structures Requiring Inspections Following Andrew

t---Jbulln3 T25bull I bullTtl

bull1$1Htl1Abull bull1SIK1779bull

~

middot-DaTA AREA~T

CRANO rsLE

AREA bull1300

bullll4W

=shy

bullPath of Hurricane Andrew with Platforms Used In Calibration

Likelihood Function vs b All Success Case STISI K STIJOQ ST134W WD90A MCJll MCJ97

I 0000

0 9000

0 8000

07000

-Bo6000

tiiii 0 5000 -

04000

0 3000

0 2000

01000

0 0000

0 05 25

ri1rt middot-middot-middot-middot-middot-

I 5 2

b

---11 middot151K(blsu~ccsI

- bull - bull - fk 130Q(bfU~lCSS)

lie bull134W(blsuccc]

lk-WD90A[blsultCS]

- - - lk middotMC397(bfUtlCS]

--o-- lk-MCJll(bbuctcsJ

--- ll (bl 6 out of 6 Mlttcs case]

_

( Likelihood Functions middotSuccess Cases - p - 1 c I

CATEGORIZATION OF ADDITIONAL PLATFORMS

I I I

bull CATEGORIES

Unexpected Survivals

Expected Survivals

bull APPROACH

Identify Corresponding Platform(s) Evaluated In the JIP

Determine Max Wave Height at Platform Location

Evaluate Whether Wave Hit the Deck

middot - Evaluate Likelihood of R and S to Differ from the Base Cases

Review Joint Information

Assess Special Features

bull IDENTIFY NUMBER OF PLATFORMS IN EACH CATEGORY

I

UPDATING OF CALIBRATION RESULTS

bull INCLUDE EFFECT OF ADDITIONAL PLATFORMS ON bullBbull

Consider Increased 1Uncertalntles In R and S

Develop bullNeW Ukellhood Functions for Different Categories

middotLet bullxbull Number of Ukely Unexpected Survival Cases and y Number of Ukely Expected Survival Cases

Determine Combined Likelihood Function lk (b Iadditional platforms) middot

bull REVISED LIKELIHOOD FUNCTION

lk(b I all cases) = lk(b I 9 platforms) middot x lk(b Ix platforms) x lk(b I y platforms)

bull ESTABLISH REVISED BIAS FACTOR B

f8 (b) u f8 (b) lk (b I all cases)

II

t t ~

shy

APPLICATIONS OF BIAS FACTORS

ii APPLICABLE TO PARTICULAR LOAD AND RESISTANCE RECIPE

bull USE BY THE API COMMITTEES AND MMS

Criteria Revisions

Recommended Design Guidelines Revisions bull

bullbull REASSESSMENT STUDIES AND DECISION MAKING

middot

bull RISK ASSESSMENT STUDIES BY THE OPERATORS

Individual Platforms

Fleet of Platforms

middot IMR Program Implementation

bullbull NOT TO BE USED IN DESIGN

I I fl J

EXAMPLE USE OF BIAS FACTORS

bull PHASE I EXAMPLE APPLICATION

bull EVALUATION OF ANNUAL PROBABILITY OF FAILURE

Establish Annual Maximum Seastate Data

Determine Coefficients to Define Base Shear

Determine Ultimate Capacity of Platform

Determine Probablllty of Failure Versus bullbbull

-bull DETERMINE UPDATED PROBABILITY OF FAILURE

Pf - P [S gt R]

Pf I P [~ gt Rb I b] f 8 (b) db

Pr - J P [S gt Rb I b] f bull 8 (b) db

bull TOTAL PROBABILITIES OF FAILURE

Total Pf - 1 bull Il [1 middotPf] - l Pf bull

Total Pf - 1 bull Il [1middot Pf] - l Pfmiddot

Total Pr = 1 - n [1middot PfbullJ = l Pfshy

I I I

EXAMPLE APPLICATION bull PHASE I

PLATFORM ST 151 K

bull GULF OF MEXICO

bull WATER DEPTH = 137 FT

bull ANNUAL MAXIMUM HS MEDIAN 175 ft COV 031

bull ASSUMED STORM DURATION 3 hoursmiddot

bull ASSUMED NUMBER OF WAVES 800

bull MEDIAN HS FOR STORM HOURS 1 2 AND 3

bull MEDIAN CURRENT VALUE bull

bull DETERMINE ULTIMATE CAPACITY FOR THREE DIRECTIONS

bull DETERMINE ANNUAL FAILURE PROBABILITIES

Broadsidemiddot Diagonal End-on

0037 0025 0007

0071 0044 0018

0022 0016 0004

DETERMINE TOTAL ANNUAL PROBABILITY OF FAILURE

Pf - 007

Pf - 013

Pr - 004

bull

EXAMPLE USE OF BIAS FACTORS - PHASE II middot

bull DETERMINE PREDICTIVE PROBABILITY OF FAILURE

Case - 1

Pf _ S S Pf (b b2) fa1a2 (b b2) db db2

g bull -1 (b1 R I S) - 1

92 - (b2 R2 I S) - 1

Case -2

Pf s ~ [bJ f11 (b) db

Pf [bJ p [g lt 0)

g - bull 1 (b1 R1 I S) - 1

1

bull1) l

I

BENEFITS OF ANDREW JIP bull PHASE II

lt

bull IMPROVED LOAD AND RESISTANCE RECIPE

bull IMPROVED CALIBRATION PROCEDURE

bull MULTIPLE BIAS FACTORS

bull PROCEDURE TO INCLUDE EFFECT OF OTHER PLATFORMS

bull PROCEDURE TO USE MULTIPLE BIAS FACTORS

  • Structure Bookmarks
    • ANDREW PHASE II JOINT1 INDUSTRY PROJECT
    • PROGRESS MEETING NUMBER 1
    • ANDREW PHASE II JOINT INDUSTRY PROJECT PROGRESS MEETING NUMBER 1 NOVEMBER 16 AGENDA
    • ANDREW PHASE II UST OF PARTICIPANTS
    • ANDREW PHASE I OBJECTIVES AND RESULTS
    • ANDREW PHASE II PROJECT OBJECTIVES
    • SUBSEQUENT PROJECTS
    • ANDREW PHASE II PROJECT TEAM
    • ANDREW PHASE II UPDATE MODELING middotAND ANALYSIS RECIPE
    • -ANDREW PHASE II CONDUCTOR RESISTANCE
    • ANDREW PHASE Il WA VE FORCE PROFILE
    • ANDREW PHASE II WA VE FORCE PROFILEmiddot
    • ANDREW PHASE II BRACE MODELING
    • ANDREW PHASE II FOUNDATIONS
    • ANDREW PHASE II middot middot JOINT MODELING
    • ANDREW PHASE Il JOINT MODELING OPTIONS DISCUSSION
    • ANDREW PHASE II JOINT MODEJJNG OPTIONS ILLUSTRATIONS
    • ANDREW PHASE II JOINT MODELING PHASE II METHOD
    • ANDREW PHASE II ANDREW STORM CONDmONS UPDATED
    • Platform Selection bullbullbull1111111111111
    • BasisCriteria for Selection bullbullbull1111111111111
    • CALIBRATION
    • WHERE TO INTRODUCE BIAS FACTOR (B
    • PHASE I -KEY CONCLUSIONS
    • bull CALIBRATION ISSUES PHASE II
    • PHASE II WORK SCOPE (CALIBRATION)
    • SOFTWARE
    • SENSITIVITY STUDIES POSTERIOR DISTRIBUTION OF B
    • WEIGHTING APPROACH bull
    • INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON
    • CATEGORIZATION OF ADDITIONAL PLATFORMS
    • UPDATING OF CALIBRATION RESULTS
    • APPLICATIONS OF BIAS FACTORS
    • EXAMPLE USE OF BIAS FACTORS
    • EXAMPLE APPLICATION bull PHASE I PLATFORM ST 151 K
    • EXAMPLE USE OF BIAS FACTORS -PHASE II middot
    • BENEFITS OF ANDREW JIP bull PHASE II
Page 15:  · . ' ANDREW PHASE II PROJECT OBJECTIVES . ' • ·UPDATE MODELING AND ANALYSIS RECIPE· JOINTS ' ' ·BRACES WAVE PROFILE FOUNDATIONS . ' ' CONDUCTORS· • ANALVZE PLATFORMS USING

--

Project ST151K Model Conductor Version 2 Mon Nov 14 142145 1994

I I I I

I I

I

I

I - - -

-Ishy-

-

Variable Displacement step 35 InelasticEvents Legend

Elastic Strut Buckling Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield Beam Clmn Fully Plastic Fracture

Project ST151K Model Conductormiddot

Version 1

Clgt - Cut Plane Force Fx Mon Nov 14 140051 1994

- -_ vii I fOIZM

0 _

~

N ID 0 bull 0

x 0

middotrt ~ () bull u 0 ~ 0 Ir

N bull

0

0 1 2 3 4 5 X Deck Disp [Ft)

I I 1 1 WbullTI- bullbull

lltcxi lr-C

011lt~~~~~~~---~~~~~~~~L~~~~~~~--~~~~~~~~i~~~~~~~--

Uniform Node Displacement

shy

Ji

r

Comparison of Pushover Curves for ST 151K Variable Load Pattern and Monotonic Load Pattern (Andrew)

Soll Improved 100

6000

5000 -i 4000

f

llS

3000 Ul

J 2000

10 c11 c1lgtolts 1000

0 v

bull ~

f

_ ----- --- ---shy

bull --Variable Lold

- - Andrew Load

I I I I I I

4 5 62 30 1

Deck Dlsplacement (ft)

bullbull r 1 II

~

ANDREW PHASE Il

WA VE FORCE PROFILE

bull DECK WA VE FORCE ~

PHASE I USED SECTION 17 (1056H degCd)

PHASE Il USE REVISED SECTION 17 (1000H Cd)

bull PHASE I USED SINGLE FORCE PROFILE FOR PUSHOVER

- USED WAVE HEIGHT CLOSE TO ULTIMATE CAPACITY

bull PHASE J

Il WILL STUDY THE

USE OF VARIABLE

PROFILE ~

- IF FORCE LEVEL INCREASES (OR DECREASES) BY A FACTOR EQUIVALENT TO A 2 FOOT CHANGE IN WA VE HEIGHT THE INCREMENTAL FO~CE PA1IERN IS CHANGED ACCORDINGLY (CAP FEATURE)

- INITIAL EXAMPLE

f~

a

middotn

a

I N -a a a amiddot

+

~ a

l

a w gt 0c 0 gt a I shyc -CJ wc w gtc(

== c( I shyZ w E W a 0 z- 8

ii sect

~ ~ bullT 0

I if

x ~

bullbullbull

II ~------------________

ANDREW PHASE II

WA VE FORCE PROFILEmiddot

CAP AUTOMATIC APPLICATION FEATURE

STAT

fR1 0 lt f lt1

STAT

STAT

f(R3-R2) middot f lt1

Project ST151K Model Wave_Loading Version 1 rue Nov 8 164107 1994 PT

E~ nss FT

SL 14amp FT

a 111 FT

EL 14 li

E-1 37 FT

El o FT

1middot1middot ______________ _ ________ __ _ __

0

200 400 600 800 1000 12000 Force at Elevatloii (kip)

200

a ll_

Wave Force at Major Elevations for Discrete Wave Heights

180

bullHmiddot40ft

Hmiddot495ft

H bull 51 ft

Hmiddot55ft_

H-59ft I

Andrew (H bull 60 86 ft) II

i I

_11---__

- middot- Hmiddot30ft

-

- bull -

middot bull bull bull middot

---Hmiddot53ft

bullbullbullbullHmiddot57ft

-----H-63ft

Elevation Above Mudllne 100

(ft)

180

140

120

80

60

40

20

Loading at Major Elevations

---H-49511

-deg-Hmiddot608611

0 200 400 600 800 1000 1200

Force It Elevatlon (kip)

--

Base Shear vs Wave Height ST151K

6000

5000 1---------Ult1mateCapacitybull 4890 kip (100 Improved Soll)-----------~--

4000a a lii ~ 3000

81 ampI

2000

1000

0 +-~~~--i~~~~~~~~~-+-~~~--lf--~~~~~~~~-+-~~~--l

30 35 40 45 50 55 60 65

Wave Height (ft)

bull middot

~

bull

II

2

g

~

It)

0

cCD E 0 E a c E euro~ CDshygtItgtmiddoto ~ti c -a Gic CD gt ~

-= c -Cll

I z

bullO 0i i ~ g 0

() CI -bull(q-ll) iuawow llu1wniiaAO

11

shy

116

114

112

g 110

bullc = 108

~ ~ 106 -sect 104

102

100

98

96

-

6555 60

Wave Height vs Center of Action ST151K

30 35 40 45 50

Wave Height (ft)

t

Comparison of Pushover Curves for ST 151K Variable Load Pattern and Monotonic Load Pattern (Andrew)

Soll Improved 100

--- --------- --- ----6000

5000

- 4000 8 - I

1 3000 Ul

81 m 2000

_J -shy

1000

---Variable Load

- - Andrew Load

0 f-~~~~~r--~~~~-t-~~~~~-t-~~~~~+-~~~~----t~~~~~-1

0 1 - 2 3 4 5 6

Deck Displacement (ft)

-middotshy

1 t I I ~

ANDREW PHASE II

BRACE MODELING

bull MARSHALL STRUTS USED FORPHASE I

bull PHASE II WILL ASSESS DIFFERENCE IN USING BUCKLING BEAMS AND MARSHALL STRUTS

middot bull COMPARE FOR 2D FRAME

bull COMPARE FOR PLATFORM 177 WHERE LOWER

BAY EXHIBITS LARGE BRACE MOMENTS

Project andrew Model beamframe Version 1 Mon Nov 14 145447 1994

~

middot Frame Test buckling beam-column model

-

Project andrew Model strutframe Version 1 Mon Nov 14 145754 1994

bull

Frame Test strut model

middot-middot _____________--_

_middotmiddotr_~ --ll--~-middot__

Buckling Beam bull

Strut (K=O 65)

Lateral Displacement

ANDREW PHASE II

FOUNDATIONS

bull ~LICIT NONLINEAR API FOUNDATION USED INiPHASE I

bull FOUNDATIONS

JIP SELECTED MINI-VANE DATA FOR

STRENGTH PROFILE

bull COMMENTS ON BENCHMARK COMPARISONS

- SIGNIFICANT VARIABILITY DUE TO FOUNDATION MODELING METHODS

II Ii

z

ctV 4x SPlO caissonfy-42s soil-api(lf-097step-114)

Inelastic Events Legend Elastic Strut Buckling

Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield

--------shyBeam Clmn Fully Plastic Fracture

~--~~~~~~~~~--~----~__i~~---~~~~__--~

Pro)ect apipile Model South Pelto 10 Version Wed May 18072235 1994 - -

1bullt

+

t

f i bullbullbull

bullbull bullbull

I I

5

~x SPlO Caissonfy=42s soil=api(lf=lOOstep=137)

Inelastic Events Legend

Elastic Strut Bucklilg

Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield

earn Clmn Full~ Plas1c Fracture

ANDREW PHASE II middot middot

JOINT MODELING

middotbull API RIGID-PLASTIC USED IN PHASE I

a ELASTIC-BRITTLE USED IN PHASE IB

amiddot MSL WILL PROVIDE NONLINEAR FORCE-DEFORMATION

AND MOMENT-DEFORMATION RELATIONS FOR PHASE II

a MODELING OPTIONS FOR PHASE II -

l

- UNCO~LEDmiddotONE DIMENSIONAL MEMBER END JOINT MODELING

J Ii

ANDREW PHASE Il

JOINT MODELING OPTIONS DISCUSSION

bull JD COUPLED INTERACTION SURFACE

- IDEAL

- DATA NOT AVAILABLE TO CALIBRATE

bull 2D COUPLED INTERACTION SURFACE

- SIGNIFICANT MODELING IMPROVEMENT

- DATAAVAILABLE

- IMPLEMENTATION PRACTICAL BUT NEEDS DEVELOPMENT AND CALIBRATION

bull 2D UNCOUPLED INTERACTION SURFACE

- SIGNIFICANT MODELING IMPROVEMENT

- DATA AVAILABLE

- IMPLEMENTATION PRACTICAL AND AVAILABLE

ANDREW PHASE II

JOINT MODEJJNG OPTIONS ILLUSTRATIONS

P

e

ANDREW PHASE II

JOINT MODELING PHASE II METHOD

e

I I I

ANDREW PHASE II

ANDREW STORM CONDmONS UPDATED

bull -PHASE I USED OCEANWEATHER HINDCAST-(Nov 1992)

bull middotmiddotPHASE II WILL USE THE REVISED HINDCAST (Nov 1994)

bull lt c lt I lt

2-2

l---4-~~~+-~~-+-~~~1--~~-+-~~--l--~T--+-~~--i24

f-tJVtMfiz~ 1Cfl (r STvb(

Plollol bullbull 5middotNOYmiddot9214bull1J Irbullbull Illbull [ANDREVl~NDREVCUSTbullMAXll 5-NOY-1992 10bull54 - 9208

HEIGHT lml

I~ MAXI

I L 1~middotamp SZ~- ~A I 13c

I J) ( r I X~0iWJ~0gtJ gt L gt gt I 1c I 120

-- N

~ deg

I I I I I I e C I ~=-41 122 -94 -92 -90 -88 -86 -84 -82 -80

-

Figure 7 contours of maximum hindcast significant wave height

_

~

~tl ~

NOltNttIJ 17lt ltNS S iJJ)y

Maximum Significant Wave Height - Hurricane Andrew 1992

30 ~

26 ~ I ~-~~~- == -shy -~ 35-_

24deg1--shy --------~- middot-middot-------middotmiddotmiddot----middot------middot-middot~middot- - - ~-----1---middot

22middot ~~~~~~~-~ -94 -92 -90 -88 -84deg-86 -82 -somiddot

c

-

Platform Selection bullbullbull1111111111111

0 BasisCriteria for Selection middot

0 Review of Phase I Platforms middot 0 Alternative Platforms

0 PMB Recommendations

0 Participants Comments

gt

L

l I

BasisCriteria for Selection bullbullbull1111111111111

0 Benefit for Calibration - Classification (UnexpectedExpected)

- Failure Mode (StructuresFoundations)

- Damaged StructuresKnown response

- Better Representation of Full ~opulation

0 Availability of Data - Existing Models

- Inspection Data

- Damage Data

PREDICTED BEHAVIOR

-

OBSERVED

BEHAVIOR

-SURVIVED DAMAGED FAILED

SURVIVED

ST134W WD90A

-MC311 -

MC397 -

--

-

-

ST151K ST130Q

-

-

-

DAMAGED - -

-

-

--

-

-

ST52 SS139 ST177B -

ST161A

-_

-

FAILED

-

-

-

ST151H ST130A ST72

Operator - Platfonns - Platfonns Survived Damagedamp

- Falled -

bull

-

-

Jle eBDIGIPllDbl -Amocoshy 30 1 Chevron -

143 12 Exxon 72 Mobll 22 1 Phllllps 7 bull

-Shell 39 Unocal 25 1

Total Partlclnts Platfonns 338 15

Other Operators bull- 310 13

-

Total Platfonns - -

648 28

fbml

Platforms Considered 6 7

tie of Total Platfonns 1 25

OVERVIEW OF PLATFORMS SURVIVEDDAMAGEDFAILED

-

shy

v

Platforms Evaluated in Phase I

-

-Platfonn Water Year Number Analysis Avallabillty of Availability Platform

- Name Depth rl of Model DetaDed rl ClassUlcatlon

Installation legs Inspection Soll Report Based on Information Phase I Calibration - -ft

- -- Suaill flalfnrm ~BSi -STISIK 137 1963 8 CAP Yes Yes Un eel Sarvlval

-STIJOO 170 1964 4 CAPmiddotshy Yes ST134data u ST134W 137 1981 4 CAP Yes Yesmiddot E

-

edSarvlval -

middotsurvival

WD90A 184 - 1964 8 CAP Yes E -

- Sarvlvill

MC3ll 343 1978 8 KARMA

MC397 468 1991 4 KARMA -

bull ~reglmmiddot- -

Yes Sure Survival

bull Yes - Sare Sanlval-

-T23STS2 63 1969 4 CAP

T2SSS139 62 1969 4 CAP -

Yes

Yes

Yes Ex

Yes Une

bull SurvivalLikely to DamalI

middot bull SanlvalLlkely to DamalI

ST161A 118 1964 8 USFOS

Eail11a pound1al(bullllm Clss ST177B 142 --1965 8 CAP

STISIH 137 1964 - 8 CAP C

STIJOA 140 19S8 8 CAP - -

Yes

Yes

ollamed

Yes

- Yes E

-

at about 1 mUe

Yes

ST134data

bull SafvlvalLlkelv to Dam- bull- -

E middotrahre

Emeried Failure -

Likely FaUare

T21ST72 61 1969 4 CAP Collamed No Likely FaUure

Platforms Identified for Evaluating Their Likelihood of Falling Under UnexpeCted Survival Category

CllEllATOa

AllB

llLOCll

Sl1lvcnJllE

NAME

YEAR

INSTAUJCD

WATER

IJBPJB

middotmiddotDISTANCE

ftOMSllOllE

(MIUIS)

Ill

(M)

AMOCO BllW

STllI

WDI075

A

bull D

4

n uo 17J

44

32

bull 20

10

bull CllEVllOH STtllO c I 1bull 21 11

STl130 D 0 161 21 11 STllJ4 I 0 137 D - 11

STIUI G I 137 32 11

STIUI I 0 121 32 11

STIUI

J 0 140 32 11

STIUI It 4 137 32 11

STllSI PROl)I 0 137 32 11

STllSI PROD-2 0 137 32 11

STll7 A 0

140 35 11

WDl117 c 65 214 34 10

DXON STshy E 51 14 bull STl165 A t3 bull 34 WD0073

WD0073

WD0073

A _ c

4

62

4

IA IA

17J

22

22

22

bullbullbull WD003-shy E 65 161 Z7

A 0 IN 31 MOBD Bl12I

ssn A bull

50

50

31

7

bull 7

MIJllPllY SS114 B 7 50 7 7

PlllUIPS SSl14t A 55 33 7

SAMEDAN ST17J A 4 117 7 bull 7

SJIELL WD0103 A 65 223 27 WDllOJ bull 65 221 27 WDll04 c 65 221 27 WDOI04 D 67 2 27

UNOCAL

SSl20I

SSl20I

bull SSl20I

SS0209

SS0209

SS0209

SS0209

SS 0253

SS 0253

E

I

PtJllE

A

bull D

G A

c

0

4

4

7J

61

0

0

69

103

97

97

9S

100

95

100

165

175

34

34

34

34

34

34

34

54

54

bullbullbullbullbullbullbullbullbull

Platforms Recommended for Phase II

llalr y_ w- Aot Av- _ I 1 Doplla Model nm bull bull _ bull I I H11 s18-1 Irht lbwl~ bullbullbull(-illtdlel OM lllMtD

II ---

-STISIIC 137 1963 I CAP Yeo Yeo I Yeo

STtshy -170 1964 4 CAP Yeo ST134dlU I Yeo

STl34W 137 Yeo Yeo -1981 4 CAP D Yeo

WD90A 114 1964 I CAP Yeo - D

MC311 30 1971 I KARMA Yeo -sre-

MC397 middot 461 1991 4 ICARMA Yeo Sireshy - - r- -

__ T13ST5l 63 1969 4 CAP Yeo Yeobull middot- ~n- D Yeo _ Tl5SSl39 62 1969 4 CAP Yeo Yeo I Yeo

_ STIT711 142 1965 I CAP Yeo I Yeo middot-1shySTl61A Ill 1964 I USFOS Yeo Yeo - D bulloshy

- r-

STISIH 137 1964 I CAP y D y -_ STl31A 140 1951 I CAP Yeo STl34dlU I Yeo

T21Sll 61 1969 4 CAP No rar-middot I Yeo

~-rmiddot-middot SS209A 95 1967 16 CAP Yeo D -SSU4R

11 UI A 41 -in D SS 2741AIM CAP bull

_ _ -shy - -shy- -shy

CALIBRATION

bull PHASEIAPPROACH

bull CALIBRATION ISSUES

bull PHASE II WORK SCOPE

bull SOFlWARE AVAILABILITY

bull EFFECT OF ADDITIONAL PLATFORMS

bull WEIGHTING APPROACH

bull EXAMPLE APPLICATION middot

(l=Ofce Mode] Wind

~

fcurrent l -middot

~SW ~leldl4

ultimate resistance force I first

design load Illgt

i _

deflection

-

bull

4

Historical Hurrlcane Dariiage and Survival of Platforms

Components Required for Calibration (Petrauskas 1992)

PRIOR Distribution

PLATFORM EXPERIENCE

BAYESIAN Survivals and Failures

UPDATING

Waves Winds and Currents bull

POSTERIOR Distribution

- A_middotJA ~ Survival Data Failure Data

Result Result

Load Resistance

middotfl

-

Bayesian Updaing middotGeneral Description (Petrauskas 1992)

L Imiddot

WHERE TO INTRODUCE BIAS FACTOR (B

On Resistance (Rmue - B (RPREDICreD I

middotOn Loads (Smue middot - B (SPRE01cTEo

On Safety Margin (RSmue - B (RSPREDICTED

CALIBRATION METHODOLOGY

gt

r

FIELD OBSERVATIONS

Observed Behavior

survivals Damages

Failures

r

bull

CAPACITY ANALYSIS

Predicted Load and Resistance Predicted Failure Modes

Survival Damage Failure cases

RELIABILITY ANALYSIS

- Probability of Failure ~

Ukellhood Failure

BAYESIAN UPDATING

Prior Distnbution of B Combined Likelihood Function

Posterior Dlstnbution of B

BIAS FACTOR

Mean Value cov bull

Represents Modeling Uncertainties

Prior Distribution of B Likelihood Functionmiddot Observing Success case STISIK and Posterior Distribution of B

2 00

uo

160

140

120

e 100 L $ uoIshy

0 60 ~ ~ 040

020

0 00

0 00 0 50

_ I 00

b

I 50 2 00

-shy lk [bl STISIK)

--shy Pnor-f(b)

--shy Pos1enor r(bl ISIK)

Prior or B Mean bull 10 COV bull 030

Posterlor of Bbull Meanbull 136COV 0 16

2 50

Likelihood Function and Posterior Distribution Example Platform STISIK

- -

~

- Prior and Posterior Distributions or B Failure cases (ST177B STlSlH ST130A T21) Damage cases (T23 T25 ST161A) Success cases (STlSlK ST130Q ST134W WD90A MC311 MC397)

400

350

300

2 so 0-

I

bull bull

bull bull gt

- bullbull ~ bull bull I bull

_ bull bull bullbull I bull bull bullII V -

bull bull

I bull ~ bull

bullbull bullI bull bullI bull bull bull

middot -~

bull__ ~ middotI middotmiddot

L 2 00

e ~ I SO

-I 00

- 0 so

000 gt

0 00 020 0 40 060 0 80 I 00 120 140 160 110 200

b -

-

shy

Prior-r

- - - - Posterior fLlll failure cases

bull bull bull bull bull bull bull Posterior flall damage cases

- bull bull - bull middot Postenor flall success cases

Posterior flall success fulure damage cases

-

Posterior of B Mean= 119 COV 0101

Posterior Distribution or Blas Factor (8) bull Different Categories and All Cases Combined

shy

l

PHASE I - KEY CONCLUSIONS

bull GLOBAL CORRECTION FACTOR

bull 15 TO 20 CONSERVATISM MEAN)

bull UNEXPECTED SURVIVALS HAVE GREATEST IMPACT

bull DAMAGED CASES MORE INFORMATIVE

bull CALIBRATION ISSUES PHASE II

bull

EFFECT OF IMPROVED CAPACITY ANALYSIS RECIPE

bull APPLICABILITY TO MULTIPLE FAILURE MODES

bull NO FOUNDATION DAMAGE CASE INCLUDED

bull IMPROVED REPRESENTATION OF FULL ANDREW DISTRIBUTION

-bull NUMBER OF PLATFORMS CONSIDERED

I 1 I

I t ) bull ~

PHASE II WORK SCOPE (CALIBRATION)

bull

bull ESTABLISH MULTIPLE BIAS FACTORS

Use Nine Base Case Platforms

middot - Capacity Analysis for Revised recipe

Foundation Blas Factor from Foundation JIP

bull DEVELOP WEIGHTING APPROACH

bull UPDATE MULTIPLE BIAS FACTORS

bullEffect of Platforms Not In the Nine Base Cases

bull EXAMPLE APPLICATION OF BIAS FACTORS I

bullbull

PHASE I PHASE II

CAPACITYbull ANALYSISRu bull

---------------------1----middot

Jc1 - iJbullraquoJ dbull

FttlS frt[f~l ratdhJJ

-

LIKELIHOOD FUNCTION

DEVELOPMENT

lk Jb Jfall1N) bull Jb)

111 Ill Isurvtvoq bull 1 - bl

-middot -------------------shy ~---------------------middot-middot

r111ir8 Cb1 111 Cbl BAYESIAN

UPDATING

~cove

R1R2 bull middot---------------------~----

Jblbull P(g cOJ

1 bull JbRI (MBSI I bull 1

~middot

llcJllJfalluro) bull P(gcOJ Ill 111 b21fallun) bull P ((g1 CO) U 1112 c OD

middot--------------------shy ----------------------shy bull

r111111r8 Cbl 111 Ill i-1nro1

deg ~1b2)laquottfrr (b1b21-lnfo) re Ill bull 1-11middot b2) bull lodegamp

middot

i1 covamp i2 cov~i

-

shy

--

Likelihood Function vs b All Success Case STISI K STl30Q STl34W WD90A MC311 MC397

I 0000

0 9000

OIOOO

0 7000

~ 0amp000

ta osoOO

~ 0 4000

0 3000

0 2000

0 1000

bull 0 0000

bull

I

middot

middot bull I

Ibull

-sibull ~ -middot-middot-middot-middot -middot-

----7----+-----~ I

Ibull bull bull I

rbull

--- Jk ISIK[blsuccess)

- bull - bull - lk 130Q[blsuccess) bull

lk 134W[blsuccess)

Jk-WD90A(blsuccc~I

- - - lk MC397[blsucccss)

--ltgt-- lk middotMC31 l[blsucccss)

--- Jk [bl 6 out of6succcsHascs)

05 I 5 2 250

b

Likelihood Functionsmiddot Success Cases - Phbullsc L

shy

bullbull bull bull bull bull bull -------------------~

bull bull bullbull bullbull bullbull bullbull bullbull bullbull bull

bull bullbull

bullbullbullbull

bullbull

bullbullbullbullbullbullbull

bullbullbullbullbullbull

bullbullbullbullbullbullbull

z 0

t z ) LL

c 0 0c i ~ i

J

~ sect ~--------------~ e l5

I

I

SOFTWARE

bull PF PROGRAM

bull CALREL

bull PF +CALREL

bull RELACS

bull

gt i

l

SENSITIVITY STUDIES

POSTERIOR DISTRIBUTION OF B

bull EFFECT OF CHANGE IN DISTRIBUTION OF PRIOR OF B

bull EFFECT OF REMOVAL OF ONE PLATFORM FROM EACH CATEGORY

bull EFFECT OF REMOVAL OF THREE DAMAGE CASES

bull EFFECT OF ADDITIONAL 13 SIMILAR PLATFORMS

-I

Prior and Posterior Distributions of B Variation In COV of PriorofB fromOl to04 -- -

4500

-4000

-

- -

3500 --middot - bull bull - bull bull Prior-r ICOV =02

3000

- - - - Prior-r ICOV =03 -

bull -~ e 2soo -

bull bull bull bull bull bull bull Pnor-r ICOV =o 4t frJ ~ 2000 Posterior rlCOV =0 2

bull bull Postenor rlCOV =0 3 bull bull 1500 bull middot bull - shy -middot- Posterior rlCOV =04

middot~ I 000 - bull

-middotmiddot ~ ~middotmiddot M -~

- middot 0 500

wbullbullbull _ I middot~ middotmiddotmiddotmiddotmiddot middotmiddotmiddotmiddotmiddot - -~ ~ __ middot_ _ bull -shy0 000

0 000 0200 0 400 0 600 0 800 IOOO 1200 1400 1600 1800 2000 -

- - b

-

=shy

Posterior Distribution of Bias Factor (B) bull Effect of Variation In COV of Prior of B

bull

bullPrior and Posterior Distributions of B Evaluating Effect of Removal of One Platform from Each Category from the Base Case of 13 Platforms

400

~ ~ 350

j IA i

3 00 bull I bull I ---Prior-r I bull l

250 1 I 1

I bull bull ---Posterior r lall 13 casesshy- ~E 2 00 bull bull bull bull bull bull bull Posterior r without STISIK

- ~ 1 bull - bull I - bull bull - bull bull Posterior r Iwithout T2I I 50 I

J-tt~i - bull - bull - Posterior r Iwithout T2S

tOO [-

I I ~ ~ I 0 50

0 00

___ r J i

l~T

~--

0 00 0 20 0 40 0 60 080 I 00 120 140 160 110 2 00

b

zshy

Posterior Distribution of Bias Factor (8) bull Effect of Removal of middotPlatform from Any Category

bullPrior and Posterior Distributions or B Evaluating Effect or Removal or 3 Damage Cases from the Base Case or 13 Platrorms

4

35

3

25

e L 2 ~ bulle shy

I 15

bull I

05

0

0 02 04 06 I 2 I 4 16 IS 2

I I

I

I

I Ibull

J -tmiddot L--

bullI

~

I

fl

II ~

I I v I

middot bull bull

bull L

~ ll- I-~lt

OS I

b

I

bull

1middot bullbullbullbull

~ __

bull

- - - - Prior-r

- - - - Postenor flall failure cases

Postenor flall success cases

--- Posterior flall 13 cases

---Posterior fI success failure cases only

IPosterior or B Mean =t11 COV =0121

-

_I

Posterir Distribution or Bias Factor (B) bull Effect or Removal or 3-Damage Cases

-

bullPrior and Posterior Distributions or B Evaluating Effect or Additional 13 Similar Observations as

Used in Andrew JIP The POsterior or B from Andrew-JIP Considered as New Prior or B

- 500

450

400

350

bull bull

bull bull bull bull J bull

bull bull f ~ I

I ~ Ibull bull

bullbull bull

l bull I -bullbull bull bull

I

I bull bull

bull I I~~bull bull bull

bull ~

I bull bull

~ ~_ ~~~

- - New Prior r bull

Posterior 114 new failure cases ~ 300- bull bull bull bull bull bull bull Posterior f 13 new damage casesl 150 ~

- bull bull - bull bull Posterior 116 new success cases i 2 00

Postenor 1113 new success failure -150 damage cases

I 00

0 50 -

0 00 0 00 0 20 0 40 0 60 0 80 I 00 120 I 40 160 110 2 00

Posterior ol B Mean= 120 COV= 0071 b

shy

shy

_rshy

Posterior Distribution of Bias Factor (B) bull Effect of Additional 13-Similar Observations

1

1 I

WEIGHTING APPROACH bull

bull ASSESSMENT OF PLATFORMS DATA

Select Platforms Most Mappable to the Analyzed Platforms

Unlikely SurvivalDamage Cases Identification

Based on Cut-Off Criteria Agemiddot Hs etc

bull INFORMATION REQUIRED FOR COMPARISON

bull CATEGORIZATION OF ADDITIONALmiddot PLATFORMS

bull UPDATING OF CALIBRATION RESULTS

--

OVERVIEW OF PLATFORMS SURVIVEDDAMAGED -~

Operator -

- Platforms

Survived CS)

Platforms

Damaged CD)

Platforms Used In

Phase IJIP -

JI~ ~Bill~IRBDll -

Amoco -

Chevron -

Exxon

Mob II

Phllllps

Shell

Unocal -

30

143

72 22

7

39 25shy -

1

12

bull

1 bull

1

1(S)+1 (D)

3 (S) +3(D)

2 (S)

bull shy

bull

bull

bull

Total Partlclnants Platforms 338 - 15 10C6S+4D) -

Other Operators

310

13

3 (D)

Total Platforms 648 28 13C6S+7D) -

Summary of Andrew JIP bull Phase I Platforms

Plairshy w v- N-ber Lea-le 1 shy car Hind_ o-w N-bull Deplh fl fl IJpe la bull n Yeadon Jadtet lbelledtf liefalit -- _ -OD

IL H - -shy _ICllshy IL IL - - rQClto ~ _

K STUIK 137 1963 8 365 608S4 Cellardeck

-middot STl-shy 170 1964 4 -~ K 39 62269 Cellardeck - --ST134W 137 1911 4 43 608S4 Smvlnlshy- _ - -

WD90A 184 1964 8 K 267Sbull34 83 SOS31 SUboftllar deck SmvMI

MC311 343 1978 8 x S7 62 - MC397 468 1991 4 x shy S33bull6SO 65911 -shySmeSmvlvol -- gt -

r-- T23ST52 63 1969 4 ton- shyKshy 52594 S03 -T25SSl39 62 1969 4 Kshy SOS94 S063 - middot ton-shy__ -

bullbull a ton--shyST161A 118 1964 8 K - 347S 51495 Cellarclecti

r~

ST1771 142 1965 8 K 39 60167 CeDordeck failme

STISIH 137 1964 8 K 395 608S4 CeDardeck Failme

K ST130A 140 1958 a SS 6095 I lblv Pai1uno - T21ST72 61 1969 4 Kshy S4S94 4974 I lhlv Pai1uno -

- -

JE

INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON

bull WATER DEPTH

bull bull YEAR OF INSTALLATIONDESIGN

bull PLATFORM ORIENTATION

bull PLATFORM LATITUDE AND LONGITUDE

bull KEY SKETCHES OF PLATFORM STRUCTURAL FRAMING

Number of Legs Plies Leg Batter Number of Conductors bull Deck Elevations (Including sub-cellar deck If any) Jacket Framing Elevations Brace type In Vertical Frames Sizes of Primary Members (leg pile vertical frames braces primary horizontals) Joint Details (can gap etc) LegPiie Annulus Grouting Status

bull middotPOST ANDREW INSPECTION REPORT Significant Wave Height Hs (ft)

bull INFORMATION ON DAMAGES PRIOR TO ANDREW

bull RESULTS FROM IN-HOUSE STUDIES IF AVAILABLE

bull ~RIGINAL DESIGN CRITERIA INFORMATION IF AVAILABLE

bull SOIL SHEAR STRENGTH PROFILE IF AVAILABLE

Platforms Identified for Evaluating Therr Likelihood of Falling Under Unexpected Survival Category

OPlRATOR

middotADA

BLOClt

llmucnJllE

NAME

YEAll

INSTAUJD

WA1ER

DBPTll

DISTANCE

JlllOM SHOllE

(MILU)

Bl

(M)

AMOCO 111113

STl11

WDI075

A

bull D

St

17

120

1n

44

20

10 I

CllllVllQfl STlllO c 61 110 21 11 STIUO D Q 1110 21 11

STl134 p Q 137 29 11

ST0151 G 61 137 11

ST0151 I 3 121 11

ST0151 J Q 140 11

STl151 It 137 11

ST0151 PJl001 Q UT 11

ST0151 PROD-2 Q UT 11

ST017 A 3 140 35 11

WDIUT c 65 214 34 10

llXXON STI055 II 51 6T 14 I

STl15 A t3 34

WDI073

WD0073

Amiddotshy Q

1111 1111

22

22

I

I

WDI073 c 1n 22 I

- WDOCl93-shy II

65 1110 2T A 3 194 31 MOBU 111121

SS IOl7

A B

501

50

50

31

T

I

T 6

MtlllPllY SS114 B T 50 T T

PlllllIPS SS0149 A 5 55 33 T

SAMllDAN STln A 107 T T

SBllLL WDl103 A 65 223 2T WDl103 B 65 221 2T WD0104 c 65 221 2T WD0104 D 6T 20 2T

UNOCAL SS020I

SS 020I SSl20I

SS 0209

SS0209

II p

PURll

A

B

3

n 61

103

97

97

95

100

34

34

34

34 34

bullbull bullbullbull

- SS0209 D u 95 34 I

- SS 1209 SSl253

SS0253

G A

c

6T

2

100

165

1T5

34

54

54

bullbull I

I I

I I

SOUTH MARCH

~LAND

N

SOUTH PaTOI ENE ISWltD AREA

MMS Defined Limits ror Offshore Structures Requiring Inspections Following Andrew

t---Jbulln3 T25bull I bullTtl

bull1$1Htl1Abull bull1SIK1779bull

~

middot-DaTA AREA~T

CRANO rsLE

AREA bull1300

bullll4W

=shy

bullPath of Hurricane Andrew with Platforms Used In Calibration

Likelihood Function vs b All Success Case STISI K STIJOQ ST134W WD90A MCJll MCJ97

I 0000

0 9000

0 8000

07000

-Bo6000

tiiii 0 5000 -

04000

0 3000

0 2000

01000

0 0000

0 05 25

ri1rt middot-middot-middot-middot-middot-

I 5 2

b

---11 middot151K(blsu~ccsI

- bull - bull - fk 130Q(bfU~lCSS)

lie bull134W(blsuccc]

lk-WD90A[blsultCS]

- - - lk middotMC397(bfUtlCS]

--o-- lk-MCJll(bbuctcsJ

--- ll (bl 6 out of 6 Mlttcs case]

_

( Likelihood Functions middotSuccess Cases - p - 1 c I

CATEGORIZATION OF ADDITIONAL PLATFORMS

I I I

bull CATEGORIES

Unexpected Survivals

Expected Survivals

bull APPROACH

Identify Corresponding Platform(s) Evaluated In the JIP

Determine Max Wave Height at Platform Location

Evaluate Whether Wave Hit the Deck

middot - Evaluate Likelihood of R and S to Differ from the Base Cases

Review Joint Information

Assess Special Features

bull IDENTIFY NUMBER OF PLATFORMS IN EACH CATEGORY

I

UPDATING OF CALIBRATION RESULTS

bull INCLUDE EFFECT OF ADDITIONAL PLATFORMS ON bullBbull

Consider Increased 1Uncertalntles In R and S

Develop bullNeW Ukellhood Functions for Different Categories

middotLet bullxbull Number of Ukely Unexpected Survival Cases and y Number of Ukely Expected Survival Cases

Determine Combined Likelihood Function lk (b Iadditional platforms) middot

bull REVISED LIKELIHOOD FUNCTION

lk(b I all cases) = lk(b I 9 platforms) middot x lk(b Ix platforms) x lk(b I y platforms)

bull ESTABLISH REVISED BIAS FACTOR B

f8 (b) u f8 (b) lk (b I all cases)

II

t t ~

shy

APPLICATIONS OF BIAS FACTORS

ii APPLICABLE TO PARTICULAR LOAD AND RESISTANCE RECIPE

bull USE BY THE API COMMITTEES AND MMS

Criteria Revisions

Recommended Design Guidelines Revisions bull

bullbull REASSESSMENT STUDIES AND DECISION MAKING

middot

bull RISK ASSESSMENT STUDIES BY THE OPERATORS

Individual Platforms

Fleet of Platforms

middot IMR Program Implementation

bullbull NOT TO BE USED IN DESIGN

I I fl J

EXAMPLE USE OF BIAS FACTORS

bull PHASE I EXAMPLE APPLICATION

bull EVALUATION OF ANNUAL PROBABILITY OF FAILURE

Establish Annual Maximum Seastate Data

Determine Coefficients to Define Base Shear

Determine Ultimate Capacity of Platform

Determine Probablllty of Failure Versus bullbbull

-bull DETERMINE UPDATED PROBABILITY OF FAILURE

Pf - P [S gt R]

Pf I P [~ gt Rb I b] f 8 (b) db

Pr - J P [S gt Rb I b] f bull 8 (b) db

bull TOTAL PROBABILITIES OF FAILURE

Total Pf - 1 bull Il [1 middotPf] - l Pf bull

Total Pf - 1 bull Il [1middot Pf] - l Pfmiddot

Total Pr = 1 - n [1middot PfbullJ = l Pfshy

I I I

EXAMPLE APPLICATION bull PHASE I

PLATFORM ST 151 K

bull GULF OF MEXICO

bull WATER DEPTH = 137 FT

bull ANNUAL MAXIMUM HS MEDIAN 175 ft COV 031

bull ASSUMED STORM DURATION 3 hoursmiddot

bull ASSUMED NUMBER OF WAVES 800

bull MEDIAN HS FOR STORM HOURS 1 2 AND 3

bull MEDIAN CURRENT VALUE bull

bull DETERMINE ULTIMATE CAPACITY FOR THREE DIRECTIONS

bull DETERMINE ANNUAL FAILURE PROBABILITIES

Broadsidemiddot Diagonal End-on

0037 0025 0007

0071 0044 0018

0022 0016 0004

DETERMINE TOTAL ANNUAL PROBABILITY OF FAILURE

Pf - 007

Pf - 013

Pr - 004

bull

EXAMPLE USE OF BIAS FACTORS - PHASE II middot

bull DETERMINE PREDICTIVE PROBABILITY OF FAILURE

Case - 1

Pf _ S S Pf (b b2) fa1a2 (b b2) db db2

g bull -1 (b1 R I S) - 1

92 - (b2 R2 I S) - 1

Case -2

Pf s ~ [bJ f11 (b) db

Pf [bJ p [g lt 0)

g - bull 1 (b1 R1 I S) - 1

1

bull1) l

I

BENEFITS OF ANDREW JIP bull PHASE II

lt

bull IMPROVED LOAD AND RESISTANCE RECIPE

bull IMPROVED CALIBRATION PROCEDURE

bull MULTIPLE BIAS FACTORS

bull PROCEDURE TO INCLUDE EFFECT OF OTHER PLATFORMS

bull PROCEDURE TO USE MULTIPLE BIAS FACTORS

  • Structure Bookmarks
    • ANDREW PHASE II JOINT1 INDUSTRY PROJECT
    • PROGRESS MEETING NUMBER 1
    • ANDREW PHASE II JOINT INDUSTRY PROJECT PROGRESS MEETING NUMBER 1 NOVEMBER 16 AGENDA
    • ANDREW PHASE II UST OF PARTICIPANTS
    • ANDREW PHASE I OBJECTIVES AND RESULTS
    • ANDREW PHASE II PROJECT OBJECTIVES
    • SUBSEQUENT PROJECTS
    • ANDREW PHASE II PROJECT TEAM
    • ANDREW PHASE II UPDATE MODELING middotAND ANALYSIS RECIPE
    • -ANDREW PHASE II CONDUCTOR RESISTANCE
    • ANDREW PHASE Il WA VE FORCE PROFILE
    • ANDREW PHASE II WA VE FORCE PROFILEmiddot
    • ANDREW PHASE II BRACE MODELING
    • ANDREW PHASE II FOUNDATIONS
    • ANDREW PHASE II middot middot JOINT MODELING
    • ANDREW PHASE Il JOINT MODELING OPTIONS DISCUSSION
    • ANDREW PHASE II JOINT MODEJJNG OPTIONS ILLUSTRATIONS
    • ANDREW PHASE II JOINT MODELING PHASE II METHOD
    • ANDREW PHASE II ANDREW STORM CONDmONS UPDATED
    • Platform Selection bullbullbull1111111111111
    • BasisCriteria for Selection bullbullbull1111111111111
    • CALIBRATION
    • WHERE TO INTRODUCE BIAS FACTOR (B
    • PHASE I -KEY CONCLUSIONS
    • bull CALIBRATION ISSUES PHASE II
    • PHASE II WORK SCOPE (CALIBRATION)
    • SOFTWARE
    • SENSITIVITY STUDIES POSTERIOR DISTRIBUTION OF B
    • WEIGHTING APPROACH bull
    • INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON
    • CATEGORIZATION OF ADDITIONAL PLATFORMS
    • UPDATING OF CALIBRATION RESULTS
    • APPLICATIONS OF BIAS FACTORS
    • EXAMPLE USE OF BIAS FACTORS
    • EXAMPLE APPLICATION bull PHASE I PLATFORM ST 151 K
    • EXAMPLE USE OF BIAS FACTORS -PHASE II middot
    • BENEFITS OF ANDREW JIP bull PHASE II
Page 16:  · . ' ANDREW PHASE II PROJECT OBJECTIVES . ' • ·UPDATE MODELING AND ANALYSIS RECIPE· JOINTS ' ' ·BRACES WAVE PROFILE FOUNDATIONS . ' ' CONDUCTORS· • ANALVZE PLATFORMS USING

Project ST151K Model Conductormiddot

Version 1

Clgt - Cut Plane Force Fx Mon Nov 14 140051 1994

- -_ vii I fOIZM

0 _

~

N ID 0 bull 0

x 0

middotrt ~ () bull u 0 ~ 0 Ir

N bull

0

0 1 2 3 4 5 X Deck Disp [Ft)

I I 1 1 WbullTI- bullbull

lltcxi lr-C

011lt~~~~~~~---~~~~~~~~L~~~~~~~--~~~~~~~~i~~~~~~~--

Uniform Node Displacement

shy

Ji

r

Comparison of Pushover Curves for ST 151K Variable Load Pattern and Monotonic Load Pattern (Andrew)

Soll Improved 100

6000

5000 -i 4000

f

llS

3000 Ul

J 2000

10 c11 c1lgtolts 1000

0 v

bull ~

f

_ ----- --- ---shy

bull --Variable Lold

- - Andrew Load

I I I I I I

4 5 62 30 1

Deck Dlsplacement (ft)

bullbull r 1 II

~

ANDREW PHASE Il

WA VE FORCE PROFILE

bull DECK WA VE FORCE ~

PHASE I USED SECTION 17 (1056H degCd)

PHASE Il USE REVISED SECTION 17 (1000H Cd)

bull PHASE I USED SINGLE FORCE PROFILE FOR PUSHOVER

- USED WAVE HEIGHT CLOSE TO ULTIMATE CAPACITY

bull PHASE J

Il WILL STUDY THE

USE OF VARIABLE

PROFILE ~

- IF FORCE LEVEL INCREASES (OR DECREASES) BY A FACTOR EQUIVALENT TO A 2 FOOT CHANGE IN WA VE HEIGHT THE INCREMENTAL FO~CE PA1IERN IS CHANGED ACCORDINGLY (CAP FEATURE)

- INITIAL EXAMPLE

f~

a

middotn

a

I N -a a a amiddot

+

~ a

l

a w gt 0c 0 gt a I shyc -CJ wc w gtc(

== c( I shyZ w E W a 0 z- 8

ii sect

~ ~ bullT 0

I if

x ~

bullbullbull

II ~------------________

ANDREW PHASE II

WA VE FORCE PROFILEmiddot

CAP AUTOMATIC APPLICATION FEATURE

STAT

fR1 0 lt f lt1

STAT

STAT

f(R3-R2) middot f lt1

Project ST151K Model Wave_Loading Version 1 rue Nov 8 164107 1994 PT

E~ nss FT

SL 14amp FT

a 111 FT

EL 14 li

E-1 37 FT

El o FT

1middot1middot ______________ _ ________ __ _ __

0

200 400 600 800 1000 12000 Force at Elevatloii (kip)

200

a ll_

Wave Force at Major Elevations for Discrete Wave Heights

180

bullHmiddot40ft

Hmiddot495ft

H bull 51 ft

Hmiddot55ft_

H-59ft I

Andrew (H bull 60 86 ft) II

i I

_11---__

- middot- Hmiddot30ft

-

- bull -

middot bull bull bull middot

---Hmiddot53ft

bullbullbullbullHmiddot57ft

-----H-63ft

Elevation Above Mudllne 100

(ft)

180

140

120

80

60

40

20

Loading at Major Elevations

---H-49511

-deg-Hmiddot608611

0 200 400 600 800 1000 1200

Force It Elevatlon (kip)

--

Base Shear vs Wave Height ST151K

6000

5000 1---------Ult1mateCapacitybull 4890 kip (100 Improved Soll)-----------~--

4000a a lii ~ 3000

81 ampI

2000

1000

0 +-~~~--i~~~~~~~~~-+-~~~--lf--~~~~~~~~-+-~~~--l

30 35 40 45 50 55 60 65

Wave Height (ft)

bull middot

~

bull

II

2

g

~

It)

0

cCD E 0 E a c E euro~ CDshygtItgtmiddoto ~ti c -a Gic CD gt ~

-= c -Cll

I z

bullO 0i i ~ g 0

() CI -bull(q-ll) iuawow llu1wniiaAO

11

shy

116

114

112

g 110

bullc = 108

~ ~ 106 -sect 104

102

100

98

96

-

6555 60

Wave Height vs Center of Action ST151K

30 35 40 45 50

Wave Height (ft)

t

Comparison of Pushover Curves for ST 151K Variable Load Pattern and Monotonic Load Pattern (Andrew)

Soll Improved 100

--- --------- --- ----6000

5000

- 4000 8 - I

1 3000 Ul

81 m 2000

_J -shy

1000

---Variable Load

- - Andrew Load

0 f-~~~~~r--~~~~-t-~~~~~-t-~~~~~+-~~~~----t~~~~~-1

0 1 - 2 3 4 5 6

Deck Displacement (ft)

-middotshy

1 t I I ~

ANDREW PHASE II

BRACE MODELING

bull MARSHALL STRUTS USED FORPHASE I

bull PHASE II WILL ASSESS DIFFERENCE IN USING BUCKLING BEAMS AND MARSHALL STRUTS

middot bull COMPARE FOR 2D FRAME

bull COMPARE FOR PLATFORM 177 WHERE LOWER

BAY EXHIBITS LARGE BRACE MOMENTS

Project andrew Model beamframe Version 1 Mon Nov 14 145447 1994

~

middot Frame Test buckling beam-column model

-

Project andrew Model strutframe Version 1 Mon Nov 14 145754 1994

bull

Frame Test strut model

middot-middot _____________--_

_middotmiddotr_~ --ll--~-middot__

Buckling Beam bull

Strut (K=O 65)

Lateral Displacement

ANDREW PHASE II

FOUNDATIONS

bull ~LICIT NONLINEAR API FOUNDATION USED INiPHASE I

bull FOUNDATIONS

JIP SELECTED MINI-VANE DATA FOR

STRENGTH PROFILE

bull COMMENTS ON BENCHMARK COMPARISONS

- SIGNIFICANT VARIABILITY DUE TO FOUNDATION MODELING METHODS

II Ii

z

ctV 4x SPlO caissonfy-42s soil-api(lf-097step-114)

Inelastic Events Legend Elastic Strut Buckling

Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield

--------shyBeam Clmn Fully Plastic Fracture

~--~~~~~~~~~--~----~__i~~---~~~~__--~

Pro)ect apipile Model South Pelto 10 Version Wed May 18072235 1994 - -

1bullt

+

t

f i bullbullbull

bullbull bullbull

I I

5

~x SPlO Caissonfy=42s soil=api(lf=lOOstep=137)

Inelastic Events Legend

Elastic Strut Bucklilg

Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield

earn Clmn Full~ Plas1c Fracture

ANDREW PHASE II middot middot

JOINT MODELING

middotbull API RIGID-PLASTIC USED IN PHASE I

a ELASTIC-BRITTLE USED IN PHASE IB

amiddot MSL WILL PROVIDE NONLINEAR FORCE-DEFORMATION

AND MOMENT-DEFORMATION RELATIONS FOR PHASE II

a MODELING OPTIONS FOR PHASE II -

l

- UNCO~LEDmiddotONE DIMENSIONAL MEMBER END JOINT MODELING

J Ii

ANDREW PHASE Il

JOINT MODELING OPTIONS DISCUSSION

bull JD COUPLED INTERACTION SURFACE

- IDEAL

- DATA NOT AVAILABLE TO CALIBRATE

bull 2D COUPLED INTERACTION SURFACE

- SIGNIFICANT MODELING IMPROVEMENT

- DATAAVAILABLE

- IMPLEMENTATION PRACTICAL BUT NEEDS DEVELOPMENT AND CALIBRATION

bull 2D UNCOUPLED INTERACTION SURFACE

- SIGNIFICANT MODELING IMPROVEMENT

- DATA AVAILABLE

- IMPLEMENTATION PRACTICAL AND AVAILABLE

ANDREW PHASE II

JOINT MODEJJNG OPTIONS ILLUSTRATIONS

P

e

ANDREW PHASE II

JOINT MODELING PHASE II METHOD

e

I I I

ANDREW PHASE II

ANDREW STORM CONDmONS UPDATED

bull -PHASE I USED OCEANWEATHER HINDCAST-(Nov 1992)

bull middotmiddotPHASE II WILL USE THE REVISED HINDCAST (Nov 1994)

bull lt c lt I lt

2-2

l---4-~~~+-~~-+-~~~1--~~-+-~~--l--~T--+-~~--i24

f-tJVtMfiz~ 1Cfl (r STvb(

Plollol bullbull 5middotNOYmiddot9214bull1J Irbullbull Illbull [ANDREVl~NDREVCUSTbullMAXll 5-NOY-1992 10bull54 - 9208

HEIGHT lml

I~ MAXI

I L 1~middotamp SZ~- ~A I 13c

I J) ( r I X~0iWJ~0gtJ gt L gt gt I 1c I 120

-- N

~ deg

I I I I I I e C I ~=-41 122 -94 -92 -90 -88 -86 -84 -82 -80

-

Figure 7 contours of maximum hindcast significant wave height

_

~

~tl ~

NOltNttIJ 17lt ltNS S iJJ)y

Maximum Significant Wave Height - Hurricane Andrew 1992

30 ~

26 ~ I ~-~~~- == -shy -~ 35-_

24deg1--shy --------~- middot-middot-------middotmiddotmiddot----middot------middot-middot~middot- - - ~-----1---middot

22middot ~~~~~~~-~ -94 -92 -90 -88 -84deg-86 -82 -somiddot

c

-

Platform Selection bullbullbull1111111111111

0 BasisCriteria for Selection middot

0 Review of Phase I Platforms middot 0 Alternative Platforms

0 PMB Recommendations

0 Participants Comments

gt

L

l I

BasisCriteria for Selection bullbullbull1111111111111

0 Benefit for Calibration - Classification (UnexpectedExpected)

- Failure Mode (StructuresFoundations)

- Damaged StructuresKnown response

- Better Representation of Full ~opulation

0 Availability of Data - Existing Models

- Inspection Data

- Damage Data

PREDICTED BEHAVIOR

-

OBSERVED

BEHAVIOR

-SURVIVED DAMAGED FAILED

SURVIVED

ST134W WD90A

-MC311 -

MC397 -

--

-

-

ST151K ST130Q

-

-

-

DAMAGED - -

-

-

--

-

-

ST52 SS139 ST177B -

ST161A

-_

-

FAILED

-

-

-

ST151H ST130A ST72

Operator - Platfonns - Platfonns Survived Damagedamp

- Falled -

bull

-

-

Jle eBDIGIPllDbl -Amocoshy 30 1 Chevron -

143 12 Exxon 72 Mobll 22 1 Phllllps 7 bull

-Shell 39 Unocal 25 1

Total Partlclnts Platfonns 338 15

Other Operators bull- 310 13

-

Total Platfonns - -

648 28

fbml

Platforms Considered 6 7

tie of Total Platfonns 1 25

OVERVIEW OF PLATFORMS SURVIVEDDAMAGEDFAILED

-

shy

v

Platforms Evaluated in Phase I

-

-Platfonn Water Year Number Analysis Avallabillty of Availability Platform

- Name Depth rl of Model DetaDed rl ClassUlcatlon

Installation legs Inspection Soll Report Based on Information Phase I Calibration - -ft

- -- Suaill flalfnrm ~BSi -STISIK 137 1963 8 CAP Yes Yes Un eel Sarvlval

-STIJOO 170 1964 4 CAPmiddotshy Yes ST134data u ST134W 137 1981 4 CAP Yes Yesmiddot E

-

edSarvlval -

middotsurvival

WD90A 184 - 1964 8 CAP Yes E -

- Sarvlvill

MC3ll 343 1978 8 KARMA

MC397 468 1991 4 KARMA -

bull ~reglmmiddot- -

Yes Sure Survival

bull Yes - Sare Sanlval-

-T23STS2 63 1969 4 CAP

T2SSS139 62 1969 4 CAP -

Yes

Yes

Yes Ex

Yes Une

bull SurvivalLikely to DamalI

middot bull SanlvalLlkely to DamalI

ST161A 118 1964 8 USFOS

Eail11a pound1al(bullllm Clss ST177B 142 --1965 8 CAP

STISIH 137 1964 - 8 CAP C

STIJOA 140 19S8 8 CAP - -

Yes

Yes

ollamed

Yes

- Yes E

-

at about 1 mUe

Yes

ST134data

bull SafvlvalLlkelv to Dam- bull- -

E middotrahre

Emeried Failure -

Likely FaUare

T21ST72 61 1969 4 CAP Collamed No Likely FaUure

Platforms Identified for Evaluating Their Likelihood of Falling Under UnexpeCted Survival Category

CllEllATOa

AllB

llLOCll

Sl1lvcnJllE

NAME

YEAR

INSTAUJCD

WATER

IJBPJB

middotmiddotDISTANCE

ftOMSllOllE

(MIUIS)

Ill

(M)

AMOCO BllW

STllI

WDI075

A

bull D

4

n uo 17J

44

32

bull 20

10

bull CllEVllOH STtllO c I 1bull 21 11

STl130 D 0 161 21 11 STllJ4 I 0 137 D - 11

STIUI G I 137 32 11

STIUI I 0 121 32 11

STIUI

J 0 140 32 11

STIUI It 4 137 32 11

STllSI PROl)I 0 137 32 11

STllSI PROD-2 0 137 32 11

STll7 A 0

140 35 11

WDl117 c 65 214 34 10

DXON STshy E 51 14 bull STl165 A t3 bull 34 WD0073

WD0073

WD0073

A _ c

4

62

4

IA IA

17J

22

22

22

bullbullbull WD003-shy E 65 161 Z7

A 0 IN 31 MOBD Bl12I

ssn A bull

50

50

31

7

bull 7

MIJllPllY SS114 B 7 50 7 7

PlllUIPS SSl14t A 55 33 7

SAMEDAN ST17J A 4 117 7 bull 7

SJIELL WD0103 A 65 223 27 WDllOJ bull 65 221 27 WDll04 c 65 221 27 WDOI04 D 67 2 27

UNOCAL

SSl20I

SSl20I

bull SSl20I

SS0209

SS0209

SS0209

SS0209

SS 0253

SS 0253

E

I

PtJllE

A

bull D

G A

c

0

4

4

7J

61

0

0

69

103

97

97

9S

100

95

100

165

175

34

34

34

34

34

34

34

54

54

bullbullbullbullbullbullbullbullbull

Platforms Recommended for Phase II

llalr y_ w- Aot Av- _ I 1 Doplla Model nm bull bull _ bull I I H11 s18-1 Irht lbwl~ bullbullbull(-illtdlel OM lllMtD

II ---

-STISIIC 137 1963 I CAP Yeo Yeo I Yeo

STtshy -170 1964 4 CAP Yeo ST134dlU I Yeo

STl34W 137 Yeo Yeo -1981 4 CAP D Yeo

WD90A 114 1964 I CAP Yeo - D

MC311 30 1971 I KARMA Yeo -sre-

MC397 middot 461 1991 4 ICARMA Yeo Sireshy - - r- -

__ T13ST5l 63 1969 4 CAP Yeo Yeobull middot- ~n- D Yeo _ Tl5SSl39 62 1969 4 CAP Yeo Yeo I Yeo

_ STIT711 142 1965 I CAP Yeo I Yeo middot-1shySTl61A Ill 1964 I USFOS Yeo Yeo - D bulloshy

- r-

STISIH 137 1964 I CAP y D y -_ STl31A 140 1951 I CAP Yeo STl34dlU I Yeo

T21Sll 61 1969 4 CAP No rar-middot I Yeo

~-rmiddot-middot SS209A 95 1967 16 CAP Yeo D -SSU4R

11 UI A 41 -in D SS 2741AIM CAP bull

_ _ -shy - -shy- -shy

CALIBRATION

bull PHASEIAPPROACH

bull CALIBRATION ISSUES

bull PHASE II WORK SCOPE

bull SOFlWARE AVAILABILITY

bull EFFECT OF ADDITIONAL PLATFORMS

bull WEIGHTING APPROACH

bull EXAMPLE APPLICATION middot

(l=Ofce Mode] Wind

~

fcurrent l -middot

~SW ~leldl4

ultimate resistance force I first

design load Illgt

i _

deflection

-

bull

4

Historical Hurrlcane Dariiage and Survival of Platforms

Components Required for Calibration (Petrauskas 1992)

PRIOR Distribution

PLATFORM EXPERIENCE

BAYESIAN Survivals and Failures

UPDATING

Waves Winds and Currents bull

POSTERIOR Distribution

- A_middotJA ~ Survival Data Failure Data

Result Result

Load Resistance

middotfl

-

Bayesian Updaing middotGeneral Description (Petrauskas 1992)

L Imiddot

WHERE TO INTRODUCE BIAS FACTOR (B

On Resistance (Rmue - B (RPREDICreD I

middotOn Loads (Smue middot - B (SPRE01cTEo

On Safety Margin (RSmue - B (RSPREDICTED

CALIBRATION METHODOLOGY

gt

r

FIELD OBSERVATIONS

Observed Behavior

survivals Damages

Failures

r

bull

CAPACITY ANALYSIS

Predicted Load and Resistance Predicted Failure Modes

Survival Damage Failure cases

RELIABILITY ANALYSIS

- Probability of Failure ~

Ukellhood Failure

BAYESIAN UPDATING

Prior Distnbution of B Combined Likelihood Function

Posterior Dlstnbution of B

BIAS FACTOR

Mean Value cov bull

Represents Modeling Uncertainties

Prior Distribution of B Likelihood Functionmiddot Observing Success case STISIK and Posterior Distribution of B

2 00

uo

160

140

120

e 100 L $ uoIshy

0 60 ~ ~ 040

020

0 00

0 00 0 50

_ I 00

b

I 50 2 00

-shy lk [bl STISIK)

--shy Pnor-f(b)

--shy Pos1enor r(bl ISIK)

Prior or B Mean bull 10 COV bull 030

Posterlor of Bbull Meanbull 136COV 0 16

2 50

Likelihood Function and Posterior Distribution Example Platform STISIK

- -

~

- Prior and Posterior Distributions or B Failure cases (ST177B STlSlH ST130A T21) Damage cases (T23 T25 ST161A) Success cases (STlSlK ST130Q ST134W WD90A MC311 MC397)

400

350

300

2 so 0-

I

bull bull

bull bull gt

- bullbull ~ bull bull I bull

_ bull bull bullbull I bull bull bullII V -

bull bull

I bull ~ bull

bullbull bullI bull bullI bull bull bull

middot -~

bull__ ~ middotI middotmiddot

L 2 00

e ~ I SO

-I 00

- 0 so

000 gt

0 00 020 0 40 060 0 80 I 00 120 140 160 110 200

b -

-

shy

Prior-r

- - - - Posterior fLlll failure cases

bull bull bull bull bull bull bull Posterior flall damage cases

- bull bull - bull middot Postenor flall success cases

Posterior flall success fulure damage cases

-

Posterior of B Mean= 119 COV 0101

Posterior Distribution or Blas Factor (8) bull Different Categories and All Cases Combined

shy

l

PHASE I - KEY CONCLUSIONS

bull GLOBAL CORRECTION FACTOR

bull 15 TO 20 CONSERVATISM MEAN)

bull UNEXPECTED SURVIVALS HAVE GREATEST IMPACT

bull DAMAGED CASES MORE INFORMATIVE

bull CALIBRATION ISSUES PHASE II

bull

EFFECT OF IMPROVED CAPACITY ANALYSIS RECIPE

bull APPLICABILITY TO MULTIPLE FAILURE MODES

bull NO FOUNDATION DAMAGE CASE INCLUDED

bull IMPROVED REPRESENTATION OF FULL ANDREW DISTRIBUTION

-bull NUMBER OF PLATFORMS CONSIDERED

I 1 I

I t ) bull ~

PHASE II WORK SCOPE (CALIBRATION)

bull

bull ESTABLISH MULTIPLE BIAS FACTORS

Use Nine Base Case Platforms

middot - Capacity Analysis for Revised recipe

Foundation Blas Factor from Foundation JIP

bull DEVELOP WEIGHTING APPROACH

bull UPDATE MULTIPLE BIAS FACTORS

bullEffect of Platforms Not In the Nine Base Cases

bull EXAMPLE APPLICATION OF BIAS FACTORS I

bullbull

PHASE I PHASE II

CAPACITYbull ANALYSISRu bull

---------------------1----middot

Jc1 - iJbullraquoJ dbull

FttlS frt[f~l ratdhJJ

-

LIKELIHOOD FUNCTION

DEVELOPMENT

lk Jb Jfall1N) bull Jb)

111 Ill Isurvtvoq bull 1 - bl

-middot -------------------shy ~---------------------middot-middot

r111ir8 Cb1 111 Cbl BAYESIAN

UPDATING

~cove

R1R2 bull middot---------------------~----

Jblbull P(g cOJ

1 bull JbRI (MBSI I bull 1

~middot

llcJllJfalluro) bull P(gcOJ Ill 111 b21fallun) bull P ((g1 CO) U 1112 c OD

middot--------------------shy ----------------------shy bull

r111111r8 Cbl 111 Ill i-1nro1

deg ~1b2)laquottfrr (b1b21-lnfo) re Ill bull 1-11middot b2) bull lodegamp

middot

i1 covamp i2 cov~i

-

shy

--

Likelihood Function vs b All Success Case STISI K STl30Q STl34W WD90A MC311 MC397

I 0000

0 9000

OIOOO

0 7000

~ 0amp000

ta osoOO

~ 0 4000

0 3000

0 2000

0 1000

bull 0 0000

bull

I

middot

middot bull I

Ibull

-sibull ~ -middot-middot-middot-middot -middot-

----7----+-----~ I

Ibull bull bull I

rbull

--- Jk ISIK[blsuccess)

- bull - bull - lk 130Q[blsuccess) bull

lk 134W[blsuccess)

Jk-WD90A(blsuccc~I

- - - lk MC397[blsucccss)

--ltgt-- lk middotMC31 l[blsucccss)

--- Jk [bl 6 out of6succcsHascs)

05 I 5 2 250

b

Likelihood Functionsmiddot Success Cases - Phbullsc L

shy

bullbull bull bull bull bull bull -------------------~

bull bull bullbull bullbull bullbull bullbull bullbull bullbull bull

bull bullbull

bullbullbullbull

bullbull

bullbullbullbullbullbullbull

bullbullbullbullbullbull

bullbullbullbullbullbullbull

z 0

t z ) LL

c 0 0c i ~ i

J

~ sect ~--------------~ e l5

I

I

SOFTWARE

bull PF PROGRAM

bull CALREL

bull PF +CALREL

bull RELACS

bull

gt i

l

SENSITIVITY STUDIES

POSTERIOR DISTRIBUTION OF B

bull EFFECT OF CHANGE IN DISTRIBUTION OF PRIOR OF B

bull EFFECT OF REMOVAL OF ONE PLATFORM FROM EACH CATEGORY

bull EFFECT OF REMOVAL OF THREE DAMAGE CASES

bull EFFECT OF ADDITIONAL 13 SIMILAR PLATFORMS

-I

Prior and Posterior Distributions of B Variation In COV of PriorofB fromOl to04 -- -

4500

-4000

-

- -

3500 --middot - bull bull - bull bull Prior-r ICOV =02

3000

- - - - Prior-r ICOV =03 -

bull -~ e 2soo -

bull bull bull bull bull bull bull Pnor-r ICOV =o 4t frJ ~ 2000 Posterior rlCOV =0 2

bull bull Postenor rlCOV =0 3 bull bull 1500 bull middot bull - shy -middot- Posterior rlCOV =04

middot~ I 000 - bull

-middotmiddot ~ ~middotmiddot M -~

- middot 0 500

wbullbullbull _ I middot~ middotmiddotmiddotmiddotmiddot middotmiddotmiddotmiddotmiddot - -~ ~ __ middot_ _ bull -shy0 000

0 000 0200 0 400 0 600 0 800 IOOO 1200 1400 1600 1800 2000 -

- - b

-

=shy

Posterior Distribution of Bias Factor (B) bull Effect of Variation In COV of Prior of B

bull

bullPrior and Posterior Distributions of B Evaluating Effect of Removal of One Platform from Each Category from the Base Case of 13 Platforms

400

~ ~ 350

j IA i

3 00 bull I bull I ---Prior-r I bull l

250 1 I 1

I bull bull ---Posterior r lall 13 casesshy- ~E 2 00 bull bull bull bull bull bull bull Posterior r without STISIK

- ~ 1 bull - bull I - bull bull - bull bull Posterior r Iwithout T2I I 50 I

J-tt~i - bull - bull - Posterior r Iwithout T2S

tOO [-

I I ~ ~ I 0 50

0 00

___ r J i

l~T

~--

0 00 0 20 0 40 0 60 080 I 00 120 140 160 110 2 00

b

zshy

Posterior Distribution of Bias Factor (8) bull Effect of Removal of middotPlatform from Any Category

bullPrior and Posterior Distributions or B Evaluating Effect or Removal or 3 Damage Cases from the Base Case or 13 Platrorms

4

35

3

25

e L 2 ~ bulle shy

I 15

bull I

05

0

0 02 04 06 I 2 I 4 16 IS 2

I I

I

I

I Ibull

J -tmiddot L--

bullI

~

I

fl

II ~

I I v I

middot bull bull

bull L

~ ll- I-~lt

OS I

b

I

bull

1middot bullbullbullbull

~ __

bull

- - - - Prior-r

- - - - Postenor flall failure cases

Postenor flall success cases

--- Posterior flall 13 cases

---Posterior fI success failure cases only

IPosterior or B Mean =t11 COV =0121

-

_I

Posterir Distribution or Bias Factor (B) bull Effect or Removal or 3-Damage Cases

-

bullPrior and Posterior Distributions or B Evaluating Effect or Additional 13 Similar Observations as

Used in Andrew JIP The POsterior or B from Andrew-JIP Considered as New Prior or B

- 500

450

400

350

bull bull

bull bull bull bull J bull

bull bull f ~ I

I ~ Ibull bull

bullbull bull

l bull I -bullbull bull bull

I

I bull bull

bull I I~~bull bull bull

bull ~

I bull bull

~ ~_ ~~~

- - New Prior r bull

Posterior 114 new failure cases ~ 300- bull bull bull bull bull bull bull Posterior f 13 new damage casesl 150 ~

- bull bull - bull bull Posterior 116 new success cases i 2 00

Postenor 1113 new success failure -150 damage cases

I 00

0 50 -

0 00 0 00 0 20 0 40 0 60 0 80 I 00 120 I 40 160 110 2 00

Posterior ol B Mean= 120 COV= 0071 b

shy

shy

_rshy

Posterior Distribution of Bias Factor (B) bull Effect of Additional 13-Similar Observations

1

1 I

WEIGHTING APPROACH bull

bull ASSESSMENT OF PLATFORMS DATA

Select Platforms Most Mappable to the Analyzed Platforms

Unlikely SurvivalDamage Cases Identification

Based on Cut-Off Criteria Agemiddot Hs etc

bull INFORMATION REQUIRED FOR COMPARISON

bull CATEGORIZATION OF ADDITIONALmiddot PLATFORMS

bull UPDATING OF CALIBRATION RESULTS

--

OVERVIEW OF PLATFORMS SURVIVEDDAMAGED -~

Operator -

- Platforms

Survived CS)

Platforms

Damaged CD)

Platforms Used In

Phase IJIP -

JI~ ~Bill~IRBDll -

Amoco -

Chevron -

Exxon

Mob II

Phllllps

Shell

Unocal -

30

143

72 22

7

39 25shy -

1

12

bull

1 bull

1

1(S)+1 (D)

3 (S) +3(D)

2 (S)

bull shy

bull

bull

bull

Total Partlclnants Platforms 338 - 15 10C6S+4D) -

Other Operators

310

13

3 (D)

Total Platforms 648 28 13C6S+7D) -

Summary of Andrew JIP bull Phase I Platforms

Plairshy w v- N-ber Lea-le 1 shy car Hind_ o-w N-bull Deplh fl fl IJpe la bull n Yeadon Jadtet lbelledtf liefalit -- _ -OD

IL H - -shy _ICllshy IL IL - - rQClto ~ _

K STUIK 137 1963 8 365 608S4 Cellardeck

-middot STl-shy 170 1964 4 -~ K 39 62269 Cellardeck - --ST134W 137 1911 4 43 608S4 Smvlnlshy- _ - -

WD90A 184 1964 8 K 267Sbull34 83 SOS31 SUboftllar deck SmvMI

MC311 343 1978 8 x S7 62 - MC397 468 1991 4 x shy S33bull6SO 65911 -shySmeSmvlvol -- gt -

r-- T23ST52 63 1969 4 ton- shyKshy 52594 S03 -T25SSl39 62 1969 4 Kshy SOS94 S063 - middot ton-shy__ -

bullbull a ton--shyST161A 118 1964 8 K - 347S 51495 Cellarclecti

r~

ST1771 142 1965 8 K 39 60167 CeDordeck failme

STISIH 137 1964 8 K 395 608S4 CeDardeck Failme

K ST130A 140 1958 a SS 6095 I lblv Pai1uno - T21ST72 61 1969 4 Kshy S4S94 4974 I lhlv Pai1uno -

- -

JE

INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON

bull WATER DEPTH

bull bull YEAR OF INSTALLATIONDESIGN

bull PLATFORM ORIENTATION

bull PLATFORM LATITUDE AND LONGITUDE

bull KEY SKETCHES OF PLATFORM STRUCTURAL FRAMING

Number of Legs Plies Leg Batter Number of Conductors bull Deck Elevations (Including sub-cellar deck If any) Jacket Framing Elevations Brace type In Vertical Frames Sizes of Primary Members (leg pile vertical frames braces primary horizontals) Joint Details (can gap etc) LegPiie Annulus Grouting Status

bull middotPOST ANDREW INSPECTION REPORT Significant Wave Height Hs (ft)

bull INFORMATION ON DAMAGES PRIOR TO ANDREW

bull RESULTS FROM IN-HOUSE STUDIES IF AVAILABLE

bull ~RIGINAL DESIGN CRITERIA INFORMATION IF AVAILABLE

bull SOIL SHEAR STRENGTH PROFILE IF AVAILABLE

Platforms Identified for Evaluating Therr Likelihood of Falling Under Unexpected Survival Category

OPlRATOR

middotADA

BLOClt

llmucnJllE

NAME

YEAll

INSTAUJD

WA1ER

DBPTll

DISTANCE

JlllOM SHOllE

(MILU)

Bl

(M)

AMOCO 111113

STl11

WDI075

A

bull D

St

17

120

1n

44

20

10 I

CllllVllQfl STlllO c 61 110 21 11 STIUO D Q 1110 21 11

STl134 p Q 137 29 11

ST0151 G 61 137 11

ST0151 I 3 121 11

ST0151 J Q 140 11

STl151 It 137 11

ST0151 PJl001 Q UT 11

ST0151 PROD-2 Q UT 11

ST017 A 3 140 35 11

WDIUT c 65 214 34 10

llXXON STI055 II 51 6T 14 I

STl15 A t3 34

WDI073

WD0073

Amiddotshy Q

1111 1111

22

22

I

I

WDI073 c 1n 22 I

- WDOCl93-shy II

65 1110 2T A 3 194 31 MOBU 111121

SS IOl7

A B

501

50

50

31

T

I

T 6

MtlllPllY SS114 B T 50 T T

PlllllIPS SS0149 A 5 55 33 T

SAMllDAN STln A 107 T T

SBllLL WDl103 A 65 223 2T WDl103 B 65 221 2T WD0104 c 65 221 2T WD0104 D 6T 20 2T

UNOCAL SS020I

SS 020I SSl20I

SS 0209

SS0209

II p

PURll

A

B

3

n 61

103

97

97

95

100

34

34

34

34 34

bullbull bullbullbull

- SS0209 D u 95 34 I

- SS 1209 SSl253

SS0253

G A

c

6T

2

100

165

1T5

34

54

54

bullbull I

I I

I I

SOUTH MARCH

~LAND

N

SOUTH PaTOI ENE ISWltD AREA

MMS Defined Limits ror Offshore Structures Requiring Inspections Following Andrew

t---Jbulln3 T25bull I bullTtl

bull1$1Htl1Abull bull1SIK1779bull

~

middot-DaTA AREA~T

CRANO rsLE

AREA bull1300

bullll4W

=shy

bullPath of Hurricane Andrew with Platforms Used In Calibration

Likelihood Function vs b All Success Case STISI K STIJOQ ST134W WD90A MCJll MCJ97

I 0000

0 9000

0 8000

07000

-Bo6000

tiiii 0 5000 -

04000

0 3000

0 2000

01000

0 0000

0 05 25

ri1rt middot-middot-middot-middot-middot-

I 5 2

b

---11 middot151K(blsu~ccsI

- bull - bull - fk 130Q(bfU~lCSS)

lie bull134W(blsuccc]

lk-WD90A[blsultCS]

- - - lk middotMC397(bfUtlCS]

--o-- lk-MCJll(bbuctcsJ

--- ll (bl 6 out of 6 Mlttcs case]

_

( Likelihood Functions middotSuccess Cases - p - 1 c I

CATEGORIZATION OF ADDITIONAL PLATFORMS

I I I

bull CATEGORIES

Unexpected Survivals

Expected Survivals

bull APPROACH

Identify Corresponding Platform(s) Evaluated In the JIP

Determine Max Wave Height at Platform Location

Evaluate Whether Wave Hit the Deck

middot - Evaluate Likelihood of R and S to Differ from the Base Cases

Review Joint Information

Assess Special Features

bull IDENTIFY NUMBER OF PLATFORMS IN EACH CATEGORY

I

UPDATING OF CALIBRATION RESULTS

bull INCLUDE EFFECT OF ADDITIONAL PLATFORMS ON bullBbull

Consider Increased 1Uncertalntles In R and S

Develop bullNeW Ukellhood Functions for Different Categories

middotLet bullxbull Number of Ukely Unexpected Survival Cases and y Number of Ukely Expected Survival Cases

Determine Combined Likelihood Function lk (b Iadditional platforms) middot

bull REVISED LIKELIHOOD FUNCTION

lk(b I all cases) = lk(b I 9 platforms) middot x lk(b Ix platforms) x lk(b I y platforms)

bull ESTABLISH REVISED BIAS FACTOR B

f8 (b) u f8 (b) lk (b I all cases)

II

t t ~

shy

APPLICATIONS OF BIAS FACTORS

ii APPLICABLE TO PARTICULAR LOAD AND RESISTANCE RECIPE

bull USE BY THE API COMMITTEES AND MMS

Criteria Revisions

Recommended Design Guidelines Revisions bull

bullbull REASSESSMENT STUDIES AND DECISION MAKING

middot

bull RISK ASSESSMENT STUDIES BY THE OPERATORS

Individual Platforms

Fleet of Platforms

middot IMR Program Implementation

bullbull NOT TO BE USED IN DESIGN

I I fl J

EXAMPLE USE OF BIAS FACTORS

bull PHASE I EXAMPLE APPLICATION

bull EVALUATION OF ANNUAL PROBABILITY OF FAILURE

Establish Annual Maximum Seastate Data

Determine Coefficients to Define Base Shear

Determine Ultimate Capacity of Platform

Determine Probablllty of Failure Versus bullbbull

-bull DETERMINE UPDATED PROBABILITY OF FAILURE

Pf - P [S gt R]

Pf I P [~ gt Rb I b] f 8 (b) db

Pr - J P [S gt Rb I b] f bull 8 (b) db

bull TOTAL PROBABILITIES OF FAILURE

Total Pf - 1 bull Il [1 middotPf] - l Pf bull

Total Pf - 1 bull Il [1middot Pf] - l Pfmiddot

Total Pr = 1 - n [1middot PfbullJ = l Pfshy

I I I

EXAMPLE APPLICATION bull PHASE I

PLATFORM ST 151 K

bull GULF OF MEXICO

bull WATER DEPTH = 137 FT

bull ANNUAL MAXIMUM HS MEDIAN 175 ft COV 031

bull ASSUMED STORM DURATION 3 hoursmiddot

bull ASSUMED NUMBER OF WAVES 800

bull MEDIAN HS FOR STORM HOURS 1 2 AND 3

bull MEDIAN CURRENT VALUE bull

bull DETERMINE ULTIMATE CAPACITY FOR THREE DIRECTIONS

bull DETERMINE ANNUAL FAILURE PROBABILITIES

Broadsidemiddot Diagonal End-on

0037 0025 0007

0071 0044 0018

0022 0016 0004

DETERMINE TOTAL ANNUAL PROBABILITY OF FAILURE

Pf - 007

Pf - 013

Pr - 004

bull

EXAMPLE USE OF BIAS FACTORS - PHASE II middot

bull DETERMINE PREDICTIVE PROBABILITY OF FAILURE

Case - 1

Pf _ S S Pf (b b2) fa1a2 (b b2) db db2

g bull -1 (b1 R I S) - 1

92 - (b2 R2 I S) - 1

Case -2

Pf s ~ [bJ f11 (b) db

Pf [bJ p [g lt 0)

g - bull 1 (b1 R1 I S) - 1

1

bull1) l

I

BENEFITS OF ANDREW JIP bull PHASE II

lt

bull IMPROVED LOAD AND RESISTANCE RECIPE

bull IMPROVED CALIBRATION PROCEDURE

bull MULTIPLE BIAS FACTORS

bull PROCEDURE TO INCLUDE EFFECT OF OTHER PLATFORMS

bull PROCEDURE TO USE MULTIPLE BIAS FACTORS

  • Structure Bookmarks
    • ANDREW PHASE II JOINT1 INDUSTRY PROJECT
    • PROGRESS MEETING NUMBER 1
    • ANDREW PHASE II JOINT INDUSTRY PROJECT PROGRESS MEETING NUMBER 1 NOVEMBER 16 AGENDA
    • ANDREW PHASE II UST OF PARTICIPANTS
    • ANDREW PHASE I OBJECTIVES AND RESULTS
    • ANDREW PHASE II PROJECT OBJECTIVES
    • SUBSEQUENT PROJECTS
    • ANDREW PHASE II PROJECT TEAM
    • ANDREW PHASE II UPDATE MODELING middotAND ANALYSIS RECIPE
    • -ANDREW PHASE II CONDUCTOR RESISTANCE
    • ANDREW PHASE Il WA VE FORCE PROFILE
    • ANDREW PHASE II WA VE FORCE PROFILEmiddot
    • ANDREW PHASE II BRACE MODELING
    • ANDREW PHASE II FOUNDATIONS
    • ANDREW PHASE II middot middot JOINT MODELING
    • ANDREW PHASE Il JOINT MODELING OPTIONS DISCUSSION
    • ANDREW PHASE II JOINT MODEJJNG OPTIONS ILLUSTRATIONS
    • ANDREW PHASE II JOINT MODELING PHASE II METHOD
    • ANDREW PHASE II ANDREW STORM CONDmONS UPDATED
    • Platform Selection bullbullbull1111111111111
    • BasisCriteria for Selection bullbullbull1111111111111
    • CALIBRATION
    • WHERE TO INTRODUCE BIAS FACTOR (B
    • PHASE I -KEY CONCLUSIONS
    • bull CALIBRATION ISSUES PHASE II
    • PHASE II WORK SCOPE (CALIBRATION)
    • SOFTWARE
    • SENSITIVITY STUDIES POSTERIOR DISTRIBUTION OF B
    • WEIGHTING APPROACH bull
    • INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON
    • CATEGORIZATION OF ADDITIONAL PLATFORMS
    • UPDATING OF CALIBRATION RESULTS
    • APPLICATIONS OF BIAS FACTORS
    • EXAMPLE USE OF BIAS FACTORS
    • EXAMPLE APPLICATION bull PHASE I PLATFORM ST 151 K
    • EXAMPLE USE OF BIAS FACTORS -PHASE II middot
    • BENEFITS OF ANDREW JIP bull PHASE II
Page 17:  · . ' ANDREW PHASE II PROJECT OBJECTIVES . ' • ·UPDATE MODELING AND ANALYSIS RECIPE· JOINTS ' ' ·BRACES WAVE PROFILE FOUNDATIONS . ' ' CONDUCTORS· • ANALVZE PLATFORMS USING

Ji

r

Comparison of Pushover Curves for ST 151K Variable Load Pattern and Monotonic Load Pattern (Andrew)

Soll Improved 100

6000

5000 -i 4000

f

llS

3000 Ul

J 2000

10 c11 c1lgtolts 1000

0 v

bull ~

f

_ ----- --- ---shy

bull --Variable Lold

- - Andrew Load

I I I I I I

4 5 62 30 1

Deck Dlsplacement (ft)

bullbull r 1 II

~

ANDREW PHASE Il

WA VE FORCE PROFILE

bull DECK WA VE FORCE ~

PHASE I USED SECTION 17 (1056H degCd)

PHASE Il USE REVISED SECTION 17 (1000H Cd)

bull PHASE I USED SINGLE FORCE PROFILE FOR PUSHOVER

- USED WAVE HEIGHT CLOSE TO ULTIMATE CAPACITY

bull PHASE J

Il WILL STUDY THE

USE OF VARIABLE

PROFILE ~

- IF FORCE LEVEL INCREASES (OR DECREASES) BY A FACTOR EQUIVALENT TO A 2 FOOT CHANGE IN WA VE HEIGHT THE INCREMENTAL FO~CE PA1IERN IS CHANGED ACCORDINGLY (CAP FEATURE)

- INITIAL EXAMPLE

f~

a

middotn

a

I N -a a a amiddot

+

~ a

l

a w gt 0c 0 gt a I shyc -CJ wc w gtc(

== c( I shyZ w E W a 0 z- 8

ii sect

~ ~ bullT 0

I if

x ~

bullbullbull

II ~------------________

ANDREW PHASE II

WA VE FORCE PROFILEmiddot

CAP AUTOMATIC APPLICATION FEATURE

STAT

fR1 0 lt f lt1

STAT

STAT

f(R3-R2) middot f lt1

Project ST151K Model Wave_Loading Version 1 rue Nov 8 164107 1994 PT

E~ nss FT

SL 14amp FT

a 111 FT

EL 14 li

E-1 37 FT

El o FT

1middot1middot ______________ _ ________ __ _ __

0

200 400 600 800 1000 12000 Force at Elevatloii (kip)

200

a ll_

Wave Force at Major Elevations for Discrete Wave Heights

180

bullHmiddot40ft

Hmiddot495ft

H bull 51 ft

Hmiddot55ft_

H-59ft I

Andrew (H bull 60 86 ft) II

i I

_11---__

- middot- Hmiddot30ft

-

- bull -

middot bull bull bull middot

---Hmiddot53ft

bullbullbullbullHmiddot57ft

-----H-63ft

Elevation Above Mudllne 100

(ft)

180

140

120

80

60

40

20

Loading at Major Elevations

---H-49511

-deg-Hmiddot608611

0 200 400 600 800 1000 1200

Force It Elevatlon (kip)

--

Base Shear vs Wave Height ST151K

6000

5000 1---------Ult1mateCapacitybull 4890 kip (100 Improved Soll)-----------~--

4000a a lii ~ 3000

81 ampI

2000

1000

0 +-~~~--i~~~~~~~~~-+-~~~--lf--~~~~~~~~-+-~~~--l

30 35 40 45 50 55 60 65

Wave Height (ft)

bull middot

~

bull

II

2

g

~

It)

0

cCD E 0 E a c E euro~ CDshygtItgtmiddoto ~ti c -a Gic CD gt ~

-= c -Cll

I z

bullO 0i i ~ g 0

() CI -bull(q-ll) iuawow llu1wniiaAO

11

shy

116

114

112

g 110

bullc = 108

~ ~ 106 -sect 104

102

100

98

96

-

6555 60

Wave Height vs Center of Action ST151K

30 35 40 45 50

Wave Height (ft)

t

Comparison of Pushover Curves for ST 151K Variable Load Pattern and Monotonic Load Pattern (Andrew)

Soll Improved 100

--- --------- --- ----6000

5000

- 4000 8 - I

1 3000 Ul

81 m 2000

_J -shy

1000

---Variable Load

- - Andrew Load

0 f-~~~~~r--~~~~-t-~~~~~-t-~~~~~+-~~~~----t~~~~~-1

0 1 - 2 3 4 5 6

Deck Displacement (ft)

-middotshy

1 t I I ~

ANDREW PHASE II

BRACE MODELING

bull MARSHALL STRUTS USED FORPHASE I

bull PHASE II WILL ASSESS DIFFERENCE IN USING BUCKLING BEAMS AND MARSHALL STRUTS

middot bull COMPARE FOR 2D FRAME

bull COMPARE FOR PLATFORM 177 WHERE LOWER

BAY EXHIBITS LARGE BRACE MOMENTS

Project andrew Model beamframe Version 1 Mon Nov 14 145447 1994

~

middot Frame Test buckling beam-column model

-

Project andrew Model strutframe Version 1 Mon Nov 14 145754 1994

bull

Frame Test strut model

middot-middot _____________--_

_middotmiddotr_~ --ll--~-middot__

Buckling Beam bull

Strut (K=O 65)

Lateral Displacement

ANDREW PHASE II

FOUNDATIONS

bull ~LICIT NONLINEAR API FOUNDATION USED INiPHASE I

bull FOUNDATIONS

JIP SELECTED MINI-VANE DATA FOR

STRENGTH PROFILE

bull COMMENTS ON BENCHMARK COMPARISONS

- SIGNIFICANT VARIABILITY DUE TO FOUNDATION MODELING METHODS

II Ii

z

ctV 4x SPlO caissonfy-42s soil-api(lf-097step-114)

Inelastic Events Legend Elastic Strut Buckling

Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield

--------shyBeam Clmn Fully Plastic Fracture

~--~~~~~~~~~--~----~__i~~---~~~~__--~

Pro)ect apipile Model South Pelto 10 Version Wed May 18072235 1994 - -

1bullt

+

t

f i bullbullbull

bullbull bullbull

I I

5

~x SPlO Caissonfy=42s soil=api(lf=lOOstep=137)

Inelastic Events Legend

Elastic Strut Bucklilg

Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield

earn Clmn Full~ Plas1c Fracture

ANDREW PHASE II middot middot

JOINT MODELING

middotbull API RIGID-PLASTIC USED IN PHASE I

a ELASTIC-BRITTLE USED IN PHASE IB

amiddot MSL WILL PROVIDE NONLINEAR FORCE-DEFORMATION

AND MOMENT-DEFORMATION RELATIONS FOR PHASE II

a MODELING OPTIONS FOR PHASE II -

l

- UNCO~LEDmiddotONE DIMENSIONAL MEMBER END JOINT MODELING

J Ii

ANDREW PHASE Il

JOINT MODELING OPTIONS DISCUSSION

bull JD COUPLED INTERACTION SURFACE

- IDEAL

- DATA NOT AVAILABLE TO CALIBRATE

bull 2D COUPLED INTERACTION SURFACE

- SIGNIFICANT MODELING IMPROVEMENT

- DATAAVAILABLE

- IMPLEMENTATION PRACTICAL BUT NEEDS DEVELOPMENT AND CALIBRATION

bull 2D UNCOUPLED INTERACTION SURFACE

- SIGNIFICANT MODELING IMPROVEMENT

- DATA AVAILABLE

- IMPLEMENTATION PRACTICAL AND AVAILABLE

ANDREW PHASE II

JOINT MODEJJNG OPTIONS ILLUSTRATIONS

P

e

ANDREW PHASE II

JOINT MODELING PHASE II METHOD

e

I I I

ANDREW PHASE II

ANDREW STORM CONDmONS UPDATED

bull -PHASE I USED OCEANWEATHER HINDCAST-(Nov 1992)

bull middotmiddotPHASE II WILL USE THE REVISED HINDCAST (Nov 1994)

bull lt c lt I lt

2-2

l---4-~~~+-~~-+-~~~1--~~-+-~~--l--~T--+-~~--i24

f-tJVtMfiz~ 1Cfl (r STvb(

Plollol bullbull 5middotNOYmiddot9214bull1J Irbullbull Illbull [ANDREVl~NDREVCUSTbullMAXll 5-NOY-1992 10bull54 - 9208

HEIGHT lml

I~ MAXI

I L 1~middotamp SZ~- ~A I 13c

I J) ( r I X~0iWJ~0gtJ gt L gt gt I 1c I 120

-- N

~ deg

I I I I I I e C I ~=-41 122 -94 -92 -90 -88 -86 -84 -82 -80

-

Figure 7 contours of maximum hindcast significant wave height

_

~

~tl ~

NOltNttIJ 17lt ltNS S iJJ)y

Maximum Significant Wave Height - Hurricane Andrew 1992

30 ~

26 ~ I ~-~~~- == -shy -~ 35-_

24deg1--shy --------~- middot-middot-------middotmiddotmiddot----middot------middot-middot~middot- - - ~-----1---middot

22middot ~~~~~~~-~ -94 -92 -90 -88 -84deg-86 -82 -somiddot

c

-

Platform Selection bullbullbull1111111111111

0 BasisCriteria for Selection middot

0 Review of Phase I Platforms middot 0 Alternative Platforms

0 PMB Recommendations

0 Participants Comments

gt

L

l I

BasisCriteria for Selection bullbullbull1111111111111

0 Benefit for Calibration - Classification (UnexpectedExpected)

- Failure Mode (StructuresFoundations)

- Damaged StructuresKnown response

- Better Representation of Full ~opulation

0 Availability of Data - Existing Models

- Inspection Data

- Damage Data

PREDICTED BEHAVIOR

-

OBSERVED

BEHAVIOR

-SURVIVED DAMAGED FAILED

SURVIVED

ST134W WD90A

-MC311 -

MC397 -

--

-

-

ST151K ST130Q

-

-

-

DAMAGED - -

-

-

--

-

-

ST52 SS139 ST177B -

ST161A

-_

-

FAILED

-

-

-

ST151H ST130A ST72

Operator - Platfonns - Platfonns Survived Damagedamp

- Falled -

bull

-

-

Jle eBDIGIPllDbl -Amocoshy 30 1 Chevron -

143 12 Exxon 72 Mobll 22 1 Phllllps 7 bull

-Shell 39 Unocal 25 1

Total Partlclnts Platfonns 338 15

Other Operators bull- 310 13

-

Total Platfonns - -

648 28

fbml

Platforms Considered 6 7

tie of Total Platfonns 1 25

OVERVIEW OF PLATFORMS SURVIVEDDAMAGEDFAILED

-

shy

v

Platforms Evaluated in Phase I

-

-Platfonn Water Year Number Analysis Avallabillty of Availability Platform

- Name Depth rl of Model DetaDed rl ClassUlcatlon

Installation legs Inspection Soll Report Based on Information Phase I Calibration - -ft

- -- Suaill flalfnrm ~BSi -STISIK 137 1963 8 CAP Yes Yes Un eel Sarvlval

-STIJOO 170 1964 4 CAPmiddotshy Yes ST134data u ST134W 137 1981 4 CAP Yes Yesmiddot E

-

edSarvlval -

middotsurvival

WD90A 184 - 1964 8 CAP Yes E -

- Sarvlvill

MC3ll 343 1978 8 KARMA

MC397 468 1991 4 KARMA -

bull ~reglmmiddot- -

Yes Sure Survival

bull Yes - Sare Sanlval-

-T23STS2 63 1969 4 CAP

T2SSS139 62 1969 4 CAP -

Yes

Yes

Yes Ex

Yes Une

bull SurvivalLikely to DamalI

middot bull SanlvalLlkely to DamalI

ST161A 118 1964 8 USFOS

Eail11a pound1al(bullllm Clss ST177B 142 --1965 8 CAP

STISIH 137 1964 - 8 CAP C

STIJOA 140 19S8 8 CAP - -

Yes

Yes

ollamed

Yes

- Yes E

-

at about 1 mUe

Yes

ST134data

bull SafvlvalLlkelv to Dam- bull- -

E middotrahre

Emeried Failure -

Likely FaUare

T21ST72 61 1969 4 CAP Collamed No Likely FaUure

Platforms Identified for Evaluating Their Likelihood of Falling Under UnexpeCted Survival Category

CllEllATOa

AllB

llLOCll

Sl1lvcnJllE

NAME

YEAR

INSTAUJCD

WATER

IJBPJB

middotmiddotDISTANCE

ftOMSllOllE

(MIUIS)

Ill

(M)

AMOCO BllW

STllI

WDI075

A

bull D

4

n uo 17J

44

32

bull 20

10

bull CllEVllOH STtllO c I 1bull 21 11

STl130 D 0 161 21 11 STllJ4 I 0 137 D - 11

STIUI G I 137 32 11

STIUI I 0 121 32 11

STIUI

J 0 140 32 11

STIUI It 4 137 32 11

STllSI PROl)I 0 137 32 11

STllSI PROD-2 0 137 32 11

STll7 A 0

140 35 11

WDl117 c 65 214 34 10

DXON STshy E 51 14 bull STl165 A t3 bull 34 WD0073

WD0073

WD0073

A _ c

4

62

4

IA IA

17J

22

22

22

bullbullbull WD003-shy E 65 161 Z7

A 0 IN 31 MOBD Bl12I

ssn A bull

50

50

31

7

bull 7

MIJllPllY SS114 B 7 50 7 7

PlllUIPS SSl14t A 55 33 7

SAMEDAN ST17J A 4 117 7 bull 7

SJIELL WD0103 A 65 223 27 WDllOJ bull 65 221 27 WDll04 c 65 221 27 WDOI04 D 67 2 27

UNOCAL

SSl20I

SSl20I

bull SSl20I

SS0209

SS0209

SS0209

SS0209

SS 0253

SS 0253

E

I

PtJllE

A

bull D

G A

c

0

4

4

7J

61

0

0

69

103

97

97

9S

100

95

100

165

175

34

34

34

34

34

34

34

54

54

bullbullbullbullbullbullbullbullbull

Platforms Recommended for Phase II

llalr y_ w- Aot Av- _ I 1 Doplla Model nm bull bull _ bull I I H11 s18-1 Irht lbwl~ bullbullbull(-illtdlel OM lllMtD

II ---

-STISIIC 137 1963 I CAP Yeo Yeo I Yeo

STtshy -170 1964 4 CAP Yeo ST134dlU I Yeo

STl34W 137 Yeo Yeo -1981 4 CAP D Yeo

WD90A 114 1964 I CAP Yeo - D

MC311 30 1971 I KARMA Yeo -sre-

MC397 middot 461 1991 4 ICARMA Yeo Sireshy - - r- -

__ T13ST5l 63 1969 4 CAP Yeo Yeobull middot- ~n- D Yeo _ Tl5SSl39 62 1969 4 CAP Yeo Yeo I Yeo

_ STIT711 142 1965 I CAP Yeo I Yeo middot-1shySTl61A Ill 1964 I USFOS Yeo Yeo - D bulloshy

- r-

STISIH 137 1964 I CAP y D y -_ STl31A 140 1951 I CAP Yeo STl34dlU I Yeo

T21Sll 61 1969 4 CAP No rar-middot I Yeo

~-rmiddot-middot SS209A 95 1967 16 CAP Yeo D -SSU4R

11 UI A 41 -in D SS 2741AIM CAP bull

_ _ -shy - -shy- -shy

CALIBRATION

bull PHASEIAPPROACH

bull CALIBRATION ISSUES

bull PHASE II WORK SCOPE

bull SOFlWARE AVAILABILITY

bull EFFECT OF ADDITIONAL PLATFORMS

bull WEIGHTING APPROACH

bull EXAMPLE APPLICATION middot

(l=Ofce Mode] Wind

~

fcurrent l -middot

~SW ~leldl4

ultimate resistance force I first

design load Illgt

i _

deflection

-

bull

4

Historical Hurrlcane Dariiage and Survival of Platforms

Components Required for Calibration (Petrauskas 1992)

PRIOR Distribution

PLATFORM EXPERIENCE

BAYESIAN Survivals and Failures

UPDATING

Waves Winds and Currents bull

POSTERIOR Distribution

- A_middotJA ~ Survival Data Failure Data

Result Result

Load Resistance

middotfl

-

Bayesian Updaing middotGeneral Description (Petrauskas 1992)

L Imiddot

WHERE TO INTRODUCE BIAS FACTOR (B

On Resistance (Rmue - B (RPREDICreD I

middotOn Loads (Smue middot - B (SPRE01cTEo

On Safety Margin (RSmue - B (RSPREDICTED

CALIBRATION METHODOLOGY

gt

r

FIELD OBSERVATIONS

Observed Behavior

survivals Damages

Failures

r

bull

CAPACITY ANALYSIS

Predicted Load and Resistance Predicted Failure Modes

Survival Damage Failure cases

RELIABILITY ANALYSIS

- Probability of Failure ~

Ukellhood Failure

BAYESIAN UPDATING

Prior Distnbution of B Combined Likelihood Function

Posterior Dlstnbution of B

BIAS FACTOR

Mean Value cov bull

Represents Modeling Uncertainties

Prior Distribution of B Likelihood Functionmiddot Observing Success case STISIK and Posterior Distribution of B

2 00

uo

160

140

120

e 100 L $ uoIshy

0 60 ~ ~ 040

020

0 00

0 00 0 50

_ I 00

b

I 50 2 00

-shy lk [bl STISIK)

--shy Pnor-f(b)

--shy Pos1enor r(bl ISIK)

Prior or B Mean bull 10 COV bull 030

Posterlor of Bbull Meanbull 136COV 0 16

2 50

Likelihood Function and Posterior Distribution Example Platform STISIK

- -

~

- Prior and Posterior Distributions or B Failure cases (ST177B STlSlH ST130A T21) Damage cases (T23 T25 ST161A) Success cases (STlSlK ST130Q ST134W WD90A MC311 MC397)

400

350

300

2 so 0-

I

bull bull

bull bull gt

- bullbull ~ bull bull I bull

_ bull bull bullbull I bull bull bullII V -

bull bull

I bull ~ bull

bullbull bullI bull bullI bull bull bull

middot -~

bull__ ~ middotI middotmiddot

L 2 00

e ~ I SO

-I 00

- 0 so

000 gt

0 00 020 0 40 060 0 80 I 00 120 140 160 110 200

b -

-

shy

Prior-r

- - - - Posterior fLlll failure cases

bull bull bull bull bull bull bull Posterior flall damage cases

- bull bull - bull middot Postenor flall success cases

Posterior flall success fulure damage cases

-

Posterior of B Mean= 119 COV 0101

Posterior Distribution or Blas Factor (8) bull Different Categories and All Cases Combined

shy

l

PHASE I - KEY CONCLUSIONS

bull GLOBAL CORRECTION FACTOR

bull 15 TO 20 CONSERVATISM MEAN)

bull UNEXPECTED SURVIVALS HAVE GREATEST IMPACT

bull DAMAGED CASES MORE INFORMATIVE

bull CALIBRATION ISSUES PHASE II

bull

EFFECT OF IMPROVED CAPACITY ANALYSIS RECIPE

bull APPLICABILITY TO MULTIPLE FAILURE MODES

bull NO FOUNDATION DAMAGE CASE INCLUDED

bull IMPROVED REPRESENTATION OF FULL ANDREW DISTRIBUTION

-bull NUMBER OF PLATFORMS CONSIDERED

I 1 I

I t ) bull ~

PHASE II WORK SCOPE (CALIBRATION)

bull

bull ESTABLISH MULTIPLE BIAS FACTORS

Use Nine Base Case Platforms

middot - Capacity Analysis for Revised recipe

Foundation Blas Factor from Foundation JIP

bull DEVELOP WEIGHTING APPROACH

bull UPDATE MULTIPLE BIAS FACTORS

bullEffect of Platforms Not In the Nine Base Cases

bull EXAMPLE APPLICATION OF BIAS FACTORS I

bullbull

PHASE I PHASE II

CAPACITYbull ANALYSISRu bull

---------------------1----middot

Jc1 - iJbullraquoJ dbull

FttlS frt[f~l ratdhJJ

-

LIKELIHOOD FUNCTION

DEVELOPMENT

lk Jb Jfall1N) bull Jb)

111 Ill Isurvtvoq bull 1 - bl

-middot -------------------shy ~---------------------middot-middot

r111ir8 Cb1 111 Cbl BAYESIAN

UPDATING

~cove

R1R2 bull middot---------------------~----

Jblbull P(g cOJ

1 bull JbRI (MBSI I bull 1

~middot

llcJllJfalluro) bull P(gcOJ Ill 111 b21fallun) bull P ((g1 CO) U 1112 c OD

middot--------------------shy ----------------------shy bull

r111111r8 Cbl 111 Ill i-1nro1

deg ~1b2)laquottfrr (b1b21-lnfo) re Ill bull 1-11middot b2) bull lodegamp

middot

i1 covamp i2 cov~i

-

shy

--

Likelihood Function vs b All Success Case STISI K STl30Q STl34W WD90A MC311 MC397

I 0000

0 9000

OIOOO

0 7000

~ 0amp000

ta osoOO

~ 0 4000

0 3000

0 2000

0 1000

bull 0 0000

bull

I

middot

middot bull I

Ibull

-sibull ~ -middot-middot-middot-middot -middot-

----7----+-----~ I

Ibull bull bull I

rbull

--- Jk ISIK[blsuccess)

- bull - bull - lk 130Q[blsuccess) bull

lk 134W[blsuccess)

Jk-WD90A(blsuccc~I

- - - lk MC397[blsucccss)

--ltgt-- lk middotMC31 l[blsucccss)

--- Jk [bl 6 out of6succcsHascs)

05 I 5 2 250

b

Likelihood Functionsmiddot Success Cases - Phbullsc L

shy

bullbull bull bull bull bull bull -------------------~

bull bull bullbull bullbull bullbull bullbull bullbull bullbull bull

bull bullbull

bullbullbullbull

bullbull

bullbullbullbullbullbullbull

bullbullbullbullbullbull

bullbullbullbullbullbullbull

z 0

t z ) LL

c 0 0c i ~ i

J

~ sect ~--------------~ e l5

I

I

SOFTWARE

bull PF PROGRAM

bull CALREL

bull PF +CALREL

bull RELACS

bull

gt i

l

SENSITIVITY STUDIES

POSTERIOR DISTRIBUTION OF B

bull EFFECT OF CHANGE IN DISTRIBUTION OF PRIOR OF B

bull EFFECT OF REMOVAL OF ONE PLATFORM FROM EACH CATEGORY

bull EFFECT OF REMOVAL OF THREE DAMAGE CASES

bull EFFECT OF ADDITIONAL 13 SIMILAR PLATFORMS

-I

Prior and Posterior Distributions of B Variation In COV of PriorofB fromOl to04 -- -

4500

-4000

-

- -

3500 --middot - bull bull - bull bull Prior-r ICOV =02

3000

- - - - Prior-r ICOV =03 -

bull -~ e 2soo -

bull bull bull bull bull bull bull Pnor-r ICOV =o 4t frJ ~ 2000 Posterior rlCOV =0 2

bull bull Postenor rlCOV =0 3 bull bull 1500 bull middot bull - shy -middot- Posterior rlCOV =04

middot~ I 000 - bull

-middotmiddot ~ ~middotmiddot M -~

- middot 0 500

wbullbullbull _ I middot~ middotmiddotmiddotmiddotmiddot middotmiddotmiddotmiddotmiddot - -~ ~ __ middot_ _ bull -shy0 000

0 000 0200 0 400 0 600 0 800 IOOO 1200 1400 1600 1800 2000 -

- - b

-

=shy

Posterior Distribution of Bias Factor (B) bull Effect of Variation In COV of Prior of B

bull

bullPrior and Posterior Distributions of B Evaluating Effect of Removal of One Platform from Each Category from the Base Case of 13 Platforms

400

~ ~ 350

j IA i

3 00 bull I bull I ---Prior-r I bull l

250 1 I 1

I bull bull ---Posterior r lall 13 casesshy- ~E 2 00 bull bull bull bull bull bull bull Posterior r without STISIK

- ~ 1 bull - bull I - bull bull - bull bull Posterior r Iwithout T2I I 50 I

J-tt~i - bull - bull - Posterior r Iwithout T2S

tOO [-

I I ~ ~ I 0 50

0 00

___ r J i

l~T

~--

0 00 0 20 0 40 0 60 080 I 00 120 140 160 110 2 00

b

zshy

Posterior Distribution of Bias Factor (8) bull Effect of Removal of middotPlatform from Any Category

bullPrior and Posterior Distributions or B Evaluating Effect or Removal or 3 Damage Cases from the Base Case or 13 Platrorms

4

35

3

25

e L 2 ~ bulle shy

I 15

bull I

05

0

0 02 04 06 I 2 I 4 16 IS 2

I I

I

I

I Ibull

J -tmiddot L--

bullI

~

I

fl

II ~

I I v I

middot bull bull

bull L

~ ll- I-~lt

OS I

b

I

bull

1middot bullbullbullbull

~ __

bull

- - - - Prior-r

- - - - Postenor flall failure cases

Postenor flall success cases

--- Posterior flall 13 cases

---Posterior fI success failure cases only

IPosterior or B Mean =t11 COV =0121

-

_I

Posterir Distribution or Bias Factor (B) bull Effect or Removal or 3-Damage Cases

-

bullPrior and Posterior Distributions or B Evaluating Effect or Additional 13 Similar Observations as

Used in Andrew JIP The POsterior or B from Andrew-JIP Considered as New Prior or B

- 500

450

400

350

bull bull

bull bull bull bull J bull

bull bull f ~ I

I ~ Ibull bull

bullbull bull

l bull I -bullbull bull bull

I

I bull bull

bull I I~~bull bull bull

bull ~

I bull bull

~ ~_ ~~~

- - New Prior r bull

Posterior 114 new failure cases ~ 300- bull bull bull bull bull bull bull Posterior f 13 new damage casesl 150 ~

- bull bull - bull bull Posterior 116 new success cases i 2 00

Postenor 1113 new success failure -150 damage cases

I 00

0 50 -

0 00 0 00 0 20 0 40 0 60 0 80 I 00 120 I 40 160 110 2 00

Posterior ol B Mean= 120 COV= 0071 b

shy

shy

_rshy

Posterior Distribution of Bias Factor (B) bull Effect of Additional 13-Similar Observations

1

1 I

WEIGHTING APPROACH bull

bull ASSESSMENT OF PLATFORMS DATA

Select Platforms Most Mappable to the Analyzed Platforms

Unlikely SurvivalDamage Cases Identification

Based on Cut-Off Criteria Agemiddot Hs etc

bull INFORMATION REQUIRED FOR COMPARISON

bull CATEGORIZATION OF ADDITIONALmiddot PLATFORMS

bull UPDATING OF CALIBRATION RESULTS

--

OVERVIEW OF PLATFORMS SURVIVEDDAMAGED -~

Operator -

- Platforms

Survived CS)

Platforms

Damaged CD)

Platforms Used In

Phase IJIP -

JI~ ~Bill~IRBDll -

Amoco -

Chevron -

Exxon

Mob II

Phllllps

Shell

Unocal -

30

143

72 22

7

39 25shy -

1

12

bull

1 bull

1

1(S)+1 (D)

3 (S) +3(D)

2 (S)

bull shy

bull

bull

bull

Total Partlclnants Platforms 338 - 15 10C6S+4D) -

Other Operators

310

13

3 (D)

Total Platforms 648 28 13C6S+7D) -

Summary of Andrew JIP bull Phase I Platforms

Plairshy w v- N-ber Lea-le 1 shy car Hind_ o-w N-bull Deplh fl fl IJpe la bull n Yeadon Jadtet lbelledtf liefalit -- _ -OD

IL H - -shy _ICllshy IL IL - - rQClto ~ _

K STUIK 137 1963 8 365 608S4 Cellardeck

-middot STl-shy 170 1964 4 -~ K 39 62269 Cellardeck - --ST134W 137 1911 4 43 608S4 Smvlnlshy- _ - -

WD90A 184 1964 8 K 267Sbull34 83 SOS31 SUboftllar deck SmvMI

MC311 343 1978 8 x S7 62 - MC397 468 1991 4 x shy S33bull6SO 65911 -shySmeSmvlvol -- gt -

r-- T23ST52 63 1969 4 ton- shyKshy 52594 S03 -T25SSl39 62 1969 4 Kshy SOS94 S063 - middot ton-shy__ -

bullbull a ton--shyST161A 118 1964 8 K - 347S 51495 Cellarclecti

r~

ST1771 142 1965 8 K 39 60167 CeDordeck failme

STISIH 137 1964 8 K 395 608S4 CeDardeck Failme

K ST130A 140 1958 a SS 6095 I lblv Pai1uno - T21ST72 61 1969 4 Kshy S4S94 4974 I lhlv Pai1uno -

- -

JE

INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON

bull WATER DEPTH

bull bull YEAR OF INSTALLATIONDESIGN

bull PLATFORM ORIENTATION

bull PLATFORM LATITUDE AND LONGITUDE

bull KEY SKETCHES OF PLATFORM STRUCTURAL FRAMING

Number of Legs Plies Leg Batter Number of Conductors bull Deck Elevations (Including sub-cellar deck If any) Jacket Framing Elevations Brace type In Vertical Frames Sizes of Primary Members (leg pile vertical frames braces primary horizontals) Joint Details (can gap etc) LegPiie Annulus Grouting Status

bull middotPOST ANDREW INSPECTION REPORT Significant Wave Height Hs (ft)

bull INFORMATION ON DAMAGES PRIOR TO ANDREW

bull RESULTS FROM IN-HOUSE STUDIES IF AVAILABLE

bull ~RIGINAL DESIGN CRITERIA INFORMATION IF AVAILABLE

bull SOIL SHEAR STRENGTH PROFILE IF AVAILABLE

Platforms Identified for Evaluating Therr Likelihood of Falling Under Unexpected Survival Category

OPlRATOR

middotADA

BLOClt

llmucnJllE

NAME

YEAll

INSTAUJD

WA1ER

DBPTll

DISTANCE

JlllOM SHOllE

(MILU)

Bl

(M)

AMOCO 111113

STl11

WDI075

A

bull D

St

17

120

1n

44

20

10 I

CllllVllQfl STlllO c 61 110 21 11 STIUO D Q 1110 21 11

STl134 p Q 137 29 11

ST0151 G 61 137 11

ST0151 I 3 121 11

ST0151 J Q 140 11

STl151 It 137 11

ST0151 PJl001 Q UT 11

ST0151 PROD-2 Q UT 11

ST017 A 3 140 35 11

WDIUT c 65 214 34 10

llXXON STI055 II 51 6T 14 I

STl15 A t3 34

WDI073

WD0073

Amiddotshy Q

1111 1111

22

22

I

I

WDI073 c 1n 22 I

- WDOCl93-shy II

65 1110 2T A 3 194 31 MOBU 111121

SS IOl7

A B

501

50

50

31

T

I

T 6

MtlllPllY SS114 B T 50 T T

PlllllIPS SS0149 A 5 55 33 T

SAMllDAN STln A 107 T T

SBllLL WDl103 A 65 223 2T WDl103 B 65 221 2T WD0104 c 65 221 2T WD0104 D 6T 20 2T

UNOCAL SS020I

SS 020I SSl20I

SS 0209

SS0209

II p

PURll

A

B

3

n 61

103

97

97

95

100

34

34

34

34 34

bullbull bullbullbull

- SS0209 D u 95 34 I

- SS 1209 SSl253

SS0253

G A

c

6T

2

100

165

1T5

34

54

54

bullbull I

I I

I I

SOUTH MARCH

~LAND

N

SOUTH PaTOI ENE ISWltD AREA

MMS Defined Limits ror Offshore Structures Requiring Inspections Following Andrew

t---Jbulln3 T25bull I bullTtl

bull1$1Htl1Abull bull1SIK1779bull

~

middot-DaTA AREA~T

CRANO rsLE

AREA bull1300

bullll4W

=shy

bullPath of Hurricane Andrew with Platforms Used In Calibration

Likelihood Function vs b All Success Case STISI K STIJOQ ST134W WD90A MCJll MCJ97

I 0000

0 9000

0 8000

07000

-Bo6000

tiiii 0 5000 -

04000

0 3000

0 2000

01000

0 0000

0 05 25

ri1rt middot-middot-middot-middot-middot-

I 5 2

b

---11 middot151K(blsu~ccsI

- bull - bull - fk 130Q(bfU~lCSS)

lie bull134W(blsuccc]

lk-WD90A[blsultCS]

- - - lk middotMC397(bfUtlCS]

--o-- lk-MCJll(bbuctcsJ

--- ll (bl 6 out of 6 Mlttcs case]

_

( Likelihood Functions middotSuccess Cases - p - 1 c I

CATEGORIZATION OF ADDITIONAL PLATFORMS

I I I

bull CATEGORIES

Unexpected Survivals

Expected Survivals

bull APPROACH

Identify Corresponding Platform(s) Evaluated In the JIP

Determine Max Wave Height at Platform Location

Evaluate Whether Wave Hit the Deck

middot - Evaluate Likelihood of R and S to Differ from the Base Cases

Review Joint Information

Assess Special Features

bull IDENTIFY NUMBER OF PLATFORMS IN EACH CATEGORY

I

UPDATING OF CALIBRATION RESULTS

bull INCLUDE EFFECT OF ADDITIONAL PLATFORMS ON bullBbull

Consider Increased 1Uncertalntles In R and S

Develop bullNeW Ukellhood Functions for Different Categories

middotLet bullxbull Number of Ukely Unexpected Survival Cases and y Number of Ukely Expected Survival Cases

Determine Combined Likelihood Function lk (b Iadditional platforms) middot

bull REVISED LIKELIHOOD FUNCTION

lk(b I all cases) = lk(b I 9 platforms) middot x lk(b Ix platforms) x lk(b I y platforms)

bull ESTABLISH REVISED BIAS FACTOR B

f8 (b) u f8 (b) lk (b I all cases)

II

t t ~

shy

APPLICATIONS OF BIAS FACTORS

ii APPLICABLE TO PARTICULAR LOAD AND RESISTANCE RECIPE

bull USE BY THE API COMMITTEES AND MMS

Criteria Revisions

Recommended Design Guidelines Revisions bull

bullbull REASSESSMENT STUDIES AND DECISION MAKING

middot

bull RISK ASSESSMENT STUDIES BY THE OPERATORS

Individual Platforms

Fleet of Platforms

middot IMR Program Implementation

bullbull NOT TO BE USED IN DESIGN

I I fl J

EXAMPLE USE OF BIAS FACTORS

bull PHASE I EXAMPLE APPLICATION

bull EVALUATION OF ANNUAL PROBABILITY OF FAILURE

Establish Annual Maximum Seastate Data

Determine Coefficients to Define Base Shear

Determine Ultimate Capacity of Platform

Determine Probablllty of Failure Versus bullbbull

-bull DETERMINE UPDATED PROBABILITY OF FAILURE

Pf - P [S gt R]

Pf I P [~ gt Rb I b] f 8 (b) db

Pr - J P [S gt Rb I b] f bull 8 (b) db

bull TOTAL PROBABILITIES OF FAILURE

Total Pf - 1 bull Il [1 middotPf] - l Pf bull

Total Pf - 1 bull Il [1middot Pf] - l Pfmiddot

Total Pr = 1 - n [1middot PfbullJ = l Pfshy

I I I

EXAMPLE APPLICATION bull PHASE I

PLATFORM ST 151 K

bull GULF OF MEXICO

bull WATER DEPTH = 137 FT

bull ANNUAL MAXIMUM HS MEDIAN 175 ft COV 031

bull ASSUMED STORM DURATION 3 hoursmiddot

bull ASSUMED NUMBER OF WAVES 800

bull MEDIAN HS FOR STORM HOURS 1 2 AND 3

bull MEDIAN CURRENT VALUE bull

bull DETERMINE ULTIMATE CAPACITY FOR THREE DIRECTIONS

bull DETERMINE ANNUAL FAILURE PROBABILITIES

Broadsidemiddot Diagonal End-on

0037 0025 0007

0071 0044 0018

0022 0016 0004

DETERMINE TOTAL ANNUAL PROBABILITY OF FAILURE

Pf - 007

Pf - 013

Pr - 004

bull

EXAMPLE USE OF BIAS FACTORS - PHASE II middot

bull DETERMINE PREDICTIVE PROBABILITY OF FAILURE

Case - 1

Pf _ S S Pf (b b2) fa1a2 (b b2) db db2

g bull -1 (b1 R I S) - 1

92 - (b2 R2 I S) - 1

Case -2

Pf s ~ [bJ f11 (b) db

Pf [bJ p [g lt 0)

g - bull 1 (b1 R1 I S) - 1

1

bull1) l

I

BENEFITS OF ANDREW JIP bull PHASE II

lt

bull IMPROVED LOAD AND RESISTANCE RECIPE

bull IMPROVED CALIBRATION PROCEDURE

bull MULTIPLE BIAS FACTORS

bull PROCEDURE TO INCLUDE EFFECT OF OTHER PLATFORMS

bull PROCEDURE TO USE MULTIPLE BIAS FACTORS

  • Structure Bookmarks
    • ANDREW PHASE II JOINT1 INDUSTRY PROJECT
    • PROGRESS MEETING NUMBER 1
    • ANDREW PHASE II JOINT INDUSTRY PROJECT PROGRESS MEETING NUMBER 1 NOVEMBER 16 AGENDA
    • ANDREW PHASE II UST OF PARTICIPANTS
    • ANDREW PHASE I OBJECTIVES AND RESULTS
    • ANDREW PHASE II PROJECT OBJECTIVES
    • SUBSEQUENT PROJECTS
    • ANDREW PHASE II PROJECT TEAM
    • ANDREW PHASE II UPDATE MODELING middotAND ANALYSIS RECIPE
    • -ANDREW PHASE II CONDUCTOR RESISTANCE
    • ANDREW PHASE Il WA VE FORCE PROFILE
    • ANDREW PHASE II WA VE FORCE PROFILEmiddot
    • ANDREW PHASE II BRACE MODELING
    • ANDREW PHASE II FOUNDATIONS
    • ANDREW PHASE II middot middot JOINT MODELING
    • ANDREW PHASE Il JOINT MODELING OPTIONS DISCUSSION
    • ANDREW PHASE II JOINT MODEJJNG OPTIONS ILLUSTRATIONS
    • ANDREW PHASE II JOINT MODELING PHASE II METHOD
    • ANDREW PHASE II ANDREW STORM CONDmONS UPDATED
    • Platform Selection bullbullbull1111111111111
    • BasisCriteria for Selection bullbullbull1111111111111
    • CALIBRATION
    • WHERE TO INTRODUCE BIAS FACTOR (B
    • PHASE I -KEY CONCLUSIONS
    • bull CALIBRATION ISSUES PHASE II
    • PHASE II WORK SCOPE (CALIBRATION)
    • SOFTWARE
    • SENSITIVITY STUDIES POSTERIOR DISTRIBUTION OF B
    • WEIGHTING APPROACH bull
    • INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON
    • CATEGORIZATION OF ADDITIONAL PLATFORMS
    • UPDATING OF CALIBRATION RESULTS
    • APPLICATIONS OF BIAS FACTORS
    • EXAMPLE USE OF BIAS FACTORS
    • EXAMPLE APPLICATION bull PHASE I PLATFORM ST 151 K
    • EXAMPLE USE OF BIAS FACTORS -PHASE II middot
    • BENEFITS OF ANDREW JIP bull PHASE II
Page 18:  · . ' ANDREW PHASE II PROJECT OBJECTIVES . ' • ·UPDATE MODELING AND ANALYSIS RECIPE· JOINTS ' ' ·BRACES WAVE PROFILE FOUNDATIONS . ' ' CONDUCTORS· • ANALVZE PLATFORMS USING

bullbull r 1 II

~

ANDREW PHASE Il

WA VE FORCE PROFILE

bull DECK WA VE FORCE ~

PHASE I USED SECTION 17 (1056H degCd)

PHASE Il USE REVISED SECTION 17 (1000H Cd)

bull PHASE I USED SINGLE FORCE PROFILE FOR PUSHOVER

- USED WAVE HEIGHT CLOSE TO ULTIMATE CAPACITY

bull PHASE J

Il WILL STUDY THE

USE OF VARIABLE

PROFILE ~

- IF FORCE LEVEL INCREASES (OR DECREASES) BY A FACTOR EQUIVALENT TO A 2 FOOT CHANGE IN WA VE HEIGHT THE INCREMENTAL FO~CE PA1IERN IS CHANGED ACCORDINGLY (CAP FEATURE)

- INITIAL EXAMPLE

f~

a

middotn

a

I N -a a a amiddot

+

~ a

l

a w gt 0c 0 gt a I shyc -CJ wc w gtc(

== c( I shyZ w E W a 0 z- 8

ii sect

~ ~ bullT 0

I if

x ~

bullbullbull

II ~------------________

ANDREW PHASE II

WA VE FORCE PROFILEmiddot

CAP AUTOMATIC APPLICATION FEATURE

STAT

fR1 0 lt f lt1

STAT

STAT

f(R3-R2) middot f lt1

Project ST151K Model Wave_Loading Version 1 rue Nov 8 164107 1994 PT

E~ nss FT

SL 14amp FT

a 111 FT

EL 14 li

E-1 37 FT

El o FT

1middot1middot ______________ _ ________ __ _ __

0

200 400 600 800 1000 12000 Force at Elevatloii (kip)

200

a ll_

Wave Force at Major Elevations for Discrete Wave Heights

180

bullHmiddot40ft

Hmiddot495ft

H bull 51 ft

Hmiddot55ft_

H-59ft I

Andrew (H bull 60 86 ft) II

i I

_11---__

- middot- Hmiddot30ft

-

- bull -

middot bull bull bull middot

---Hmiddot53ft

bullbullbullbullHmiddot57ft

-----H-63ft

Elevation Above Mudllne 100

(ft)

180

140

120

80

60

40

20

Loading at Major Elevations

---H-49511

-deg-Hmiddot608611

0 200 400 600 800 1000 1200

Force It Elevatlon (kip)

--

Base Shear vs Wave Height ST151K

6000

5000 1---------Ult1mateCapacitybull 4890 kip (100 Improved Soll)-----------~--

4000a a lii ~ 3000

81 ampI

2000

1000

0 +-~~~--i~~~~~~~~~-+-~~~--lf--~~~~~~~~-+-~~~--l

30 35 40 45 50 55 60 65

Wave Height (ft)

bull middot

~

bull

II

2

g

~

It)

0

cCD E 0 E a c E euro~ CDshygtItgtmiddoto ~ti c -a Gic CD gt ~

-= c -Cll

I z

bullO 0i i ~ g 0

() CI -bull(q-ll) iuawow llu1wniiaAO

11

shy

116

114

112

g 110

bullc = 108

~ ~ 106 -sect 104

102

100

98

96

-

6555 60

Wave Height vs Center of Action ST151K

30 35 40 45 50

Wave Height (ft)

t

Comparison of Pushover Curves for ST 151K Variable Load Pattern and Monotonic Load Pattern (Andrew)

Soll Improved 100

--- --------- --- ----6000

5000

- 4000 8 - I

1 3000 Ul

81 m 2000

_J -shy

1000

---Variable Load

- - Andrew Load

0 f-~~~~~r--~~~~-t-~~~~~-t-~~~~~+-~~~~----t~~~~~-1

0 1 - 2 3 4 5 6

Deck Displacement (ft)

-middotshy

1 t I I ~

ANDREW PHASE II

BRACE MODELING

bull MARSHALL STRUTS USED FORPHASE I

bull PHASE II WILL ASSESS DIFFERENCE IN USING BUCKLING BEAMS AND MARSHALL STRUTS

middot bull COMPARE FOR 2D FRAME

bull COMPARE FOR PLATFORM 177 WHERE LOWER

BAY EXHIBITS LARGE BRACE MOMENTS

Project andrew Model beamframe Version 1 Mon Nov 14 145447 1994

~

middot Frame Test buckling beam-column model

-

Project andrew Model strutframe Version 1 Mon Nov 14 145754 1994

bull

Frame Test strut model

middot-middot _____________--_

_middotmiddotr_~ --ll--~-middot__

Buckling Beam bull

Strut (K=O 65)

Lateral Displacement

ANDREW PHASE II

FOUNDATIONS

bull ~LICIT NONLINEAR API FOUNDATION USED INiPHASE I

bull FOUNDATIONS

JIP SELECTED MINI-VANE DATA FOR

STRENGTH PROFILE

bull COMMENTS ON BENCHMARK COMPARISONS

- SIGNIFICANT VARIABILITY DUE TO FOUNDATION MODELING METHODS

II Ii

z

ctV 4x SPlO caissonfy-42s soil-api(lf-097step-114)

Inelastic Events Legend Elastic Strut Buckling

Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield

--------shyBeam Clmn Fully Plastic Fracture

~--~~~~~~~~~--~----~__i~~---~~~~__--~

Pro)ect apipile Model South Pelto 10 Version Wed May 18072235 1994 - -

1bullt

+

t

f i bullbullbull

bullbull bullbull

I I

5

~x SPlO Caissonfy=42s soil=api(lf=lOOstep=137)

Inelastic Events Legend

Elastic Strut Bucklilg

Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield

earn Clmn Full~ Plas1c Fracture

ANDREW PHASE II middot middot

JOINT MODELING

middotbull API RIGID-PLASTIC USED IN PHASE I

a ELASTIC-BRITTLE USED IN PHASE IB

amiddot MSL WILL PROVIDE NONLINEAR FORCE-DEFORMATION

AND MOMENT-DEFORMATION RELATIONS FOR PHASE II

a MODELING OPTIONS FOR PHASE II -

l

- UNCO~LEDmiddotONE DIMENSIONAL MEMBER END JOINT MODELING

J Ii

ANDREW PHASE Il

JOINT MODELING OPTIONS DISCUSSION

bull JD COUPLED INTERACTION SURFACE

- IDEAL

- DATA NOT AVAILABLE TO CALIBRATE

bull 2D COUPLED INTERACTION SURFACE

- SIGNIFICANT MODELING IMPROVEMENT

- DATAAVAILABLE

- IMPLEMENTATION PRACTICAL BUT NEEDS DEVELOPMENT AND CALIBRATION

bull 2D UNCOUPLED INTERACTION SURFACE

- SIGNIFICANT MODELING IMPROVEMENT

- DATA AVAILABLE

- IMPLEMENTATION PRACTICAL AND AVAILABLE

ANDREW PHASE II

JOINT MODEJJNG OPTIONS ILLUSTRATIONS

P

e

ANDREW PHASE II

JOINT MODELING PHASE II METHOD

e

I I I

ANDREW PHASE II

ANDREW STORM CONDmONS UPDATED

bull -PHASE I USED OCEANWEATHER HINDCAST-(Nov 1992)

bull middotmiddotPHASE II WILL USE THE REVISED HINDCAST (Nov 1994)

bull lt c lt I lt

2-2

l---4-~~~+-~~-+-~~~1--~~-+-~~--l--~T--+-~~--i24

f-tJVtMfiz~ 1Cfl (r STvb(

Plollol bullbull 5middotNOYmiddot9214bull1J Irbullbull Illbull [ANDREVl~NDREVCUSTbullMAXll 5-NOY-1992 10bull54 - 9208

HEIGHT lml

I~ MAXI

I L 1~middotamp SZ~- ~A I 13c

I J) ( r I X~0iWJ~0gtJ gt L gt gt I 1c I 120

-- N

~ deg

I I I I I I e C I ~=-41 122 -94 -92 -90 -88 -86 -84 -82 -80

-

Figure 7 contours of maximum hindcast significant wave height

_

~

~tl ~

NOltNttIJ 17lt ltNS S iJJ)y

Maximum Significant Wave Height - Hurricane Andrew 1992

30 ~

26 ~ I ~-~~~- == -shy -~ 35-_

24deg1--shy --------~- middot-middot-------middotmiddotmiddot----middot------middot-middot~middot- - - ~-----1---middot

22middot ~~~~~~~-~ -94 -92 -90 -88 -84deg-86 -82 -somiddot

c

-

Platform Selection bullbullbull1111111111111

0 BasisCriteria for Selection middot

0 Review of Phase I Platforms middot 0 Alternative Platforms

0 PMB Recommendations

0 Participants Comments

gt

L

l I

BasisCriteria for Selection bullbullbull1111111111111

0 Benefit for Calibration - Classification (UnexpectedExpected)

- Failure Mode (StructuresFoundations)

- Damaged StructuresKnown response

- Better Representation of Full ~opulation

0 Availability of Data - Existing Models

- Inspection Data

- Damage Data

PREDICTED BEHAVIOR

-

OBSERVED

BEHAVIOR

-SURVIVED DAMAGED FAILED

SURVIVED

ST134W WD90A

-MC311 -

MC397 -

--

-

-

ST151K ST130Q

-

-

-

DAMAGED - -

-

-

--

-

-

ST52 SS139 ST177B -

ST161A

-_

-

FAILED

-

-

-

ST151H ST130A ST72

Operator - Platfonns - Platfonns Survived Damagedamp

- Falled -

bull

-

-

Jle eBDIGIPllDbl -Amocoshy 30 1 Chevron -

143 12 Exxon 72 Mobll 22 1 Phllllps 7 bull

-Shell 39 Unocal 25 1

Total Partlclnts Platfonns 338 15

Other Operators bull- 310 13

-

Total Platfonns - -

648 28

fbml

Platforms Considered 6 7

tie of Total Platfonns 1 25

OVERVIEW OF PLATFORMS SURVIVEDDAMAGEDFAILED

-

shy

v

Platforms Evaluated in Phase I

-

-Platfonn Water Year Number Analysis Avallabillty of Availability Platform

- Name Depth rl of Model DetaDed rl ClassUlcatlon

Installation legs Inspection Soll Report Based on Information Phase I Calibration - -ft

- -- Suaill flalfnrm ~BSi -STISIK 137 1963 8 CAP Yes Yes Un eel Sarvlval

-STIJOO 170 1964 4 CAPmiddotshy Yes ST134data u ST134W 137 1981 4 CAP Yes Yesmiddot E

-

edSarvlval -

middotsurvival

WD90A 184 - 1964 8 CAP Yes E -

- Sarvlvill

MC3ll 343 1978 8 KARMA

MC397 468 1991 4 KARMA -

bull ~reglmmiddot- -

Yes Sure Survival

bull Yes - Sare Sanlval-

-T23STS2 63 1969 4 CAP

T2SSS139 62 1969 4 CAP -

Yes

Yes

Yes Ex

Yes Une

bull SurvivalLikely to DamalI

middot bull SanlvalLlkely to DamalI

ST161A 118 1964 8 USFOS

Eail11a pound1al(bullllm Clss ST177B 142 --1965 8 CAP

STISIH 137 1964 - 8 CAP C

STIJOA 140 19S8 8 CAP - -

Yes

Yes

ollamed

Yes

- Yes E

-

at about 1 mUe

Yes

ST134data

bull SafvlvalLlkelv to Dam- bull- -

E middotrahre

Emeried Failure -

Likely FaUare

T21ST72 61 1969 4 CAP Collamed No Likely FaUure

Platforms Identified for Evaluating Their Likelihood of Falling Under UnexpeCted Survival Category

CllEllATOa

AllB

llLOCll

Sl1lvcnJllE

NAME

YEAR

INSTAUJCD

WATER

IJBPJB

middotmiddotDISTANCE

ftOMSllOllE

(MIUIS)

Ill

(M)

AMOCO BllW

STllI

WDI075

A

bull D

4

n uo 17J

44

32

bull 20

10

bull CllEVllOH STtllO c I 1bull 21 11

STl130 D 0 161 21 11 STllJ4 I 0 137 D - 11

STIUI G I 137 32 11

STIUI I 0 121 32 11

STIUI

J 0 140 32 11

STIUI It 4 137 32 11

STllSI PROl)I 0 137 32 11

STllSI PROD-2 0 137 32 11

STll7 A 0

140 35 11

WDl117 c 65 214 34 10

DXON STshy E 51 14 bull STl165 A t3 bull 34 WD0073

WD0073

WD0073

A _ c

4

62

4

IA IA

17J

22

22

22

bullbullbull WD003-shy E 65 161 Z7

A 0 IN 31 MOBD Bl12I

ssn A bull

50

50

31

7

bull 7

MIJllPllY SS114 B 7 50 7 7

PlllUIPS SSl14t A 55 33 7

SAMEDAN ST17J A 4 117 7 bull 7

SJIELL WD0103 A 65 223 27 WDllOJ bull 65 221 27 WDll04 c 65 221 27 WDOI04 D 67 2 27

UNOCAL

SSl20I

SSl20I

bull SSl20I

SS0209

SS0209

SS0209

SS0209

SS 0253

SS 0253

E

I

PtJllE

A

bull D

G A

c

0

4

4

7J

61

0

0

69

103

97

97

9S

100

95

100

165

175

34

34

34

34

34

34

34

54

54

bullbullbullbullbullbullbullbullbull

Platforms Recommended for Phase II

llalr y_ w- Aot Av- _ I 1 Doplla Model nm bull bull _ bull I I H11 s18-1 Irht lbwl~ bullbullbull(-illtdlel OM lllMtD

II ---

-STISIIC 137 1963 I CAP Yeo Yeo I Yeo

STtshy -170 1964 4 CAP Yeo ST134dlU I Yeo

STl34W 137 Yeo Yeo -1981 4 CAP D Yeo

WD90A 114 1964 I CAP Yeo - D

MC311 30 1971 I KARMA Yeo -sre-

MC397 middot 461 1991 4 ICARMA Yeo Sireshy - - r- -

__ T13ST5l 63 1969 4 CAP Yeo Yeobull middot- ~n- D Yeo _ Tl5SSl39 62 1969 4 CAP Yeo Yeo I Yeo

_ STIT711 142 1965 I CAP Yeo I Yeo middot-1shySTl61A Ill 1964 I USFOS Yeo Yeo - D bulloshy

- r-

STISIH 137 1964 I CAP y D y -_ STl31A 140 1951 I CAP Yeo STl34dlU I Yeo

T21Sll 61 1969 4 CAP No rar-middot I Yeo

~-rmiddot-middot SS209A 95 1967 16 CAP Yeo D -SSU4R

11 UI A 41 -in D SS 2741AIM CAP bull

_ _ -shy - -shy- -shy

CALIBRATION

bull PHASEIAPPROACH

bull CALIBRATION ISSUES

bull PHASE II WORK SCOPE

bull SOFlWARE AVAILABILITY

bull EFFECT OF ADDITIONAL PLATFORMS

bull WEIGHTING APPROACH

bull EXAMPLE APPLICATION middot

(l=Ofce Mode] Wind

~

fcurrent l -middot

~SW ~leldl4

ultimate resistance force I first

design load Illgt

i _

deflection

-

bull

4

Historical Hurrlcane Dariiage and Survival of Platforms

Components Required for Calibration (Petrauskas 1992)

PRIOR Distribution

PLATFORM EXPERIENCE

BAYESIAN Survivals and Failures

UPDATING

Waves Winds and Currents bull

POSTERIOR Distribution

- A_middotJA ~ Survival Data Failure Data

Result Result

Load Resistance

middotfl

-

Bayesian Updaing middotGeneral Description (Petrauskas 1992)

L Imiddot

WHERE TO INTRODUCE BIAS FACTOR (B

On Resistance (Rmue - B (RPREDICreD I

middotOn Loads (Smue middot - B (SPRE01cTEo

On Safety Margin (RSmue - B (RSPREDICTED

CALIBRATION METHODOLOGY

gt

r

FIELD OBSERVATIONS

Observed Behavior

survivals Damages

Failures

r

bull

CAPACITY ANALYSIS

Predicted Load and Resistance Predicted Failure Modes

Survival Damage Failure cases

RELIABILITY ANALYSIS

- Probability of Failure ~

Ukellhood Failure

BAYESIAN UPDATING

Prior Distnbution of B Combined Likelihood Function

Posterior Dlstnbution of B

BIAS FACTOR

Mean Value cov bull

Represents Modeling Uncertainties

Prior Distribution of B Likelihood Functionmiddot Observing Success case STISIK and Posterior Distribution of B

2 00

uo

160

140

120

e 100 L $ uoIshy

0 60 ~ ~ 040

020

0 00

0 00 0 50

_ I 00

b

I 50 2 00

-shy lk [bl STISIK)

--shy Pnor-f(b)

--shy Pos1enor r(bl ISIK)

Prior or B Mean bull 10 COV bull 030

Posterlor of Bbull Meanbull 136COV 0 16

2 50

Likelihood Function and Posterior Distribution Example Platform STISIK

- -

~

- Prior and Posterior Distributions or B Failure cases (ST177B STlSlH ST130A T21) Damage cases (T23 T25 ST161A) Success cases (STlSlK ST130Q ST134W WD90A MC311 MC397)

400

350

300

2 so 0-

I

bull bull

bull bull gt

- bullbull ~ bull bull I bull

_ bull bull bullbull I bull bull bullII V -

bull bull

I bull ~ bull

bullbull bullI bull bullI bull bull bull

middot -~

bull__ ~ middotI middotmiddot

L 2 00

e ~ I SO

-I 00

- 0 so

000 gt

0 00 020 0 40 060 0 80 I 00 120 140 160 110 200

b -

-

shy

Prior-r

- - - - Posterior fLlll failure cases

bull bull bull bull bull bull bull Posterior flall damage cases

- bull bull - bull middot Postenor flall success cases

Posterior flall success fulure damage cases

-

Posterior of B Mean= 119 COV 0101

Posterior Distribution or Blas Factor (8) bull Different Categories and All Cases Combined

shy

l

PHASE I - KEY CONCLUSIONS

bull GLOBAL CORRECTION FACTOR

bull 15 TO 20 CONSERVATISM MEAN)

bull UNEXPECTED SURVIVALS HAVE GREATEST IMPACT

bull DAMAGED CASES MORE INFORMATIVE

bull CALIBRATION ISSUES PHASE II

bull

EFFECT OF IMPROVED CAPACITY ANALYSIS RECIPE

bull APPLICABILITY TO MULTIPLE FAILURE MODES

bull NO FOUNDATION DAMAGE CASE INCLUDED

bull IMPROVED REPRESENTATION OF FULL ANDREW DISTRIBUTION

-bull NUMBER OF PLATFORMS CONSIDERED

I 1 I

I t ) bull ~

PHASE II WORK SCOPE (CALIBRATION)

bull

bull ESTABLISH MULTIPLE BIAS FACTORS

Use Nine Base Case Platforms

middot - Capacity Analysis for Revised recipe

Foundation Blas Factor from Foundation JIP

bull DEVELOP WEIGHTING APPROACH

bull UPDATE MULTIPLE BIAS FACTORS

bullEffect of Platforms Not In the Nine Base Cases

bull EXAMPLE APPLICATION OF BIAS FACTORS I

bullbull

PHASE I PHASE II

CAPACITYbull ANALYSISRu bull

---------------------1----middot

Jc1 - iJbullraquoJ dbull

FttlS frt[f~l ratdhJJ

-

LIKELIHOOD FUNCTION

DEVELOPMENT

lk Jb Jfall1N) bull Jb)

111 Ill Isurvtvoq bull 1 - bl

-middot -------------------shy ~---------------------middot-middot

r111ir8 Cb1 111 Cbl BAYESIAN

UPDATING

~cove

R1R2 bull middot---------------------~----

Jblbull P(g cOJ

1 bull JbRI (MBSI I bull 1

~middot

llcJllJfalluro) bull P(gcOJ Ill 111 b21fallun) bull P ((g1 CO) U 1112 c OD

middot--------------------shy ----------------------shy bull

r111111r8 Cbl 111 Ill i-1nro1

deg ~1b2)laquottfrr (b1b21-lnfo) re Ill bull 1-11middot b2) bull lodegamp

middot

i1 covamp i2 cov~i

-

shy

--

Likelihood Function vs b All Success Case STISI K STl30Q STl34W WD90A MC311 MC397

I 0000

0 9000

OIOOO

0 7000

~ 0amp000

ta osoOO

~ 0 4000

0 3000

0 2000

0 1000

bull 0 0000

bull

I

middot

middot bull I

Ibull

-sibull ~ -middot-middot-middot-middot -middot-

----7----+-----~ I

Ibull bull bull I

rbull

--- Jk ISIK[blsuccess)

- bull - bull - lk 130Q[blsuccess) bull

lk 134W[blsuccess)

Jk-WD90A(blsuccc~I

- - - lk MC397[blsucccss)

--ltgt-- lk middotMC31 l[blsucccss)

--- Jk [bl 6 out of6succcsHascs)

05 I 5 2 250

b

Likelihood Functionsmiddot Success Cases - Phbullsc L

shy

bullbull bull bull bull bull bull -------------------~

bull bull bullbull bullbull bullbull bullbull bullbull bullbull bull

bull bullbull

bullbullbullbull

bullbull

bullbullbullbullbullbullbull

bullbullbullbullbullbull

bullbullbullbullbullbullbull

z 0

t z ) LL

c 0 0c i ~ i

J

~ sect ~--------------~ e l5

I

I

SOFTWARE

bull PF PROGRAM

bull CALREL

bull PF +CALREL

bull RELACS

bull

gt i

l

SENSITIVITY STUDIES

POSTERIOR DISTRIBUTION OF B

bull EFFECT OF CHANGE IN DISTRIBUTION OF PRIOR OF B

bull EFFECT OF REMOVAL OF ONE PLATFORM FROM EACH CATEGORY

bull EFFECT OF REMOVAL OF THREE DAMAGE CASES

bull EFFECT OF ADDITIONAL 13 SIMILAR PLATFORMS

-I

Prior and Posterior Distributions of B Variation In COV of PriorofB fromOl to04 -- -

4500

-4000

-

- -

3500 --middot - bull bull - bull bull Prior-r ICOV =02

3000

- - - - Prior-r ICOV =03 -

bull -~ e 2soo -

bull bull bull bull bull bull bull Pnor-r ICOV =o 4t frJ ~ 2000 Posterior rlCOV =0 2

bull bull Postenor rlCOV =0 3 bull bull 1500 bull middot bull - shy -middot- Posterior rlCOV =04

middot~ I 000 - bull

-middotmiddot ~ ~middotmiddot M -~

- middot 0 500

wbullbullbull _ I middot~ middotmiddotmiddotmiddotmiddot middotmiddotmiddotmiddotmiddot - -~ ~ __ middot_ _ bull -shy0 000

0 000 0200 0 400 0 600 0 800 IOOO 1200 1400 1600 1800 2000 -

- - b

-

=shy

Posterior Distribution of Bias Factor (B) bull Effect of Variation In COV of Prior of B

bull

bullPrior and Posterior Distributions of B Evaluating Effect of Removal of One Platform from Each Category from the Base Case of 13 Platforms

400

~ ~ 350

j IA i

3 00 bull I bull I ---Prior-r I bull l

250 1 I 1

I bull bull ---Posterior r lall 13 casesshy- ~E 2 00 bull bull bull bull bull bull bull Posterior r without STISIK

- ~ 1 bull - bull I - bull bull - bull bull Posterior r Iwithout T2I I 50 I

J-tt~i - bull - bull - Posterior r Iwithout T2S

tOO [-

I I ~ ~ I 0 50

0 00

___ r J i

l~T

~--

0 00 0 20 0 40 0 60 080 I 00 120 140 160 110 2 00

b

zshy

Posterior Distribution of Bias Factor (8) bull Effect of Removal of middotPlatform from Any Category

bullPrior and Posterior Distributions or B Evaluating Effect or Removal or 3 Damage Cases from the Base Case or 13 Platrorms

4

35

3

25

e L 2 ~ bulle shy

I 15

bull I

05

0

0 02 04 06 I 2 I 4 16 IS 2

I I

I

I

I Ibull

J -tmiddot L--

bullI

~

I

fl

II ~

I I v I

middot bull bull

bull L

~ ll- I-~lt

OS I

b

I

bull

1middot bullbullbullbull

~ __

bull

- - - - Prior-r

- - - - Postenor flall failure cases

Postenor flall success cases

--- Posterior flall 13 cases

---Posterior fI success failure cases only

IPosterior or B Mean =t11 COV =0121

-

_I

Posterir Distribution or Bias Factor (B) bull Effect or Removal or 3-Damage Cases

-

bullPrior and Posterior Distributions or B Evaluating Effect or Additional 13 Similar Observations as

Used in Andrew JIP The POsterior or B from Andrew-JIP Considered as New Prior or B

- 500

450

400

350

bull bull

bull bull bull bull J bull

bull bull f ~ I

I ~ Ibull bull

bullbull bull

l bull I -bullbull bull bull

I

I bull bull

bull I I~~bull bull bull

bull ~

I bull bull

~ ~_ ~~~

- - New Prior r bull

Posterior 114 new failure cases ~ 300- bull bull bull bull bull bull bull Posterior f 13 new damage casesl 150 ~

- bull bull - bull bull Posterior 116 new success cases i 2 00

Postenor 1113 new success failure -150 damage cases

I 00

0 50 -

0 00 0 00 0 20 0 40 0 60 0 80 I 00 120 I 40 160 110 2 00

Posterior ol B Mean= 120 COV= 0071 b

shy

shy

_rshy

Posterior Distribution of Bias Factor (B) bull Effect of Additional 13-Similar Observations

1

1 I

WEIGHTING APPROACH bull

bull ASSESSMENT OF PLATFORMS DATA

Select Platforms Most Mappable to the Analyzed Platforms

Unlikely SurvivalDamage Cases Identification

Based on Cut-Off Criteria Agemiddot Hs etc

bull INFORMATION REQUIRED FOR COMPARISON

bull CATEGORIZATION OF ADDITIONALmiddot PLATFORMS

bull UPDATING OF CALIBRATION RESULTS

--

OVERVIEW OF PLATFORMS SURVIVEDDAMAGED -~

Operator -

- Platforms

Survived CS)

Platforms

Damaged CD)

Platforms Used In

Phase IJIP -

JI~ ~Bill~IRBDll -

Amoco -

Chevron -

Exxon

Mob II

Phllllps

Shell

Unocal -

30

143

72 22

7

39 25shy -

1

12

bull

1 bull

1

1(S)+1 (D)

3 (S) +3(D)

2 (S)

bull shy

bull

bull

bull

Total Partlclnants Platforms 338 - 15 10C6S+4D) -

Other Operators

310

13

3 (D)

Total Platforms 648 28 13C6S+7D) -

Summary of Andrew JIP bull Phase I Platforms

Plairshy w v- N-ber Lea-le 1 shy car Hind_ o-w N-bull Deplh fl fl IJpe la bull n Yeadon Jadtet lbelledtf liefalit -- _ -OD

IL H - -shy _ICllshy IL IL - - rQClto ~ _

K STUIK 137 1963 8 365 608S4 Cellardeck

-middot STl-shy 170 1964 4 -~ K 39 62269 Cellardeck - --ST134W 137 1911 4 43 608S4 Smvlnlshy- _ - -

WD90A 184 1964 8 K 267Sbull34 83 SOS31 SUboftllar deck SmvMI

MC311 343 1978 8 x S7 62 - MC397 468 1991 4 x shy S33bull6SO 65911 -shySmeSmvlvol -- gt -

r-- T23ST52 63 1969 4 ton- shyKshy 52594 S03 -T25SSl39 62 1969 4 Kshy SOS94 S063 - middot ton-shy__ -

bullbull a ton--shyST161A 118 1964 8 K - 347S 51495 Cellarclecti

r~

ST1771 142 1965 8 K 39 60167 CeDordeck failme

STISIH 137 1964 8 K 395 608S4 CeDardeck Failme

K ST130A 140 1958 a SS 6095 I lblv Pai1uno - T21ST72 61 1969 4 Kshy S4S94 4974 I lhlv Pai1uno -

- -

JE

INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON

bull WATER DEPTH

bull bull YEAR OF INSTALLATIONDESIGN

bull PLATFORM ORIENTATION

bull PLATFORM LATITUDE AND LONGITUDE

bull KEY SKETCHES OF PLATFORM STRUCTURAL FRAMING

Number of Legs Plies Leg Batter Number of Conductors bull Deck Elevations (Including sub-cellar deck If any) Jacket Framing Elevations Brace type In Vertical Frames Sizes of Primary Members (leg pile vertical frames braces primary horizontals) Joint Details (can gap etc) LegPiie Annulus Grouting Status

bull middotPOST ANDREW INSPECTION REPORT Significant Wave Height Hs (ft)

bull INFORMATION ON DAMAGES PRIOR TO ANDREW

bull RESULTS FROM IN-HOUSE STUDIES IF AVAILABLE

bull ~RIGINAL DESIGN CRITERIA INFORMATION IF AVAILABLE

bull SOIL SHEAR STRENGTH PROFILE IF AVAILABLE

Platforms Identified for Evaluating Therr Likelihood of Falling Under Unexpected Survival Category

OPlRATOR

middotADA

BLOClt

llmucnJllE

NAME

YEAll

INSTAUJD

WA1ER

DBPTll

DISTANCE

JlllOM SHOllE

(MILU)

Bl

(M)

AMOCO 111113

STl11

WDI075

A

bull D

St

17

120

1n

44

20

10 I

CllllVllQfl STlllO c 61 110 21 11 STIUO D Q 1110 21 11

STl134 p Q 137 29 11

ST0151 G 61 137 11

ST0151 I 3 121 11

ST0151 J Q 140 11

STl151 It 137 11

ST0151 PJl001 Q UT 11

ST0151 PROD-2 Q UT 11

ST017 A 3 140 35 11

WDIUT c 65 214 34 10

llXXON STI055 II 51 6T 14 I

STl15 A t3 34

WDI073

WD0073

Amiddotshy Q

1111 1111

22

22

I

I

WDI073 c 1n 22 I

- WDOCl93-shy II

65 1110 2T A 3 194 31 MOBU 111121

SS IOl7

A B

501

50

50

31

T

I

T 6

MtlllPllY SS114 B T 50 T T

PlllllIPS SS0149 A 5 55 33 T

SAMllDAN STln A 107 T T

SBllLL WDl103 A 65 223 2T WDl103 B 65 221 2T WD0104 c 65 221 2T WD0104 D 6T 20 2T

UNOCAL SS020I

SS 020I SSl20I

SS 0209

SS0209

II p

PURll

A

B

3

n 61

103

97

97

95

100

34

34

34

34 34

bullbull bullbullbull

- SS0209 D u 95 34 I

- SS 1209 SSl253

SS0253

G A

c

6T

2

100

165

1T5

34

54

54

bullbull I

I I

I I

SOUTH MARCH

~LAND

N

SOUTH PaTOI ENE ISWltD AREA

MMS Defined Limits ror Offshore Structures Requiring Inspections Following Andrew

t---Jbulln3 T25bull I bullTtl

bull1$1Htl1Abull bull1SIK1779bull

~

middot-DaTA AREA~T

CRANO rsLE

AREA bull1300

bullll4W

=shy

bullPath of Hurricane Andrew with Platforms Used In Calibration

Likelihood Function vs b All Success Case STISI K STIJOQ ST134W WD90A MCJll MCJ97

I 0000

0 9000

0 8000

07000

-Bo6000

tiiii 0 5000 -

04000

0 3000

0 2000

01000

0 0000

0 05 25

ri1rt middot-middot-middot-middot-middot-

I 5 2

b

---11 middot151K(blsu~ccsI

- bull - bull - fk 130Q(bfU~lCSS)

lie bull134W(blsuccc]

lk-WD90A[blsultCS]

- - - lk middotMC397(bfUtlCS]

--o-- lk-MCJll(bbuctcsJ

--- ll (bl 6 out of 6 Mlttcs case]

_

( Likelihood Functions middotSuccess Cases - p - 1 c I

CATEGORIZATION OF ADDITIONAL PLATFORMS

I I I

bull CATEGORIES

Unexpected Survivals

Expected Survivals

bull APPROACH

Identify Corresponding Platform(s) Evaluated In the JIP

Determine Max Wave Height at Platform Location

Evaluate Whether Wave Hit the Deck

middot - Evaluate Likelihood of R and S to Differ from the Base Cases

Review Joint Information

Assess Special Features

bull IDENTIFY NUMBER OF PLATFORMS IN EACH CATEGORY

I

UPDATING OF CALIBRATION RESULTS

bull INCLUDE EFFECT OF ADDITIONAL PLATFORMS ON bullBbull

Consider Increased 1Uncertalntles In R and S

Develop bullNeW Ukellhood Functions for Different Categories

middotLet bullxbull Number of Ukely Unexpected Survival Cases and y Number of Ukely Expected Survival Cases

Determine Combined Likelihood Function lk (b Iadditional platforms) middot

bull REVISED LIKELIHOOD FUNCTION

lk(b I all cases) = lk(b I 9 platforms) middot x lk(b Ix platforms) x lk(b I y platforms)

bull ESTABLISH REVISED BIAS FACTOR B

f8 (b) u f8 (b) lk (b I all cases)

II

t t ~

shy

APPLICATIONS OF BIAS FACTORS

ii APPLICABLE TO PARTICULAR LOAD AND RESISTANCE RECIPE

bull USE BY THE API COMMITTEES AND MMS

Criteria Revisions

Recommended Design Guidelines Revisions bull

bullbull REASSESSMENT STUDIES AND DECISION MAKING

middot

bull RISK ASSESSMENT STUDIES BY THE OPERATORS

Individual Platforms

Fleet of Platforms

middot IMR Program Implementation

bullbull NOT TO BE USED IN DESIGN

I I fl J

EXAMPLE USE OF BIAS FACTORS

bull PHASE I EXAMPLE APPLICATION

bull EVALUATION OF ANNUAL PROBABILITY OF FAILURE

Establish Annual Maximum Seastate Data

Determine Coefficients to Define Base Shear

Determine Ultimate Capacity of Platform

Determine Probablllty of Failure Versus bullbbull

-bull DETERMINE UPDATED PROBABILITY OF FAILURE

Pf - P [S gt R]

Pf I P [~ gt Rb I b] f 8 (b) db

Pr - J P [S gt Rb I b] f bull 8 (b) db

bull TOTAL PROBABILITIES OF FAILURE

Total Pf - 1 bull Il [1 middotPf] - l Pf bull

Total Pf - 1 bull Il [1middot Pf] - l Pfmiddot

Total Pr = 1 - n [1middot PfbullJ = l Pfshy

I I I

EXAMPLE APPLICATION bull PHASE I

PLATFORM ST 151 K

bull GULF OF MEXICO

bull WATER DEPTH = 137 FT

bull ANNUAL MAXIMUM HS MEDIAN 175 ft COV 031

bull ASSUMED STORM DURATION 3 hoursmiddot

bull ASSUMED NUMBER OF WAVES 800

bull MEDIAN HS FOR STORM HOURS 1 2 AND 3

bull MEDIAN CURRENT VALUE bull

bull DETERMINE ULTIMATE CAPACITY FOR THREE DIRECTIONS

bull DETERMINE ANNUAL FAILURE PROBABILITIES

Broadsidemiddot Diagonal End-on

0037 0025 0007

0071 0044 0018

0022 0016 0004

DETERMINE TOTAL ANNUAL PROBABILITY OF FAILURE

Pf - 007

Pf - 013

Pr - 004

bull

EXAMPLE USE OF BIAS FACTORS - PHASE II middot

bull DETERMINE PREDICTIVE PROBABILITY OF FAILURE

Case - 1

Pf _ S S Pf (b b2) fa1a2 (b b2) db db2

g bull -1 (b1 R I S) - 1

92 - (b2 R2 I S) - 1

Case -2

Pf s ~ [bJ f11 (b) db

Pf [bJ p [g lt 0)

g - bull 1 (b1 R1 I S) - 1

1

bull1) l

I

BENEFITS OF ANDREW JIP bull PHASE II

lt

bull IMPROVED LOAD AND RESISTANCE RECIPE

bull IMPROVED CALIBRATION PROCEDURE

bull MULTIPLE BIAS FACTORS

bull PROCEDURE TO INCLUDE EFFECT OF OTHER PLATFORMS

bull PROCEDURE TO USE MULTIPLE BIAS FACTORS

  • Structure Bookmarks
    • ANDREW PHASE II JOINT1 INDUSTRY PROJECT
    • PROGRESS MEETING NUMBER 1
    • ANDREW PHASE II JOINT INDUSTRY PROJECT PROGRESS MEETING NUMBER 1 NOVEMBER 16 AGENDA
    • ANDREW PHASE II UST OF PARTICIPANTS
    • ANDREW PHASE I OBJECTIVES AND RESULTS
    • ANDREW PHASE II PROJECT OBJECTIVES
    • SUBSEQUENT PROJECTS
    • ANDREW PHASE II PROJECT TEAM
    • ANDREW PHASE II UPDATE MODELING middotAND ANALYSIS RECIPE
    • -ANDREW PHASE II CONDUCTOR RESISTANCE
    • ANDREW PHASE Il WA VE FORCE PROFILE
    • ANDREW PHASE II WA VE FORCE PROFILEmiddot
    • ANDREW PHASE II BRACE MODELING
    • ANDREW PHASE II FOUNDATIONS
    • ANDREW PHASE II middot middot JOINT MODELING
    • ANDREW PHASE Il JOINT MODELING OPTIONS DISCUSSION
    • ANDREW PHASE II JOINT MODEJJNG OPTIONS ILLUSTRATIONS
    • ANDREW PHASE II JOINT MODELING PHASE II METHOD
    • ANDREW PHASE II ANDREW STORM CONDmONS UPDATED
    • Platform Selection bullbullbull1111111111111
    • BasisCriteria for Selection bullbullbull1111111111111
    • CALIBRATION
    • WHERE TO INTRODUCE BIAS FACTOR (B
    • PHASE I -KEY CONCLUSIONS
    • bull CALIBRATION ISSUES PHASE II
    • PHASE II WORK SCOPE (CALIBRATION)
    • SOFTWARE
    • SENSITIVITY STUDIES POSTERIOR DISTRIBUTION OF B
    • WEIGHTING APPROACH bull
    • INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON
    • CATEGORIZATION OF ADDITIONAL PLATFORMS
    • UPDATING OF CALIBRATION RESULTS
    • APPLICATIONS OF BIAS FACTORS
    • EXAMPLE USE OF BIAS FACTORS
    • EXAMPLE APPLICATION bull PHASE I PLATFORM ST 151 K
    • EXAMPLE USE OF BIAS FACTORS -PHASE II middot
    • BENEFITS OF ANDREW JIP bull PHASE II
Page 19:  · . ' ANDREW PHASE II PROJECT OBJECTIVES . ' • ·UPDATE MODELING AND ANALYSIS RECIPE· JOINTS ' ' ·BRACES WAVE PROFILE FOUNDATIONS . ' ' CONDUCTORS· • ANALVZE PLATFORMS USING

f~

a

middotn

a

I N -a a a amiddot

+

~ a

l

a w gt 0c 0 gt a I shyc -CJ wc w gtc(

== c( I shyZ w E W a 0 z- 8

ii sect

~ ~ bullT 0

I if

x ~

bullbullbull

II ~------------________

ANDREW PHASE II

WA VE FORCE PROFILEmiddot

CAP AUTOMATIC APPLICATION FEATURE

STAT

fR1 0 lt f lt1

STAT

STAT

f(R3-R2) middot f lt1

Project ST151K Model Wave_Loading Version 1 rue Nov 8 164107 1994 PT

E~ nss FT

SL 14amp FT

a 111 FT

EL 14 li

E-1 37 FT

El o FT

1middot1middot ______________ _ ________ __ _ __

0

200 400 600 800 1000 12000 Force at Elevatloii (kip)

200

a ll_

Wave Force at Major Elevations for Discrete Wave Heights

180

bullHmiddot40ft

Hmiddot495ft

H bull 51 ft

Hmiddot55ft_

H-59ft I

Andrew (H bull 60 86 ft) II

i I

_11---__

- middot- Hmiddot30ft

-

- bull -

middot bull bull bull middot

---Hmiddot53ft

bullbullbullbullHmiddot57ft

-----H-63ft

Elevation Above Mudllne 100

(ft)

180

140

120

80

60

40

20

Loading at Major Elevations

---H-49511

-deg-Hmiddot608611

0 200 400 600 800 1000 1200

Force It Elevatlon (kip)

--

Base Shear vs Wave Height ST151K

6000

5000 1---------Ult1mateCapacitybull 4890 kip (100 Improved Soll)-----------~--

4000a a lii ~ 3000

81 ampI

2000

1000

0 +-~~~--i~~~~~~~~~-+-~~~--lf--~~~~~~~~-+-~~~--l

30 35 40 45 50 55 60 65

Wave Height (ft)

bull middot

~

bull

II

2

g

~

It)

0

cCD E 0 E a c E euro~ CDshygtItgtmiddoto ~ti c -a Gic CD gt ~

-= c -Cll

I z

bullO 0i i ~ g 0

() CI -bull(q-ll) iuawow llu1wniiaAO

11

shy

116

114

112

g 110

bullc = 108

~ ~ 106 -sect 104

102

100

98

96

-

6555 60

Wave Height vs Center of Action ST151K

30 35 40 45 50

Wave Height (ft)

t

Comparison of Pushover Curves for ST 151K Variable Load Pattern and Monotonic Load Pattern (Andrew)

Soll Improved 100

--- --------- --- ----6000

5000

- 4000 8 - I

1 3000 Ul

81 m 2000

_J -shy

1000

---Variable Load

- - Andrew Load

0 f-~~~~~r--~~~~-t-~~~~~-t-~~~~~+-~~~~----t~~~~~-1

0 1 - 2 3 4 5 6

Deck Displacement (ft)

-middotshy

1 t I I ~

ANDREW PHASE II

BRACE MODELING

bull MARSHALL STRUTS USED FORPHASE I

bull PHASE II WILL ASSESS DIFFERENCE IN USING BUCKLING BEAMS AND MARSHALL STRUTS

middot bull COMPARE FOR 2D FRAME

bull COMPARE FOR PLATFORM 177 WHERE LOWER

BAY EXHIBITS LARGE BRACE MOMENTS

Project andrew Model beamframe Version 1 Mon Nov 14 145447 1994

~

middot Frame Test buckling beam-column model

-

Project andrew Model strutframe Version 1 Mon Nov 14 145754 1994

bull

Frame Test strut model

middot-middot _____________--_

_middotmiddotr_~ --ll--~-middot__

Buckling Beam bull

Strut (K=O 65)

Lateral Displacement

ANDREW PHASE II

FOUNDATIONS

bull ~LICIT NONLINEAR API FOUNDATION USED INiPHASE I

bull FOUNDATIONS

JIP SELECTED MINI-VANE DATA FOR

STRENGTH PROFILE

bull COMMENTS ON BENCHMARK COMPARISONS

- SIGNIFICANT VARIABILITY DUE TO FOUNDATION MODELING METHODS

II Ii

z

ctV 4x SPlO caissonfy-42s soil-api(lf-097step-114)

Inelastic Events Legend Elastic Strut Buckling

Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield

--------shyBeam Clmn Fully Plastic Fracture

~--~~~~~~~~~--~----~__i~~---~~~~__--~

Pro)ect apipile Model South Pelto 10 Version Wed May 18072235 1994 - -

1bullt

+

t

f i bullbullbull

bullbull bullbull

I I

5

~x SPlO Caissonfy=42s soil=api(lf=lOOstep=137)

Inelastic Events Legend

Elastic Strut Bucklilg

Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield

earn Clmn Full~ Plas1c Fracture

ANDREW PHASE II middot middot

JOINT MODELING

middotbull API RIGID-PLASTIC USED IN PHASE I

a ELASTIC-BRITTLE USED IN PHASE IB

amiddot MSL WILL PROVIDE NONLINEAR FORCE-DEFORMATION

AND MOMENT-DEFORMATION RELATIONS FOR PHASE II

a MODELING OPTIONS FOR PHASE II -

l

- UNCO~LEDmiddotONE DIMENSIONAL MEMBER END JOINT MODELING

J Ii

ANDREW PHASE Il

JOINT MODELING OPTIONS DISCUSSION

bull JD COUPLED INTERACTION SURFACE

- IDEAL

- DATA NOT AVAILABLE TO CALIBRATE

bull 2D COUPLED INTERACTION SURFACE

- SIGNIFICANT MODELING IMPROVEMENT

- DATAAVAILABLE

- IMPLEMENTATION PRACTICAL BUT NEEDS DEVELOPMENT AND CALIBRATION

bull 2D UNCOUPLED INTERACTION SURFACE

- SIGNIFICANT MODELING IMPROVEMENT

- DATA AVAILABLE

- IMPLEMENTATION PRACTICAL AND AVAILABLE

ANDREW PHASE II

JOINT MODEJJNG OPTIONS ILLUSTRATIONS

P

e

ANDREW PHASE II

JOINT MODELING PHASE II METHOD

e

I I I

ANDREW PHASE II

ANDREW STORM CONDmONS UPDATED

bull -PHASE I USED OCEANWEATHER HINDCAST-(Nov 1992)

bull middotmiddotPHASE II WILL USE THE REVISED HINDCAST (Nov 1994)

bull lt c lt I lt

2-2

l---4-~~~+-~~-+-~~~1--~~-+-~~--l--~T--+-~~--i24

f-tJVtMfiz~ 1Cfl (r STvb(

Plollol bullbull 5middotNOYmiddot9214bull1J Irbullbull Illbull [ANDREVl~NDREVCUSTbullMAXll 5-NOY-1992 10bull54 - 9208

HEIGHT lml

I~ MAXI

I L 1~middotamp SZ~- ~A I 13c

I J) ( r I X~0iWJ~0gtJ gt L gt gt I 1c I 120

-- N

~ deg

I I I I I I e C I ~=-41 122 -94 -92 -90 -88 -86 -84 -82 -80

-

Figure 7 contours of maximum hindcast significant wave height

_

~

~tl ~

NOltNttIJ 17lt ltNS S iJJ)y

Maximum Significant Wave Height - Hurricane Andrew 1992

30 ~

26 ~ I ~-~~~- == -shy -~ 35-_

24deg1--shy --------~- middot-middot-------middotmiddotmiddot----middot------middot-middot~middot- - - ~-----1---middot

22middot ~~~~~~~-~ -94 -92 -90 -88 -84deg-86 -82 -somiddot

c

-

Platform Selection bullbullbull1111111111111

0 BasisCriteria for Selection middot

0 Review of Phase I Platforms middot 0 Alternative Platforms

0 PMB Recommendations

0 Participants Comments

gt

L

l I

BasisCriteria for Selection bullbullbull1111111111111

0 Benefit for Calibration - Classification (UnexpectedExpected)

- Failure Mode (StructuresFoundations)

- Damaged StructuresKnown response

- Better Representation of Full ~opulation

0 Availability of Data - Existing Models

- Inspection Data

- Damage Data

PREDICTED BEHAVIOR

-

OBSERVED

BEHAVIOR

-SURVIVED DAMAGED FAILED

SURVIVED

ST134W WD90A

-MC311 -

MC397 -

--

-

-

ST151K ST130Q

-

-

-

DAMAGED - -

-

-

--

-

-

ST52 SS139 ST177B -

ST161A

-_

-

FAILED

-

-

-

ST151H ST130A ST72

Operator - Platfonns - Platfonns Survived Damagedamp

- Falled -

bull

-

-

Jle eBDIGIPllDbl -Amocoshy 30 1 Chevron -

143 12 Exxon 72 Mobll 22 1 Phllllps 7 bull

-Shell 39 Unocal 25 1

Total Partlclnts Platfonns 338 15

Other Operators bull- 310 13

-

Total Platfonns - -

648 28

fbml

Platforms Considered 6 7

tie of Total Platfonns 1 25

OVERVIEW OF PLATFORMS SURVIVEDDAMAGEDFAILED

-

shy

v

Platforms Evaluated in Phase I

-

-Platfonn Water Year Number Analysis Avallabillty of Availability Platform

- Name Depth rl of Model DetaDed rl ClassUlcatlon

Installation legs Inspection Soll Report Based on Information Phase I Calibration - -ft

- -- Suaill flalfnrm ~BSi -STISIK 137 1963 8 CAP Yes Yes Un eel Sarvlval

-STIJOO 170 1964 4 CAPmiddotshy Yes ST134data u ST134W 137 1981 4 CAP Yes Yesmiddot E

-

edSarvlval -

middotsurvival

WD90A 184 - 1964 8 CAP Yes E -

- Sarvlvill

MC3ll 343 1978 8 KARMA

MC397 468 1991 4 KARMA -

bull ~reglmmiddot- -

Yes Sure Survival

bull Yes - Sare Sanlval-

-T23STS2 63 1969 4 CAP

T2SSS139 62 1969 4 CAP -

Yes

Yes

Yes Ex

Yes Une

bull SurvivalLikely to DamalI

middot bull SanlvalLlkely to DamalI

ST161A 118 1964 8 USFOS

Eail11a pound1al(bullllm Clss ST177B 142 --1965 8 CAP

STISIH 137 1964 - 8 CAP C

STIJOA 140 19S8 8 CAP - -

Yes

Yes

ollamed

Yes

- Yes E

-

at about 1 mUe

Yes

ST134data

bull SafvlvalLlkelv to Dam- bull- -

E middotrahre

Emeried Failure -

Likely FaUare

T21ST72 61 1969 4 CAP Collamed No Likely FaUure

Platforms Identified for Evaluating Their Likelihood of Falling Under UnexpeCted Survival Category

CllEllATOa

AllB

llLOCll

Sl1lvcnJllE

NAME

YEAR

INSTAUJCD

WATER

IJBPJB

middotmiddotDISTANCE

ftOMSllOllE

(MIUIS)

Ill

(M)

AMOCO BllW

STllI

WDI075

A

bull D

4

n uo 17J

44

32

bull 20

10

bull CllEVllOH STtllO c I 1bull 21 11

STl130 D 0 161 21 11 STllJ4 I 0 137 D - 11

STIUI G I 137 32 11

STIUI I 0 121 32 11

STIUI

J 0 140 32 11

STIUI It 4 137 32 11

STllSI PROl)I 0 137 32 11

STllSI PROD-2 0 137 32 11

STll7 A 0

140 35 11

WDl117 c 65 214 34 10

DXON STshy E 51 14 bull STl165 A t3 bull 34 WD0073

WD0073

WD0073

A _ c

4

62

4

IA IA

17J

22

22

22

bullbullbull WD003-shy E 65 161 Z7

A 0 IN 31 MOBD Bl12I

ssn A bull

50

50

31

7

bull 7

MIJllPllY SS114 B 7 50 7 7

PlllUIPS SSl14t A 55 33 7

SAMEDAN ST17J A 4 117 7 bull 7

SJIELL WD0103 A 65 223 27 WDllOJ bull 65 221 27 WDll04 c 65 221 27 WDOI04 D 67 2 27

UNOCAL

SSl20I

SSl20I

bull SSl20I

SS0209

SS0209

SS0209

SS0209

SS 0253

SS 0253

E

I

PtJllE

A

bull D

G A

c

0

4

4

7J

61

0

0

69

103

97

97

9S

100

95

100

165

175

34

34

34

34

34

34

34

54

54

bullbullbullbullbullbullbullbullbull

Platforms Recommended for Phase II

llalr y_ w- Aot Av- _ I 1 Doplla Model nm bull bull _ bull I I H11 s18-1 Irht lbwl~ bullbullbull(-illtdlel OM lllMtD

II ---

-STISIIC 137 1963 I CAP Yeo Yeo I Yeo

STtshy -170 1964 4 CAP Yeo ST134dlU I Yeo

STl34W 137 Yeo Yeo -1981 4 CAP D Yeo

WD90A 114 1964 I CAP Yeo - D

MC311 30 1971 I KARMA Yeo -sre-

MC397 middot 461 1991 4 ICARMA Yeo Sireshy - - r- -

__ T13ST5l 63 1969 4 CAP Yeo Yeobull middot- ~n- D Yeo _ Tl5SSl39 62 1969 4 CAP Yeo Yeo I Yeo

_ STIT711 142 1965 I CAP Yeo I Yeo middot-1shySTl61A Ill 1964 I USFOS Yeo Yeo - D bulloshy

- r-

STISIH 137 1964 I CAP y D y -_ STl31A 140 1951 I CAP Yeo STl34dlU I Yeo

T21Sll 61 1969 4 CAP No rar-middot I Yeo

~-rmiddot-middot SS209A 95 1967 16 CAP Yeo D -SSU4R

11 UI A 41 -in D SS 2741AIM CAP bull

_ _ -shy - -shy- -shy

CALIBRATION

bull PHASEIAPPROACH

bull CALIBRATION ISSUES

bull PHASE II WORK SCOPE

bull SOFlWARE AVAILABILITY

bull EFFECT OF ADDITIONAL PLATFORMS

bull WEIGHTING APPROACH

bull EXAMPLE APPLICATION middot

(l=Ofce Mode] Wind

~

fcurrent l -middot

~SW ~leldl4

ultimate resistance force I first

design load Illgt

i _

deflection

-

bull

4

Historical Hurrlcane Dariiage and Survival of Platforms

Components Required for Calibration (Petrauskas 1992)

PRIOR Distribution

PLATFORM EXPERIENCE

BAYESIAN Survivals and Failures

UPDATING

Waves Winds and Currents bull

POSTERIOR Distribution

- A_middotJA ~ Survival Data Failure Data

Result Result

Load Resistance

middotfl

-

Bayesian Updaing middotGeneral Description (Petrauskas 1992)

L Imiddot

WHERE TO INTRODUCE BIAS FACTOR (B

On Resistance (Rmue - B (RPREDICreD I

middotOn Loads (Smue middot - B (SPRE01cTEo

On Safety Margin (RSmue - B (RSPREDICTED

CALIBRATION METHODOLOGY

gt

r

FIELD OBSERVATIONS

Observed Behavior

survivals Damages

Failures

r

bull

CAPACITY ANALYSIS

Predicted Load and Resistance Predicted Failure Modes

Survival Damage Failure cases

RELIABILITY ANALYSIS

- Probability of Failure ~

Ukellhood Failure

BAYESIAN UPDATING

Prior Distnbution of B Combined Likelihood Function

Posterior Dlstnbution of B

BIAS FACTOR

Mean Value cov bull

Represents Modeling Uncertainties

Prior Distribution of B Likelihood Functionmiddot Observing Success case STISIK and Posterior Distribution of B

2 00

uo

160

140

120

e 100 L $ uoIshy

0 60 ~ ~ 040

020

0 00

0 00 0 50

_ I 00

b

I 50 2 00

-shy lk [bl STISIK)

--shy Pnor-f(b)

--shy Pos1enor r(bl ISIK)

Prior or B Mean bull 10 COV bull 030

Posterlor of Bbull Meanbull 136COV 0 16

2 50

Likelihood Function and Posterior Distribution Example Platform STISIK

- -

~

- Prior and Posterior Distributions or B Failure cases (ST177B STlSlH ST130A T21) Damage cases (T23 T25 ST161A) Success cases (STlSlK ST130Q ST134W WD90A MC311 MC397)

400

350

300

2 so 0-

I

bull bull

bull bull gt

- bullbull ~ bull bull I bull

_ bull bull bullbull I bull bull bullII V -

bull bull

I bull ~ bull

bullbull bullI bull bullI bull bull bull

middot -~

bull__ ~ middotI middotmiddot

L 2 00

e ~ I SO

-I 00

- 0 so

000 gt

0 00 020 0 40 060 0 80 I 00 120 140 160 110 200

b -

-

shy

Prior-r

- - - - Posterior fLlll failure cases

bull bull bull bull bull bull bull Posterior flall damage cases

- bull bull - bull middot Postenor flall success cases

Posterior flall success fulure damage cases

-

Posterior of B Mean= 119 COV 0101

Posterior Distribution or Blas Factor (8) bull Different Categories and All Cases Combined

shy

l

PHASE I - KEY CONCLUSIONS

bull GLOBAL CORRECTION FACTOR

bull 15 TO 20 CONSERVATISM MEAN)

bull UNEXPECTED SURVIVALS HAVE GREATEST IMPACT

bull DAMAGED CASES MORE INFORMATIVE

bull CALIBRATION ISSUES PHASE II

bull

EFFECT OF IMPROVED CAPACITY ANALYSIS RECIPE

bull APPLICABILITY TO MULTIPLE FAILURE MODES

bull NO FOUNDATION DAMAGE CASE INCLUDED

bull IMPROVED REPRESENTATION OF FULL ANDREW DISTRIBUTION

-bull NUMBER OF PLATFORMS CONSIDERED

I 1 I

I t ) bull ~

PHASE II WORK SCOPE (CALIBRATION)

bull

bull ESTABLISH MULTIPLE BIAS FACTORS

Use Nine Base Case Platforms

middot - Capacity Analysis for Revised recipe

Foundation Blas Factor from Foundation JIP

bull DEVELOP WEIGHTING APPROACH

bull UPDATE MULTIPLE BIAS FACTORS

bullEffect of Platforms Not In the Nine Base Cases

bull EXAMPLE APPLICATION OF BIAS FACTORS I

bullbull

PHASE I PHASE II

CAPACITYbull ANALYSISRu bull

---------------------1----middot

Jc1 - iJbullraquoJ dbull

FttlS frt[f~l ratdhJJ

-

LIKELIHOOD FUNCTION

DEVELOPMENT

lk Jb Jfall1N) bull Jb)

111 Ill Isurvtvoq bull 1 - bl

-middot -------------------shy ~---------------------middot-middot

r111ir8 Cb1 111 Cbl BAYESIAN

UPDATING

~cove

R1R2 bull middot---------------------~----

Jblbull P(g cOJ

1 bull JbRI (MBSI I bull 1

~middot

llcJllJfalluro) bull P(gcOJ Ill 111 b21fallun) bull P ((g1 CO) U 1112 c OD

middot--------------------shy ----------------------shy bull

r111111r8 Cbl 111 Ill i-1nro1

deg ~1b2)laquottfrr (b1b21-lnfo) re Ill bull 1-11middot b2) bull lodegamp

middot

i1 covamp i2 cov~i

-

shy

--

Likelihood Function vs b All Success Case STISI K STl30Q STl34W WD90A MC311 MC397

I 0000

0 9000

OIOOO

0 7000

~ 0amp000

ta osoOO

~ 0 4000

0 3000

0 2000

0 1000

bull 0 0000

bull

I

middot

middot bull I

Ibull

-sibull ~ -middot-middot-middot-middot -middot-

----7----+-----~ I

Ibull bull bull I

rbull

--- Jk ISIK[blsuccess)

- bull - bull - lk 130Q[blsuccess) bull

lk 134W[blsuccess)

Jk-WD90A(blsuccc~I

- - - lk MC397[blsucccss)

--ltgt-- lk middotMC31 l[blsucccss)

--- Jk [bl 6 out of6succcsHascs)

05 I 5 2 250

b

Likelihood Functionsmiddot Success Cases - Phbullsc L

shy

bullbull bull bull bull bull bull -------------------~

bull bull bullbull bullbull bullbull bullbull bullbull bullbull bull

bull bullbull

bullbullbullbull

bullbull

bullbullbullbullbullbullbull

bullbullbullbullbullbull

bullbullbullbullbullbullbull

z 0

t z ) LL

c 0 0c i ~ i

J

~ sect ~--------------~ e l5

I

I

SOFTWARE

bull PF PROGRAM

bull CALREL

bull PF +CALREL

bull RELACS

bull

gt i

l

SENSITIVITY STUDIES

POSTERIOR DISTRIBUTION OF B

bull EFFECT OF CHANGE IN DISTRIBUTION OF PRIOR OF B

bull EFFECT OF REMOVAL OF ONE PLATFORM FROM EACH CATEGORY

bull EFFECT OF REMOVAL OF THREE DAMAGE CASES

bull EFFECT OF ADDITIONAL 13 SIMILAR PLATFORMS

-I

Prior and Posterior Distributions of B Variation In COV of PriorofB fromOl to04 -- -

4500

-4000

-

- -

3500 --middot - bull bull - bull bull Prior-r ICOV =02

3000

- - - - Prior-r ICOV =03 -

bull -~ e 2soo -

bull bull bull bull bull bull bull Pnor-r ICOV =o 4t frJ ~ 2000 Posterior rlCOV =0 2

bull bull Postenor rlCOV =0 3 bull bull 1500 bull middot bull - shy -middot- Posterior rlCOV =04

middot~ I 000 - bull

-middotmiddot ~ ~middotmiddot M -~

- middot 0 500

wbullbullbull _ I middot~ middotmiddotmiddotmiddotmiddot middotmiddotmiddotmiddotmiddot - -~ ~ __ middot_ _ bull -shy0 000

0 000 0200 0 400 0 600 0 800 IOOO 1200 1400 1600 1800 2000 -

- - b

-

=shy

Posterior Distribution of Bias Factor (B) bull Effect of Variation In COV of Prior of B

bull

bullPrior and Posterior Distributions of B Evaluating Effect of Removal of One Platform from Each Category from the Base Case of 13 Platforms

400

~ ~ 350

j IA i

3 00 bull I bull I ---Prior-r I bull l

250 1 I 1

I bull bull ---Posterior r lall 13 casesshy- ~E 2 00 bull bull bull bull bull bull bull Posterior r without STISIK

- ~ 1 bull - bull I - bull bull - bull bull Posterior r Iwithout T2I I 50 I

J-tt~i - bull - bull - Posterior r Iwithout T2S

tOO [-

I I ~ ~ I 0 50

0 00

___ r J i

l~T

~--

0 00 0 20 0 40 0 60 080 I 00 120 140 160 110 2 00

b

zshy

Posterior Distribution of Bias Factor (8) bull Effect of Removal of middotPlatform from Any Category

bullPrior and Posterior Distributions or B Evaluating Effect or Removal or 3 Damage Cases from the Base Case or 13 Platrorms

4

35

3

25

e L 2 ~ bulle shy

I 15

bull I

05

0

0 02 04 06 I 2 I 4 16 IS 2

I I

I

I

I Ibull

J -tmiddot L--

bullI

~

I

fl

II ~

I I v I

middot bull bull

bull L

~ ll- I-~lt

OS I

b

I

bull

1middot bullbullbullbull

~ __

bull

- - - - Prior-r

- - - - Postenor flall failure cases

Postenor flall success cases

--- Posterior flall 13 cases

---Posterior fI success failure cases only

IPosterior or B Mean =t11 COV =0121

-

_I

Posterir Distribution or Bias Factor (B) bull Effect or Removal or 3-Damage Cases

-

bullPrior and Posterior Distributions or B Evaluating Effect or Additional 13 Similar Observations as

Used in Andrew JIP The POsterior or B from Andrew-JIP Considered as New Prior or B

- 500

450

400

350

bull bull

bull bull bull bull J bull

bull bull f ~ I

I ~ Ibull bull

bullbull bull

l bull I -bullbull bull bull

I

I bull bull

bull I I~~bull bull bull

bull ~

I bull bull

~ ~_ ~~~

- - New Prior r bull

Posterior 114 new failure cases ~ 300- bull bull bull bull bull bull bull Posterior f 13 new damage casesl 150 ~

- bull bull - bull bull Posterior 116 new success cases i 2 00

Postenor 1113 new success failure -150 damage cases

I 00

0 50 -

0 00 0 00 0 20 0 40 0 60 0 80 I 00 120 I 40 160 110 2 00

Posterior ol B Mean= 120 COV= 0071 b

shy

shy

_rshy

Posterior Distribution of Bias Factor (B) bull Effect of Additional 13-Similar Observations

1

1 I

WEIGHTING APPROACH bull

bull ASSESSMENT OF PLATFORMS DATA

Select Platforms Most Mappable to the Analyzed Platforms

Unlikely SurvivalDamage Cases Identification

Based on Cut-Off Criteria Agemiddot Hs etc

bull INFORMATION REQUIRED FOR COMPARISON

bull CATEGORIZATION OF ADDITIONALmiddot PLATFORMS

bull UPDATING OF CALIBRATION RESULTS

--

OVERVIEW OF PLATFORMS SURVIVEDDAMAGED -~

Operator -

- Platforms

Survived CS)

Platforms

Damaged CD)

Platforms Used In

Phase IJIP -

JI~ ~Bill~IRBDll -

Amoco -

Chevron -

Exxon

Mob II

Phllllps

Shell

Unocal -

30

143

72 22

7

39 25shy -

1

12

bull

1 bull

1

1(S)+1 (D)

3 (S) +3(D)

2 (S)

bull shy

bull

bull

bull

Total Partlclnants Platforms 338 - 15 10C6S+4D) -

Other Operators

310

13

3 (D)

Total Platforms 648 28 13C6S+7D) -

Summary of Andrew JIP bull Phase I Platforms

Plairshy w v- N-ber Lea-le 1 shy car Hind_ o-w N-bull Deplh fl fl IJpe la bull n Yeadon Jadtet lbelledtf liefalit -- _ -OD

IL H - -shy _ICllshy IL IL - - rQClto ~ _

K STUIK 137 1963 8 365 608S4 Cellardeck

-middot STl-shy 170 1964 4 -~ K 39 62269 Cellardeck - --ST134W 137 1911 4 43 608S4 Smvlnlshy- _ - -

WD90A 184 1964 8 K 267Sbull34 83 SOS31 SUboftllar deck SmvMI

MC311 343 1978 8 x S7 62 - MC397 468 1991 4 x shy S33bull6SO 65911 -shySmeSmvlvol -- gt -

r-- T23ST52 63 1969 4 ton- shyKshy 52594 S03 -T25SSl39 62 1969 4 Kshy SOS94 S063 - middot ton-shy__ -

bullbull a ton--shyST161A 118 1964 8 K - 347S 51495 Cellarclecti

r~

ST1771 142 1965 8 K 39 60167 CeDordeck failme

STISIH 137 1964 8 K 395 608S4 CeDardeck Failme

K ST130A 140 1958 a SS 6095 I lblv Pai1uno - T21ST72 61 1969 4 Kshy S4S94 4974 I lhlv Pai1uno -

- -

JE

INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON

bull WATER DEPTH

bull bull YEAR OF INSTALLATIONDESIGN

bull PLATFORM ORIENTATION

bull PLATFORM LATITUDE AND LONGITUDE

bull KEY SKETCHES OF PLATFORM STRUCTURAL FRAMING

Number of Legs Plies Leg Batter Number of Conductors bull Deck Elevations (Including sub-cellar deck If any) Jacket Framing Elevations Brace type In Vertical Frames Sizes of Primary Members (leg pile vertical frames braces primary horizontals) Joint Details (can gap etc) LegPiie Annulus Grouting Status

bull middotPOST ANDREW INSPECTION REPORT Significant Wave Height Hs (ft)

bull INFORMATION ON DAMAGES PRIOR TO ANDREW

bull RESULTS FROM IN-HOUSE STUDIES IF AVAILABLE

bull ~RIGINAL DESIGN CRITERIA INFORMATION IF AVAILABLE

bull SOIL SHEAR STRENGTH PROFILE IF AVAILABLE

Platforms Identified for Evaluating Therr Likelihood of Falling Under Unexpected Survival Category

OPlRATOR

middotADA

BLOClt

llmucnJllE

NAME

YEAll

INSTAUJD

WA1ER

DBPTll

DISTANCE

JlllOM SHOllE

(MILU)

Bl

(M)

AMOCO 111113

STl11

WDI075

A

bull D

St

17

120

1n

44

20

10 I

CllllVllQfl STlllO c 61 110 21 11 STIUO D Q 1110 21 11

STl134 p Q 137 29 11

ST0151 G 61 137 11

ST0151 I 3 121 11

ST0151 J Q 140 11

STl151 It 137 11

ST0151 PJl001 Q UT 11

ST0151 PROD-2 Q UT 11

ST017 A 3 140 35 11

WDIUT c 65 214 34 10

llXXON STI055 II 51 6T 14 I

STl15 A t3 34

WDI073

WD0073

Amiddotshy Q

1111 1111

22

22

I

I

WDI073 c 1n 22 I

- WDOCl93-shy II

65 1110 2T A 3 194 31 MOBU 111121

SS IOl7

A B

501

50

50

31

T

I

T 6

MtlllPllY SS114 B T 50 T T

PlllllIPS SS0149 A 5 55 33 T

SAMllDAN STln A 107 T T

SBllLL WDl103 A 65 223 2T WDl103 B 65 221 2T WD0104 c 65 221 2T WD0104 D 6T 20 2T

UNOCAL SS020I

SS 020I SSl20I

SS 0209

SS0209

II p

PURll

A

B

3

n 61

103

97

97

95

100

34

34

34

34 34

bullbull bullbullbull

- SS0209 D u 95 34 I

- SS 1209 SSl253

SS0253

G A

c

6T

2

100

165

1T5

34

54

54

bullbull I

I I

I I

SOUTH MARCH

~LAND

N

SOUTH PaTOI ENE ISWltD AREA

MMS Defined Limits ror Offshore Structures Requiring Inspections Following Andrew

t---Jbulln3 T25bull I bullTtl

bull1$1Htl1Abull bull1SIK1779bull

~

middot-DaTA AREA~T

CRANO rsLE

AREA bull1300

bullll4W

=shy

bullPath of Hurricane Andrew with Platforms Used In Calibration

Likelihood Function vs b All Success Case STISI K STIJOQ ST134W WD90A MCJll MCJ97

I 0000

0 9000

0 8000

07000

-Bo6000

tiiii 0 5000 -

04000

0 3000

0 2000

01000

0 0000

0 05 25

ri1rt middot-middot-middot-middot-middot-

I 5 2

b

---11 middot151K(blsu~ccsI

- bull - bull - fk 130Q(bfU~lCSS)

lie bull134W(blsuccc]

lk-WD90A[blsultCS]

- - - lk middotMC397(bfUtlCS]

--o-- lk-MCJll(bbuctcsJ

--- ll (bl 6 out of 6 Mlttcs case]

_

( Likelihood Functions middotSuccess Cases - p - 1 c I

CATEGORIZATION OF ADDITIONAL PLATFORMS

I I I

bull CATEGORIES

Unexpected Survivals

Expected Survivals

bull APPROACH

Identify Corresponding Platform(s) Evaluated In the JIP

Determine Max Wave Height at Platform Location

Evaluate Whether Wave Hit the Deck

middot - Evaluate Likelihood of R and S to Differ from the Base Cases

Review Joint Information

Assess Special Features

bull IDENTIFY NUMBER OF PLATFORMS IN EACH CATEGORY

I

UPDATING OF CALIBRATION RESULTS

bull INCLUDE EFFECT OF ADDITIONAL PLATFORMS ON bullBbull

Consider Increased 1Uncertalntles In R and S

Develop bullNeW Ukellhood Functions for Different Categories

middotLet bullxbull Number of Ukely Unexpected Survival Cases and y Number of Ukely Expected Survival Cases

Determine Combined Likelihood Function lk (b Iadditional platforms) middot

bull REVISED LIKELIHOOD FUNCTION

lk(b I all cases) = lk(b I 9 platforms) middot x lk(b Ix platforms) x lk(b I y platforms)

bull ESTABLISH REVISED BIAS FACTOR B

f8 (b) u f8 (b) lk (b I all cases)

II

t t ~

shy

APPLICATIONS OF BIAS FACTORS

ii APPLICABLE TO PARTICULAR LOAD AND RESISTANCE RECIPE

bull USE BY THE API COMMITTEES AND MMS

Criteria Revisions

Recommended Design Guidelines Revisions bull

bullbull REASSESSMENT STUDIES AND DECISION MAKING

middot

bull RISK ASSESSMENT STUDIES BY THE OPERATORS

Individual Platforms

Fleet of Platforms

middot IMR Program Implementation

bullbull NOT TO BE USED IN DESIGN

I I fl J

EXAMPLE USE OF BIAS FACTORS

bull PHASE I EXAMPLE APPLICATION

bull EVALUATION OF ANNUAL PROBABILITY OF FAILURE

Establish Annual Maximum Seastate Data

Determine Coefficients to Define Base Shear

Determine Ultimate Capacity of Platform

Determine Probablllty of Failure Versus bullbbull

-bull DETERMINE UPDATED PROBABILITY OF FAILURE

Pf - P [S gt R]

Pf I P [~ gt Rb I b] f 8 (b) db

Pr - J P [S gt Rb I b] f bull 8 (b) db

bull TOTAL PROBABILITIES OF FAILURE

Total Pf - 1 bull Il [1 middotPf] - l Pf bull

Total Pf - 1 bull Il [1middot Pf] - l Pfmiddot

Total Pr = 1 - n [1middot PfbullJ = l Pfshy

I I I

EXAMPLE APPLICATION bull PHASE I

PLATFORM ST 151 K

bull GULF OF MEXICO

bull WATER DEPTH = 137 FT

bull ANNUAL MAXIMUM HS MEDIAN 175 ft COV 031

bull ASSUMED STORM DURATION 3 hoursmiddot

bull ASSUMED NUMBER OF WAVES 800

bull MEDIAN HS FOR STORM HOURS 1 2 AND 3

bull MEDIAN CURRENT VALUE bull

bull DETERMINE ULTIMATE CAPACITY FOR THREE DIRECTIONS

bull DETERMINE ANNUAL FAILURE PROBABILITIES

Broadsidemiddot Diagonal End-on

0037 0025 0007

0071 0044 0018

0022 0016 0004

DETERMINE TOTAL ANNUAL PROBABILITY OF FAILURE

Pf - 007

Pf - 013

Pr - 004

bull

EXAMPLE USE OF BIAS FACTORS - PHASE II middot

bull DETERMINE PREDICTIVE PROBABILITY OF FAILURE

Case - 1

Pf _ S S Pf (b b2) fa1a2 (b b2) db db2

g bull -1 (b1 R I S) - 1

92 - (b2 R2 I S) - 1

Case -2

Pf s ~ [bJ f11 (b) db

Pf [bJ p [g lt 0)

g - bull 1 (b1 R1 I S) - 1

1

bull1) l

I

BENEFITS OF ANDREW JIP bull PHASE II

lt

bull IMPROVED LOAD AND RESISTANCE RECIPE

bull IMPROVED CALIBRATION PROCEDURE

bull MULTIPLE BIAS FACTORS

bull PROCEDURE TO INCLUDE EFFECT OF OTHER PLATFORMS

bull PROCEDURE TO USE MULTIPLE BIAS FACTORS

  • Structure Bookmarks
    • ANDREW PHASE II JOINT1 INDUSTRY PROJECT
    • PROGRESS MEETING NUMBER 1
    • ANDREW PHASE II JOINT INDUSTRY PROJECT PROGRESS MEETING NUMBER 1 NOVEMBER 16 AGENDA
    • ANDREW PHASE II UST OF PARTICIPANTS
    • ANDREW PHASE I OBJECTIVES AND RESULTS
    • ANDREW PHASE II PROJECT OBJECTIVES
    • SUBSEQUENT PROJECTS
    • ANDREW PHASE II PROJECT TEAM
    • ANDREW PHASE II UPDATE MODELING middotAND ANALYSIS RECIPE
    • -ANDREW PHASE II CONDUCTOR RESISTANCE
    • ANDREW PHASE Il WA VE FORCE PROFILE
    • ANDREW PHASE II WA VE FORCE PROFILEmiddot
    • ANDREW PHASE II BRACE MODELING
    • ANDREW PHASE II FOUNDATIONS
    • ANDREW PHASE II middot middot JOINT MODELING
    • ANDREW PHASE Il JOINT MODELING OPTIONS DISCUSSION
    • ANDREW PHASE II JOINT MODEJJNG OPTIONS ILLUSTRATIONS
    • ANDREW PHASE II JOINT MODELING PHASE II METHOD
    • ANDREW PHASE II ANDREW STORM CONDmONS UPDATED
    • Platform Selection bullbullbull1111111111111
    • BasisCriteria for Selection bullbullbull1111111111111
    • CALIBRATION
    • WHERE TO INTRODUCE BIAS FACTOR (B
    • PHASE I -KEY CONCLUSIONS
    • bull CALIBRATION ISSUES PHASE II
    • PHASE II WORK SCOPE (CALIBRATION)
    • SOFTWARE
    • SENSITIVITY STUDIES POSTERIOR DISTRIBUTION OF B
    • WEIGHTING APPROACH bull
    • INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON
    • CATEGORIZATION OF ADDITIONAL PLATFORMS
    • UPDATING OF CALIBRATION RESULTS
    • APPLICATIONS OF BIAS FACTORS
    • EXAMPLE USE OF BIAS FACTORS
    • EXAMPLE APPLICATION bull PHASE I PLATFORM ST 151 K
    • EXAMPLE USE OF BIAS FACTORS -PHASE II middot
    • BENEFITS OF ANDREW JIP bull PHASE II
Page 20:  · . ' ANDREW PHASE II PROJECT OBJECTIVES . ' • ·UPDATE MODELING AND ANALYSIS RECIPE· JOINTS ' ' ·BRACES WAVE PROFILE FOUNDATIONS . ' ' CONDUCTORS· • ANALVZE PLATFORMS USING

bullbullbull

II ~------------________

ANDREW PHASE II

WA VE FORCE PROFILEmiddot

CAP AUTOMATIC APPLICATION FEATURE

STAT

fR1 0 lt f lt1

STAT

STAT

f(R3-R2) middot f lt1

Project ST151K Model Wave_Loading Version 1 rue Nov 8 164107 1994 PT

E~ nss FT

SL 14amp FT

a 111 FT

EL 14 li

E-1 37 FT

El o FT

1middot1middot ______________ _ ________ __ _ __

0

200 400 600 800 1000 12000 Force at Elevatloii (kip)

200

a ll_

Wave Force at Major Elevations for Discrete Wave Heights

180

bullHmiddot40ft

Hmiddot495ft

H bull 51 ft

Hmiddot55ft_

H-59ft I

Andrew (H bull 60 86 ft) II

i I

_11---__

- middot- Hmiddot30ft

-

- bull -

middot bull bull bull middot

---Hmiddot53ft

bullbullbullbullHmiddot57ft

-----H-63ft

Elevation Above Mudllne 100

(ft)

180

140

120

80

60

40

20

Loading at Major Elevations

---H-49511

-deg-Hmiddot608611

0 200 400 600 800 1000 1200

Force It Elevatlon (kip)

--

Base Shear vs Wave Height ST151K

6000

5000 1---------Ult1mateCapacitybull 4890 kip (100 Improved Soll)-----------~--

4000a a lii ~ 3000

81 ampI

2000

1000

0 +-~~~--i~~~~~~~~~-+-~~~--lf--~~~~~~~~-+-~~~--l

30 35 40 45 50 55 60 65

Wave Height (ft)

bull middot

~

bull

II

2

g

~

It)

0

cCD E 0 E a c E euro~ CDshygtItgtmiddoto ~ti c -a Gic CD gt ~

-= c -Cll

I z

bullO 0i i ~ g 0

() CI -bull(q-ll) iuawow llu1wniiaAO

11

shy

116

114

112

g 110

bullc = 108

~ ~ 106 -sect 104

102

100

98

96

-

6555 60

Wave Height vs Center of Action ST151K

30 35 40 45 50

Wave Height (ft)

t

Comparison of Pushover Curves for ST 151K Variable Load Pattern and Monotonic Load Pattern (Andrew)

Soll Improved 100

--- --------- --- ----6000

5000

- 4000 8 - I

1 3000 Ul

81 m 2000

_J -shy

1000

---Variable Load

- - Andrew Load

0 f-~~~~~r--~~~~-t-~~~~~-t-~~~~~+-~~~~----t~~~~~-1

0 1 - 2 3 4 5 6

Deck Displacement (ft)

-middotshy

1 t I I ~

ANDREW PHASE II

BRACE MODELING

bull MARSHALL STRUTS USED FORPHASE I

bull PHASE II WILL ASSESS DIFFERENCE IN USING BUCKLING BEAMS AND MARSHALL STRUTS

middot bull COMPARE FOR 2D FRAME

bull COMPARE FOR PLATFORM 177 WHERE LOWER

BAY EXHIBITS LARGE BRACE MOMENTS

Project andrew Model beamframe Version 1 Mon Nov 14 145447 1994

~

middot Frame Test buckling beam-column model

-

Project andrew Model strutframe Version 1 Mon Nov 14 145754 1994

bull

Frame Test strut model

middot-middot _____________--_

_middotmiddotr_~ --ll--~-middot__

Buckling Beam bull

Strut (K=O 65)

Lateral Displacement

ANDREW PHASE II

FOUNDATIONS

bull ~LICIT NONLINEAR API FOUNDATION USED INiPHASE I

bull FOUNDATIONS

JIP SELECTED MINI-VANE DATA FOR

STRENGTH PROFILE

bull COMMENTS ON BENCHMARK COMPARISONS

- SIGNIFICANT VARIABILITY DUE TO FOUNDATION MODELING METHODS

II Ii

z

ctV 4x SPlO caissonfy-42s soil-api(lf-097step-114)

Inelastic Events Legend Elastic Strut Buckling

Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield

--------shyBeam Clmn Fully Plastic Fracture

~--~~~~~~~~~--~----~__i~~---~~~~__--~

Pro)ect apipile Model South Pelto 10 Version Wed May 18072235 1994 - -

1bullt

+

t

f i bullbullbull

bullbull bullbull

I I

5

~x SPlO Caissonfy=42s soil=api(lf=lOOstep=137)

Inelastic Events Legend

Elastic Strut Bucklilg

Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield

earn Clmn Full~ Plas1c Fracture

ANDREW PHASE II middot middot

JOINT MODELING

middotbull API RIGID-PLASTIC USED IN PHASE I

a ELASTIC-BRITTLE USED IN PHASE IB

amiddot MSL WILL PROVIDE NONLINEAR FORCE-DEFORMATION

AND MOMENT-DEFORMATION RELATIONS FOR PHASE II

a MODELING OPTIONS FOR PHASE II -

l

- UNCO~LEDmiddotONE DIMENSIONAL MEMBER END JOINT MODELING

J Ii

ANDREW PHASE Il

JOINT MODELING OPTIONS DISCUSSION

bull JD COUPLED INTERACTION SURFACE

- IDEAL

- DATA NOT AVAILABLE TO CALIBRATE

bull 2D COUPLED INTERACTION SURFACE

- SIGNIFICANT MODELING IMPROVEMENT

- DATAAVAILABLE

- IMPLEMENTATION PRACTICAL BUT NEEDS DEVELOPMENT AND CALIBRATION

bull 2D UNCOUPLED INTERACTION SURFACE

- SIGNIFICANT MODELING IMPROVEMENT

- DATA AVAILABLE

- IMPLEMENTATION PRACTICAL AND AVAILABLE

ANDREW PHASE II

JOINT MODEJJNG OPTIONS ILLUSTRATIONS

P

e

ANDREW PHASE II

JOINT MODELING PHASE II METHOD

e

I I I

ANDREW PHASE II

ANDREW STORM CONDmONS UPDATED

bull -PHASE I USED OCEANWEATHER HINDCAST-(Nov 1992)

bull middotmiddotPHASE II WILL USE THE REVISED HINDCAST (Nov 1994)

bull lt c lt I lt

2-2

l---4-~~~+-~~-+-~~~1--~~-+-~~--l--~T--+-~~--i24

f-tJVtMfiz~ 1Cfl (r STvb(

Plollol bullbull 5middotNOYmiddot9214bull1J Irbullbull Illbull [ANDREVl~NDREVCUSTbullMAXll 5-NOY-1992 10bull54 - 9208

HEIGHT lml

I~ MAXI

I L 1~middotamp SZ~- ~A I 13c

I J) ( r I X~0iWJ~0gtJ gt L gt gt I 1c I 120

-- N

~ deg

I I I I I I e C I ~=-41 122 -94 -92 -90 -88 -86 -84 -82 -80

-

Figure 7 contours of maximum hindcast significant wave height

_

~

~tl ~

NOltNttIJ 17lt ltNS S iJJ)y

Maximum Significant Wave Height - Hurricane Andrew 1992

30 ~

26 ~ I ~-~~~- == -shy -~ 35-_

24deg1--shy --------~- middot-middot-------middotmiddotmiddot----middot------middot-middot~middot- - - ~-----1---middot

22middot ~~~~~~~-~ -94 -92 -90 -88 -84deg-86 -82 -somiddot

c

-

Platform Selection bullbullbull1111111111111

0 BasisCriteria for Selection middot

0 Review of Phase I Platforms middot 0 Alternative Platforms

0 PMB Recommendations

0 Participants Comments

gt

L

l I

BasisCriteria for Selection bullbullbull1111111111111

0 Benefit for Calibration - Classification (UnexpectedExpected)

- Failure Mode (StructuresFoundations)

- Damaged StructuresKnown response

- Better Representation of Full ~opulation

0 Availability of Data - Existing Models

- Inspection Data

- Damage Data

PREDICTED BEHAVIOR

-

OBSERVED

BEHAVIOR

-SURVIVED DAMAGED FAILED

SURVIVED

ST134W WD90A

-MC311 -

MC397 -

--

-

-

ST151K ST130Q

-

-

-

DAMAGED - -

-

-

--

-

-

ST52 SS139 ST177B -

ST161A

-_

-

FAILED

-

-

-

ST151H ST130A ST72

Operator - Platfonns - Platfonns Survived Damagedamp

- Falled -

bull

-

-

Jle eBDIGIPllDbl -Amocoshy 30 1 Chevron -

143 12 Exxon 72 Mobll 22 1 Phllllps 7 bull

-Shell 39 Unocal 25 1

Total Partlclnts Platfonns 338 15

Other Operators bull- 310 13

-

Total Platfonns - -

648 28

fbml

Platforms Considered 6 7

tie of Total Platfonns 1 25

OVERVIEW OF PLATFORMS SURVIVEDDAMAGEDFAILED

-

shy

v

Platforms Evaluated in Phase I

-

-Platfonn Water Year Number Analysis Avallabillty of Availability Platform

- Name Depth rl of Model DetaDed rl ClassUlcatlon

Installation legs Inspection Soll Report Based on Information Phase I Calibration - -ft

- -- Suaill flalfnrm ~BSi -STISIK 137 1963 8 CAP Yes Yes Un eel Sarvlval

-STIJOO 170 1964 4 CAPmiddotshy Yes ST134data u ST134W 137 1981 4 CAP Yes Yesmiddot E

-

edSarvlval -

middotsurvival

WD90A 184 - 1964 8 CAP Yes E -

- Sarvlvill

MC3ll 343 1978 8 KARMA

MC397 468 1991 4 KARMA -

bull ~reglmmiddot- -

Yes Sure Survival

bull Yes - Sare Sanlval-

-T23STS2 63 1969 4 CAP

T2SSS139 62 1969 4 CAP -

Yes

Yes

Yes Ex

Yes Une

bull SurvivalLikely to DamalI

middot bull SanlvalLlkely to DamalI

ST161A 118 1964 8 USFOS

Eail11a pound1al(bullllm Clss ST177B 142 --1965 8 CAP

STISIH 137 1964 - 8 CAP C

STIJOA 140 19S8 8 CAP - -

Yes

Yes

ollamed

Yes

- Yes E

-

at about 1 mUe

Yes

ST134data

bull SafvlvalLlkelv to Dam- bull- -

E middotrahre

Emeried Failure -

Likely FaUare

T21ST72 61 1969 4 CAP Collamed No Likely FaUure

Platforms Identified for Evaluating Their Likelihood of Falling Under UnexpeCted Survival Category

CllEllATOa

AllB

llLOCll

Sl1lvcnJllE

NAME

YEAR

INSTAUJCD

WATER

IJBPJB

middotmiddotDISTANCE

ftOMSllOllE

(MIUIS)

Ill

(M)

AMOCO BllW

STllI

WDI075

A

bull D

4

n uo 17J

44

32

bull 20

10

bull CllEVllOH STtllO c I 1bull 21 11

STl130 D 0 161 21 11 STllJ4 I 0 137 D - 11

STIUI G I 137 32 11

STIUI I 0 121 32 11

STIUI

J 0 140 32 11

STIUI It 4 137 32 11

STllSI PROl)I 0 137 32 11

STllSI PROD-2 0 137 32 11

STll7 A 0

140 35 11

WDl117 c 65 214 34 10

DXON STshy E 51 14 bull STl165 A t3 bull 34 WD0073

WD0073

WD0073

A _ c

4

62

4

IA IA

17J

22

22

22

bullbullbull WD003-shy E 65 161 Z7

A 0 IN 31 MOBD Bl12I

ssn A bull

50

50

31

7

bull 7

MIJllPllY SS114 B 7 50 7 7

PlllUIPS SSl14t A 55 33 7

SAMEDAN ST17J A 4 117 7 bull 7

SJIELL WD0103 A 65 223 27 WDllOJ bull 65 221 27 WDll04 c 65 221 27 WDOI04 D 67 2 27

UNOCAL

SSl20I

SSl20I

bull SSl20I

SS0209

SS0209

SS0209

SS0209

SS 0253

SS 0253

E

I

PtJllE

A

bull D

G A

c

0

4

4

7J

61

0

0

69

103

97

97

9S

100

95

100

165

175

34

34

34

34

34

34

34

54

54

bullbullbullbullbullbullbullbullbull

Platforms Recommended for Phase II

llalr y_ w- Aot Av- _ I 1 Doplla Model nm bull bull _ bull I I H11 s18-1 Irht lbwl~ bullbullbull(-illtdlel OM lllMtD

II ---

-STISIIC 137 1963 I CAP Yeo Yeo I Yeo

STtshy -170 1964 4 CAP Yeo ST134dlU I Yeo

STl34W 137 Yeo Yeo -1981 4 CAP D Yeo

WD90A 114 1964 I CAP Yeo - D

MC311 30 1971 I KARMA Yeo -sre-

MC397 middot 461 1991 4 ICARMA Yeo Sireshy - - r- -

__ T13ST5l 63 1969 4 CAP Yeo Yeobull middot- ~n- D Yeo _ Tl5SSl39 62 1969 4 CAP Yeo Yeo I Yeo

_ STIT711 142 1965 I CAP Yeo I Yeo middot-1shySTl61A Ill 1964 I USFOS Yeo Yeo - D bulloshy

- r-

STISIH 137 1964 I CAP y D y -_ STl31A 140 1951 I CAP Yeo STl34dlU I Yeo

T21Sll 61 1969 4 CAP No rar-middot I Yeo

~-rmiddot-middot SS209A 95 1967 16 CAP Yeo D -SSU4R

11 UI A 41 -in D SS 2741AIM CAP bull

_ _ -shy - -shy- -shy

CALIBRATION

bull PHASEIAPPROACH

bull CALIBRATION ISSUES

bull PHASE II WORK SCOPE

bull SOFlWARE AVAILABILITY

bull EFFECT OF ADDITIONAL PLATFORMS

bull WEIGHTING APPROACH

bull EXAMPLE APPLICATION middot

(l=Ofce Mode] Wind

~

fcurrent l -middot

~SW ~leldl4

ultimate resistance force I first

design load Illgt

i _

deflection

-

bull

4

Historical Hurrlcane Dariiage and Survival of Platforms

Components Required for Calibration (Petrauskas 1992)

PRIOR Distribution

PLATFORM EXPERIENCE

BAYESIAN Survivals and Failures

UPDATING

Waves Winds and Currents bull

POSTERIOR Distribution

- A_middotJA ~ Survival Data Failure Data

Result Result

Load Resistance

middotfl

-

Bayesian Updaing middotGeneral Description (Petrauskas 1992)

L Imiddot

WHERE TO INTRODUCE BIAS FACTOR (B

On Resistance (Rmue - B (RPREDICreD I

middotOn Loads (Smue middot - B (SPRE01cTEo

On Safety Margin (RSmue - B (RSPREDICTED

CALIBRATION METHODOLOGY

gt

r

FIELD OBSERVATIONS

Observed Behavior

survivals Damages

Failures

r

bull

CAPACITY ANALYSIS

Predicted Load and Resistance Predicted Failure Modes

Survival Damage Failure cases

RELIABILITY ANALYSIS

- Probability of Failure ~

Ukellhood Failure

BAYESIAN UPDATING

Prior Distnbution of B Combined Likelihood Function

Posterior Dlstnbution of B

BIAS FACTOR

Mean Value cov bull

Represents Modeling Uncertainties

Prior Distribution of B Likelihood Functionmiddot Observing Success case STISIK and Posterior Distribution of B

2 00

uo

160

140

120

e 100 L $ uoIshy

0 60 ~ ~ 040

020

0 00

0 00 0 50

_ I 00

b

I 50 2 00

-shy lk [bl STISIK)

--shy Pnor-f(b)

--shy Pos1enor r(bl ISIK)

Prior or B Mean bull 10 COV bull 030

Posterlor of Bbull Meanbull 136COV 0 16

2 50

Likelihood Function and Posterior Distribution Example Platform STISIK

- -

~

- Prior and Posterior Distributions or B Failure cases (ST177B STlSlH ST130A T21) Damage cases (T23 T25 ST161A) Success cases (STlSlK ST130Q ST134W WD90A MC311 MC397)

400

350

300

2 so 0-

I

bull bull

bull bull gt

- bullbull ~ bull bull I bull

_ bull bull bullbull I bull bull bullII V -

bull bull

I bull ~ bull

bullbull bullI bull bullI bull bull bull

middot -~

bull__ ~ middotI middotmiddot

L 2 00

e ~ I SO

-I 00

- 0 so

000 gt

0 00 020 0 40 060 0 80 I 00 120 140 160 110 200

b -

-

shy

Prior-r

- - - - Posterior fLlll failure cases

bull bull bull bull bull bull bull Posterior flall damage cases

- bull bull - bull middot Postenor flall success cases

Posterior flall success fulure damage cases

-

Posterior of B Mean= 119 COV 0101

Posterior Distribution or Blas Factor (8) bull Different Categories and All Cases Combined

shy

l

PHASE I - KEY CONCLUSIONS

bull GLOBAL CORRECTION FACTOR

bull 15 TO 20 CONSERVATISM MEAN)

bull UNEXPECTED SURVIVALS HAVE GREATEST IMPACT

bull DAMAGED CASES MORE INFORMATIVE

bull CALIBRATION ISSUES PHASE II

bull

EFFECT OF IMPROVED CAPACITY ANALYSIS RECIPE

bull APPLICABILITY TO MULTIPLE FAILURE MODES

bull NO FOUNDATION DAMAGE CASE INCLUDED

bull IMPROVED REPRESENTATION OF FULL ANDREW DISTRIBUTION

-bull NUMBER OF PLATFORMS CONSIDERED

I 1 I

I t ) bull ~

PHASE II WORK SCOPE (CALIBRATION)

bull

bull ESTABLISH MULTIPLE BIAS FACTORS

Use Nine Base Case Platforms

middot - Capacity Analysis for Revised recipe

Foundation Blas Factor from Foundation JIP

bull DEVELOP WEIGHTING APPROACH

bull UPDATE MULTIPLE BIAS FACTORS

bullEffect of Platforms Not In the Nine Base Cases

bull EXAMPLE APPLICATION OF BIAS FACTORS I

bullbull

PHASE I PHASE II

CAPACITYbull ANALYSISRu bull

---------------------1----middot

Jc1 - iJbullraquoJ dbull

FttlS frt[f~l ratdhJJ

-

LIKELIHOOD FUNCTION

DEVELOPMENT

lk Jb Jfall1N) bull Jb)

111 Ill Isurvtvoq bull 1 - bl

-middot -------------------shy ~---------------------middot-middot

r111ir8 Cb1 111 Cbl BAYESIAN

UPDATING

~cove

R1R2 bull middot---------------------~----

Jblbull P(g cOJ

1 bull JbRI (MBSI I bull 1

~middot

llcJllJfalluro) bull P(gcOJ Ill 111 b21fallun) bull P ((g1 CO) U 1112 c OD

middot--------------------shy ----------------------shy bull

r111111r8 Cbl 111 Ill i-1nro1

deg ~1b2)laquottfrr (b1b21-lnfo) re Ill bull 1-11middot b2) bull lodegamp

middot

i1 covamp i2 cov~i

-

shy

--

Likelihood Function vs b All Success Case STISI K STl30Q STl34W WD90A MC311 MC397

I 0000

0 9000

OIOOO

0 7000

~ 0amp000

ta osoOO

~ 0 4000

0 3000

0 2000

0 1000

bull 0 0000

bull

I

middot

middot bull I

Ibull

-sibull ~ -middot-middot-middot-middot -middot-

----7----+-----~ I

Ibull bull bull I

rbull

--- Jk ISIK[blsuccess)

- bull - bull - lk 130Q[blsuccess) bull

lk 134W[blsuccess)

Jk-WD90A(blsuccc~I

- - - lk MC397[blsucccss)

--ltgt-- lk middotMC31 l[blsucccss)

--- Jk [bl 6 out of6succcsHascs)

05 I 5 2 250

b

Likelihood Functionsmiddot Success Cases - Phbullsc L

shy

bullbull bull bull bull bull bull -------------------~

bull bull bullbull bullbull bullbull bullbull bullbull bullbull bull

bull bullbull

bullbullbullbull

bullbull

bullbullbullbullbullbullbull

bullbullbullbullbullbull

bullbullbullbullbullbullbull

z 0

t z ) LL

c 0 0c i ~ i

J

~ sect ~--------------~ e l5

I

I

SOFTWARE

bull PF PROGRAM

bull CALREL

bull PF +CALREL

bull RELACS

bull

gt i

l

SENSITIVITY STUDIES

POSTERIOR DISTRIBUTION OF B

bull EFFECT OF CHANGE IN DISTRIBUTION OF PRIOR OF B

bull EFFECT OF REMOVAL OF ONE PLATFORM FROM EACH CATEGORY

bull EFFECT OF REMOVAL OF THREE DAMAGE CASES

bull EFFECT OF ADDITIONAL 13 SIMILAR PLATFORMS

-I

Prior and Posterior Distributions of B Variation In COV of PriorofB fromOl to04 -- -

4500

-4000

-

- -

3500 --middot - bull bull - bull bull Prior-r ICOV =02

3000

- - - - Prior-r ICOV =03 -

bull -~ e 2soo -

bull bull bull bull bull bull bull Pnor-r ICOV =o 4t frJ ~ 2000 Posterior rlCOV =0 2

bull bull Postenor rlCOV =0 3 bull bull 1500 bull middot bull - shy -middot- Posterior rlCOV =04

middot~ I 000 - bull

-middotmiddot ~ ~middotmiddot M -~

- middot 0 500

wbullbullbull _ I middot~ middotmiddotmiddotmiddotmiddot middotmiddotmiddotmiddotmiddot - -~ ~ __ middot_ _ bull -shy0 000

0 000 0200 0 400 0 600 0 800 IOOO 1200 1400 1600 1800 2000 -

- - b

-

=shy

Posterior Distribution of Bias Factor (B) bull Effect of Variation In COV of Prior of B

bull

bullPrior and Posterior Distributions of B Evaluating Effect of Removal of One Platform from Each Category from the Base Case of 13 Platforms

400

~ ~ 350

j IA i

3 00 bull I bull I ---Prior-r I bull l

250 1 I 1

I bull bull ---Posterior r lall 13 casesshy- ~E 2 00 bull bull bull bull bull bull bull Posterior r without STISIK

- ~ 1 bull - bull I - bull bull - bull bull Posterior r Iwithout T2I I 50 I

J-tt~i - bull - bull - Posterior r Iwithout T2S

tOO [-

I I ~ ~ I 0 50

0 00

___ r J i

l~T

~--

0 00 0 20 0 40 0 60 080 I 00 120 140 160 110 2 00

b

zshy

Posterior Distribution of Bias Factor (8) bull Effect of Removal of middotPlatform from Any Category

bullPrior and Posterior Distributions or B Evaluating Effect or Removal or 3 Damage Cases from the Base Case or 13 Platrorms

4

35

3

25

e L 2 ~ bulle shy

I 15

bull I

05

0

0 02 04 06 I 2 I 4 16 IS 2

I I

I

I

I Ibull

J -tmiddot L--

bullI

~

I

fl

II ~

I I v I

middot bull bull

bull L

~ ll- I-~lt

OS I

b

I

bull

1middot bullbullbullbull

~ __

bull

- - - - Prior-r

- - - - Postenor flall failure cases

Postenor flall success cases

--- Posterior flall 13 cases

---Posterior fI success failure cases only

IPosterior or B Mean =t11 COV =0121

-

_I

Posterir Distribution or Bias Factor (B) bull Effect or Removal or 3-Damage Cases

-

bullPrior and Posterior Distributions or B Evaluating Effect or Additional 13 Similar Observations as

Used in Andrew JIP The POsterior or B from Andrew-JIP Considered as New Prior or B

- 500

450

400

350

bull bull

bull bull bull bull J bull

bull bull f ~ I

I ~ Ibull bull

bullbull bull

l bull I -bullbull bull bull

I

I bull bull

bull I I~~bull bull bull

bull ~

I bull bull

~ ~_ ~~~

- - New Prior r bull

Posterior 114 new failure cases ~ 300- bull bull bull bull bull bull bull Posterior f 13 new damage casesl 150 ~

- bull bull - bull bull Posterior 116 new success cases i 2 00

Postenor 1113 new success failure -150 damage cases

I 00

0 50 -

0 00 0 00 0 20 0 40 0 60 0 80 I 00 120 I 40 160 110 2 00

Posterior ol B Mean= 120 COV= 0071 b

shy

shy

_rshy

Posterior Distribution of Bias Factor (B) bull Effect of Additional 13-Similar Observations

1

1 I

WEIGHTING APPROACH bull

bull ASSESSMENT OF PLATFORMS DATA

Select Platforms Most Mappable to the Analyzed Platforms

Unlikely SurvivalDamage Cases Identification

Based on Cut-Off Criteria Agemiddot Hs etc

bull INFORMATION REQUIRED FOR COMPARISON

bull CATEGORIZATION OF ADDITIONALmiddot PLATFORMS

bull UPDATING OF CALIBRATION RESULTS

--

OVERVIEW OF PLATFORMS SURVIVEDDAMAGED -~

Operator -

- Platforms

Survived CS)

Platforms

Damaged CD)

Platforms Used In

Phase IJIP -

JI~ ~Bill~IRBDll -

Amoco -

Chevron -

Exxon

Mob II

Phllllps

Shell

Unocal -

30

143

72 22

7

39 25shy -

1

12

bull

1 bull

1

1(S)+1 (D)

3 (S) +3(D)

2 (S)

bull shy

bull

bull

bull

Total Partlclnants Platforms 338 - 15 10C6S+4D) -

Other Operators

310

13

3 (D)

Total Platforms 648 28 13C6S+7D) -

Summary of Andrew JIP bull Phase I Platforms

Plairshy w v- N-ber Lea-le 1 shy car Hind_ o-w N-bull Deplh fl fl IJpe la bull n Yeadon Jadtet lbelledtf liefalit -- _ -OD

IL H - -shy _ICllshy IL IL - - rQClto ~ _

K STUIK 137 1963 8 365 608S4 Cellardeck

-middot STl-shy 170 1964 4 -~ K 39 62269 Cellardeck - --ST134W 137 1911 4 43 608S4 Smvlnlshy- _ - -

WD90A 184 1964 8 K 267Sbull34 83 SOS31 SUboftllar deck SmvMI

MC311 343 1978 8 x S7 62 - MC397 468 1991 4 x shy S33bull6SO 65911 -shySmeSmvlvol -- gt -

r-- T23ST52 63 1969 4 ton- shyKshy 52594 S03 -T25SSl39 62 1969 4 Kshy SOS94 S063 - middot ton-shy__ -

bullbull a ton--shyST161A 118 1964 8 K - 347S 51495 Cellarclecti

r~

ST1771 142 1965 8 K 39 60167 CeDordeck failme

STISIH 137 1964 8 K 395 608S4 CeDardeck Failme

K ST130A 140 1958 a SS 6095 I lblv Pai1uno - T21ST72 61 1969 4 Kshy S4S94 4974 I lhlv Pai1uno -

- -

JE

INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON

bull WATER DEPTH

bull bull YEAR OF INSTALLATIONDESIGN

bull PLATFORM ORIENTATION

bull PLATFORM LATITUDE AND LONGITUDE

bull KEY SKETCHES OF PLATFORM STRUCTURAL FRAMING

Number of Legs Plies Leg Batter Number of Conductors bull Deck Elevations (Including sub-cellar deck If any) Jacket Framing Elevations Brace type In Vertical Frames Sizes of Primary Members (leg pile vertical frames braces primary horizontals) Joint Details (can gap etc) LegPiie Annulus Grouting Status

bull middotPOST ANDREW INSPECTION REPORT Significant Wave Height Hs (ft)

bull INFORMATION ON DAMAGES PRIOR TO ANDREW

bull RESULTS FROM IN-HOUSE STUDIES IF AVAILABLE

bull ~RIGINAL DESIGN CRITERIA INFORMATION IF AVAILABLE

bull SOIL SHEAR STRENGTH PROFILE IF AVAILABLE

Platforms Identified for Evaluating Therr Likelihood of Falling Under Unexpected Survival Category

OPlRATOR

middotADA

BLOClt

llmucnJllE

NAME

YEAll

INSTAUJD

WA1ER

DBPTll

DISTANCE

JlllOM SHOllE

(MILU)

Bl

(M)

AMOCO 111113

STl11

WDI075

A

bull D

St

17

120

1n

44

20

10 I

CllllVllQfl STlllO c 61 110 21 11 STIUO D Q 1110 21 11

STl134 p Q 137 29 11

ST0151 G 61 137 11

ST0151 I 3 121 11

ST0151 J Q 140 11

STl151 It 137 11

ST0151 PJl001 Q UT 11

ST0151 PROD-2 Q UT 11

ST017 A 3 140 35 11

WDIUT c 65 214 34 10

llXXON STI055 II 51 6T 14 I

STl15 A t3 34

WDI073

WD0073

Amiddotshy Q

1111 1111

22

22

I

I

WDI073 c 1n 22 I

- WDOCl93-shy II

65 1110 2T A 3 194 31 MOBU 111121

SS IOl7

A B

501

50

50

31

T

I

T 6

MtlllPllY SS114 B T 50 T T

PlllllIPS SS0149 A 5 55 33 T

SAMllDAN STln A 107 T T

SBllLL WDl103 A 65 223 2T WDl103 B 65 221 2T WD0104 c 65 221 2T WD0104 D 6T 20 2T

UNOCAL SS020I

SS 020I SSl20I

SS 0209

SS0209

II p

PURll

A

B

3

n 61

103

97

97

95

100

34

34

34

34 34

bullbull bullbullbull

- SS0209 D u 95 34 I

- SS 1209 SSl253

SS0253

G A

c

6T

2

100

165

1T5

34

54

54

bullbull I

I I

I I

SOUTH MARCH

~LAND

N

SOUTH PaTOI ENE ISWltD AREA

MMS Defined Limits ror Offshore Structures Requiring Inspections Following Andrew

t---Jbulln3 T25bull I bullTtl

bull1$1Htl1Abull bull1SIK1779bull

~

middot-DaTA AREA~T

CRANO rsLE

AREA bull1300

bullll4W

=shy

bullPath of Hurricane Andrew with Platforms Used In Calibration

Likelihood Function vs b All Success Case STISI K STIJOQ ST134W WD90A MCJll MCJ97

I 0000

0 9000

0 8000

07000

-Bo6000

tiiii 0 5000 -

04000

0 3000

0 2000

01000

0 0000

0 05 25

ri1rt middot-middot-middot-middot-middot-

I 5 2

b

---11 middot151K(blsu~ccsI

- bull - bull - fk 130Q(bfU~lCSS)

lie bull134W(blsuccc]

lk-WD90A[blsultCS]

- - - lk middotMC397(bfUtlCS]

--o-- lk-MCJll(bbuctcsJ

--- ll (bl 6 out of 6 Mlttcs case]

_

( Likelihood Functions middotSuccess Cases - p - 1 c I

CATEGORIZATION OF ADDITIONAL PLATFORMS

I I I

bull CATEGORIES

Unexpected Survivals

Expected Survivals

bull APPROACH

Identify Corresponding Platform(s) Evaluated In the JIP

Determine Max Wave Height at Platform Location

Evaluate Whether Wave Hit the Deck

middot - Evaluate Likelihood of R and S to Differ from the Base Cases

Review Joint Information

Assess Special Features

bull IDENTIFY NUMBER OF PLATFORMS IN EACH CATEGORY

I

UPDATING OF CALIBRATION RESULTS

bull INCLUDE EFFECT OF ADDITIONAL PLATFORMS ON bullBbull

Consider Increased 1Uncertalntles In R and S

Develop bullNeW Ukellhood Functions for Different Categories

middotLet bullxbull Number of Ukely Unexpected Survival Cases and y Number of Ukely Expected Survival Cases

Determine Combined Likelihood Function lk (b Iadditional platforms) middot

bull REVISED LIKELIHOOD FUNCTION

lk(b I all cases) = lk(b I 9 platforms) middot x lk(b Ix platforms) x lk(b I y platforms)

bull ESTABLISH REVISED BIAS FACTOR B

f8 (b) u f8 (b) lk (b I all cases)

II

t t ~

shy

APPLICATIONS OF BIAS FACTORS

ii APPLICABLE TO PARTICULAR LOAD AND RESISTANCE RECIPE

bull USE BY THE API COMMITTEES AND MMS

Criteria Revisions

Recommended Design Guidelines Revisions bull

bullbull REASSESSMENT STUDIES AND DECISION MAKING

middot

bull RISK ASSESSMENT STUDIES BY THE OPERATORS

Individual Platforms

Fleet of Platforms

middot IMR Program Implementation

bullbull NOT TO BE USED IN DESIGN

I I fl J

EXAMPLE USE OF BIAS FACTORS

bull PHASE I EXAMPLE APPLICATION

bull EVALUATION OF ANNUAL PROBABILITY OF FAILURE

Establish Annual Maximum Seastate Data

Determine Coefficients to Define Base Shear

Determine Ultimate Capacity of Platform

Determine Probablllty of Failure Versus bullbbull

-bull DETERMINE UPDATED PROBABILITY OF FAILURE

Pf - P [S gt R]

Pf I P [~ gt Rb I b] f 8 (b) db

Pr - J P [S gt Rb I b] f bull 8 (b) db

bull TOTAL PROBABILITIES OF FAILURE

Total Pf - 1 bull Il [1 middotPf] - l Pf bull

Total Pf - 1 bull Il [1middot Pf] - l Pfmiddot

Total Pr = 1 - n [1middot PfbullJ = l Pfshy

I I I

EXAMPLE APPLICATION bull PHASE I

PLATFORM ST 151 K

bull GULF OF MEXICO

bull WATER DEPTH = 137 FT

bull ANNUAL MAXIMUM HS MEDIAN 175 ft COV 031

bull ASSUMED STORM DURATION 3 hoursmiddot

bull ASSUMED NUMBER OF WAVES 800

bull MEDIAN HS FOR STORM HOURS 1 2 AND 3

bull MEDIAN CURRENT VALUE bull

bull DETERMINE ULTIMATE CAPACITY FOR THREE DIRECTIONS

bull DETERMINE ANNUAL FAILURE PROBABILITIES

Broadsidemiddot Diagonal End-on

0037 0025 0007

0071 0044 0018

0022 0016 0004

DETERMINE TOTAL ANNUAL PROBABILITY OF FAILURE

Pf - 007

Pf - 013

Pr - 004

bull

EXAMPLE USE OF BIAS FACTORS - PHASE II middot

bull DETERMINE PREDICTIVE PROBABILITY OF FAILURE

Case - 1

Pf _ S S Pf (b b2) fa1a2 (b b2) db db2

g bull -1 (b1 R I S) - 1

92 - (b2 R2 I S) - 1

Case -2

Pf s ~ [bJ f11 (b) db

Pf [bJ p [g lt 0)

g - bull 1 (b1 R1 I S) - 1

1

bull1) l

I

BENEFITS OF ANDREW JIP bull PHASE II

lt

bull IMPROVED LOAD AND RESISTANCE RECIPE

bull IMPROVED CALIBRATION PROCEDURE

bull MULTIPLE BIAS FACTORS

bull PROCEDURE TO INCLUDE EFFECT OF OTHER PLATFORMS

bull PROCEDURE TO USE MULTIPLE BIAS FACTORS

  • Structure Bookmarks
    • ANDREW PHASE II JOINT1 INDUSTRY PROJECT
    • PROGRESS MEETING NUMBER 1
    • ANDREW PHASE II JOINT INDUSTRY PROJECT PROGRESS MEETING NUMBER 1 NOVEMBER 16 AGENDA
    • ANDREW PHASE II UST OF PARTICIPANTS
    • ANDREW PHASE I OBJECTIVES AND RESULTS
    • ANDREW PHASE II PROJECT OBJECTIVES
    • SUBSEQUENT PROJECTS
    • ANDREW PHASE II PROJECT TEAM
    • ANDREW PHASE II UPDATE MODELING middotAND ANALYSIS RECIPE
    • -ANDREW PHASE II CONDUCTOR RESISTANCE
    • ANDREW PHASE Il WA VE FORCE PROFILE
    • ANDREW PHASE II WA VE FORCE PROFILEmiddot
    • ANDREW PHASE II BRACE MODELING
    • ANDREW PHASE II FOUNDATIONS
    • ANDREW PHASE II middot middot JOINT MODELING
    • ANDREW PHASE Il JOINT MODELING OPTIONS DISCUSSION
    • ANDREW PHASE II JOINT MODEJJNG OPTIONS ILLUSTRATIONS
    • ANDREW PHASE II JOINT MODELING PHASE II METHOD
    • ANDREW PHASE II ANDREW STORM CONDmONS UPDATED
    • Platform Selection bullbullbull1111111111111
    • BasisCriteria for Selection bullbullbull1111111111111
    • CALIBRATION
    • WHERE TO INTRODUCE BIAS FACTOR (B
    • PHASE I -KEY CONCLUSIONS
    • bull CALIBRATION ISSUES PHASE II
    • PHASE II WORK SCOPE (CALIBRATION)
    • SOFTWARE
    • SENSITIVITY STUDIES POSTERIOR DISTRIBUTION OF B
    • WEIGHTING APPROACH bull
    • INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON
    • CATEGORIZATION OF ADDITIONAL PLATFORMS
    • UPDATING OF CALIBRATION RESULTS
    • APPLICATIONS OF BIAS FACTORS
    • EXAMPLE USE OF BIAS FACTORS
    • EXAMPLE APPLICATION bull PHASE I PLATFORM ST 151 K
    • EXAMPLE USE OF BIAS FACTORS -PHASE II middot
    • BENEFITS OF ANDREW JIP bull PHASE II
Page 21:  · . ' ANDREW PHASE II PROJECT OBJECTIVES . ' • ·UPDATE MODELING AND ANALYSIS RECIPE· JOINTS ' ' ·BRACES WAVE PROFILE FOUNDATIONS . ' ' CONDUCTORS· • ANALVZE PLATFORMS USING

Project ST151K Model Wave_Loading Version 1 rue Nov 8 164107 1994 PT

E~ nss FT

SL 14amp FT

a 111 FT

EL 14 li

E-1 37 FT

El o FT

1middot1middot ______________ _ ________ __ _ __

0

200 400 600 800 1000 12000 Force at Elevatloii (kip)

200

a ll_

Wave Force at Major Elevations for Discrete Wave Heights

180

bullHmiddot40ft

Hmiddot495ft

H bull 51 ft

Hmiddot55ft_

H-59ft I

Andrew (H bull 60 86 ft) II

i I

_11---__

- middot- Hmiddot30ft

-

- bull -

middot bull bull bull middot

---Hmiddot53ft

bullbullbullbullHmiddot57ft

-----H-63ft

Elevation Above Mudllne 100

(ft)

180

140

120

80

60

40

20

Loading at Major Elevations

---H-49511

-deg-Hmiddot608611

0 200 400 600 800 1000 1200

Force It Elevatlon (kip)

--

Base Shear vs Wave Height ST151K

6000

5000 1---------Ult1mateCapacitybull 4890 kip (100 Improved Soll)-----------~--

4000a a lii ~ 3000

81 ampI

2000

1000

0 +-~~~--i~~~~~~~~~-+-~~~--lf--~~~~~~~~-+-~~~--l

30 35 40 45 50 55 60 65

Wave Height (ft)

bull middot

~

bull

II

2

g

~

It)

0

cCD E 0 E a c E euro~ CDshygtItgtmiddoto ~ti c -a Gic CD gt ~

-= c -Cll

I z

bullO 0i i ~ g 0

() CI -bull(q-ll) iuawow llu1wniiaAO

11

shy

116

114

112

g 110

bullc = 108

~ ~ 106 -sect 104

102

100

98

96

-

6555 60

Wave Height vs Center of Action ST151K

30 35 40 45 50

Wave Height (ft)

t

Comparison of Pushover Curves for ST 151K Variable Load Pattern and Monotonic Load Pattern (Andrew)

Soll Improved 100

--- --------- --- ----6000

5000

- 4000 8 - I

1 3000 Ul

81 m 2000

_J -shy

1000

---Variable Load

- - Andrew Load

0 f-~~~~~r--~~~~-t-~~~~~-t-~~~~~+-~~~~----t~~~~~-1

0 1 - 2 3 4 5 6

Deck Displacement (ft)

-middotshy

1 t I I ~

ANDREW PHASE II

BRACE MODELING

bull MARSHALL STRUTS USED FORPHASE I

bull PHASE II WILL ASSESS DIFFERENCE IN USING BUCKLING BEAMS AND MARSHALL STRUTS

middot bull COMPARE FOR 2D FRAME

bull COMPARE FOR PLATFORM 177 WHERE LOWER

BAY EXHIBITS LARGE BRACE MOMENTS

Project andrew Model beamframe Version 1 Mon Nov 14 145447 1994

~

middot Frame Test buckling beam-column model

-

Project andrew Model strutframe Version 1 Mon Nov 14 145754 1994

bull

Frame Test strut model

middot-middot _____________--_

_middotmiddotr_~ --ll--~-middot__

Buckling Beam bull

Strut (K=O 65)

Lateral Displacement

ANDREW PHASE II

FOUNDATIONS

bull ~LICIT NONLINEAR API FOUNDATION USED INiPHASE I

bull FOUNDATIONS

JIP SELECTED MINI-VANE DATA FOR

STRENGTH PROFILE

bull COMMENTS ON BENCHMARK COMPARISONS

- SIGNIFICANT VARIABILITY DUE TO FOUNDATION MODELING METHODS

II Ii

z

ctV 4x SPlO caissonfy-42s soil-api(lf-097step-114)

Inelastic Events Legend Elastic Strut Buckling

Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield

--------shyBeam Clmn Fully Plastic Fracture

~--~~~~~~~~~--~----~__i~~---~~~~__--~

Pro)ect apipile Model South Pelto 10 Version Wed May 18072235 1994 - -

1bullt

+

t

f i bullbullbull

bullbull bullbull

I I

5

~x SPlO Caissonfy=42s soil=api(lf=lOOstep=137)

Inelastic Events Legend

Elastic Strut Bucklilg

Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield

earn Clmn Full~ Plas1c Fracture

ANDREW PHASE II middot middot

JOINT MODELING

middotbull API RIGID-PLASTIC USED IN PHASE I

a ELASTIC-BRITTLE USED IN PHASE IB

amiddot MSL WILL PROVIDE NONLINEAR FORCE-DEFORMATION

AND MOMENT-DEFORMATION RELATIONS FOR PHASE II

a MODELING OPTIONS FOR PHASE II -

l

- UNCO~LEDmiddotONE DIMENSIONAL MEMBER END JOINT MODELING

J Ii

ANDREW PHASE Il

JOINT MODELING OPTIONS DISCUSSION

bull JD COUPLED INTERACTION SURFACE

- IDEAL

- DATA NOT AVAILABLE TO CALIBRATE

bull 2D COUPLED INTERACTION SURFACE

- SIGNIFICANT MODELING IMPROVEMENT

- DATAAVAILABLE

- IMPLEMENTATION PRACTICAL BUT NEEDS DEVELOPMENT AND CALIBRATION

bull 2D UNCOUPLED INTERACTION SURFACE

- SIGNIFICANT MODELING IMPROVEMENT

- DATA AVAILABLE

- IMPLEMENTATION PRACTICAL AND AVAILABLE

ANDREW PHASE II

JOINT MODEJJNG OPTIONS ILLUSTRATIONS

P

e

ANDREW PHASE II

JOINT MODELING PHASE II METHOD

e

I I I

ANDREW PHASE II

ANDREW STORM CONDmONS UPDATED

bull -PHASE I USED OCEANWEATHER HINDCAST-(Nov 1992)

bull middotmiddotPHASE II WILL USE THE REVISED HINDCAST (Nov 1994)

bull lt c lt I lt

2-2

l---4-~~~+-~~-+-~~~1--~~-+-~~--l--~T--+-~~--i24

f-tJVtMfiz~ 1Cfl (r STvb(

Plollol bullbull 5middotNOYmiddot9214bull1J Irbullbull Illbull [ANDREVl~NDREVCUSTbullMAXll 5-NOY-1992 10bull54 - 9208

HEIGHT lml

I~ MAXI

I L 1~middotamp SZ~- ~A I 13c

I J) ( r I X~0iWJ~0gtJ gt L gt gt I 1c I 120

-- N

~ deg

I I I I I I e C I ~=-41 122 -94 -92 -90 -88 -86 -84 -82 -80

-

Figure 7 contours of maximum hindcast significant wave height

_

~

~tl ~

NOltNttIJ 17lt ltNS S iJJ)y

Maximum Significant Wave Height - Hurricane Andrew 1992

30 ~

26 ~ I ~-~~~- == -shy -~ 35-_

24deg1--shy --------~- middot-middot-------middotmiddotmiddot----middot------middot-middot~middot- - - ~-----1---middot

22middot ~~~~~~~-~ -94 -92 -90 -88 -84deg-86 -82 -somiddot

c

-

Platform Selection bullbullbull1111111111111

0 BasisCriteria for Selection middot

0 Review of Phase I Platforms middot 0 Alternative Platforms

0 PMB Recommendations

0 Participants Comments

gt

L

l I

BasisCriteria for Selection bullbullbull1111111111111

0 Benefit for Calibration - Classification (UnexpectedExpected)

- Failure Mode (StructuresFoundations)

- Damaged StructuresKnown response

- Better Representation of Full ~opulation

0 Availability of Data - Existing Models

- Inspection Data

- Damage Data

PREDICTED BEHAVIOR

-

OBSERVED

BEHAVIOR

-SURVIVED DAMAGED FAILED

SURVIVED

ST134W WD90A

-MC311 -

MC397 -

--

-

-

ST151K ST130Q

-

-

-

DAMAGED - -

-

-

--

-

-

ST52 SS139 ST177B -

ST161A

-_

-

FAILED

-

-

-

ST151H ST130A ST72

Operator - Platfonns - Platfonns Survived Damagedamp

- Falled -

bull

-

-

Jle eBDIGIPllDbl -Amocoshy 30 1 Chevron -

143 12 Exxon 72 Mobll 22 1 Phllllps 7 bull

-Shell 39 Unocal 25 1

Total Partlclnts Platfonns 338 15

Other Operators bull- 310 13

-

Total Platfonns - -

648 28

fbml

Platforms Considered 6 7

tie of Total Platfonns 1 25

OVERVIEW OF PLATFORMS SURVIVEDDAMAGEDFAILED

-

shy

v

Platforms Evaluated in Phase I

-

-Platfonn Water Year Number Analysis Avallabillty of Availability Platform

- Name Depth rl of Model DetaDed rl ClassUlcatlon

Installation legs Inspection Soll Report Based on Information Phase I Calibration - -ft

- -- Suaill flalfnrm ~BSi -STISIK 137 1963 8 CAP Yes Yes Un eel Sarvlval

-STIJOO 170 1964 4 CAPmiddotshy Yes ST134data u ST134W 137 1981 4 CAP Yes Yesmiddot E

-

edSarvlval -

middotsurvival

WD90A 184 - 1964 8 CAP Yes E -

- Sarvlvill

MC3ll 343 1978 8 KARMA

MC397 468 1991 4 KARMA -

bull ~reglmmiddot- -

Yes Sure Survival

bull Yes - Sare Sanlval-

-T23STS2 63 1969 4 CAP

T2SSS139 62 1969 4 CAP -

Yes

Yes

Yes Ex

Yes Une

bull SurvivalLikely to DamalI

middot bull SanlvalLlkely to DamalI

ST161A 118 1964 8 USFOS

Eail11a pound1al(bullllm Clss ST177B 142 --1965 8 CAP

STISIH 137 1964 - 8 CAP C

STIJOA 140 19S8 8 CAP - -

Yes

Yes

ollamed

Yes

- Yes E

-

at about 1 mUe

Yes

ST134data

bull SafvlvalLlkelv to Dam- bull- -

E middotrahre

Emeried Failure -

Likely FaUare

T21ST72 61 1969 4 CAP Collamed No Likely FaUure

Platforms Identified for Evaluating Their Likelihood of Falling Under UnexpeCted Survival Category

CllEllATOa

AllB

llLOCll

Sl1lvcnJllE

NAME

YEAR

INSTAUJCD

WATER

IJBPJB

middotmiddotDISTANCE

ftOMSllOllE

(MIUIS)

Ill

(M)

AMOCO BllW

STllI

WDI075

A

bull D

4

n uo 17J

44

32

bull 20

10

bull CllEVllOH STtllO c I 1bull 21 11

STl130 D 0 161 21 11 STllJ4 I 0 137 D - 11

STIUI G I 137 32 11

STIUI I 0 121 32 11

STIUI

J 0 140 32 11

STIUI It 4 137 32 11

STllSI PROl)I 0 137 32 11

STllSI PROD-2 0 137 32 11

STll7 A 0

140 35 11

WDl117 c 65 214 34 10

DXON STshy E 51 14 bull STl165 A t3 bull 34 WD0073

WD0073

WD0073

A _ c

4

62

4

IA IA

17J

22

22

22

bullbullbull WD003-shy E 65 161 Z7

A 0 IN 31 MOBD Bl12I

ssn A bull

50

50

31

7

bull 7

MIJllPllY SS114 B 7 50 7 7

PlllUIPS SSl14t A 55 33 7

SAMEDAN ST17J A 4 117 7 bull 7

SJIELL WD0103 A 65 223 27 WDllOJ bull 65 221 27 WDll04 c 65 221 27 WDOI04 D 67 2 27

UNOCAL

SSl20I

SSl20I

bull SSl20I

SS0209

SS0209

SS0209

SS0209

SS 0253

SS 0253

E

I

PtJllE

A

bull D

G A

c

0

4

4

7J

61

0

0

69

103

97

97

9S

100

95

100

165

175

34

34

34

34

34

34

34

54

54

bullbullbullbullbullbullbullbullbull

Platforms Recommended for Phase II

llalr y_ w- Aot Av- _ I 1 Doplla Model nm bull bull _ bull I I H11 s18-1 Irht lbwl~ bullbullbull(-illtdlel OM lllMtD

II ---

-STISIIC 137 1963 I CAP Yeo Yeo I Yeo

STtshy -170 1964 4 CAP Yeo ST134dlU I Yeo

STl34W 137 Yeo Yeo -1981 4 CAP D Yeo

WD90A 114 1964 I CAP Yeo - D

MC311 30 1971 I KARMA Yeo -sre-

MC397 middot 461 1991 4 ICARMA Yeo Sireshy - - r- -

__ T13ST5l 63 1969 4 CAP Yeo Yeobull middot- ~n- D Yeo _ Tl5SSl39 62 1969 4 CAP Yeo Yeo I Yeo

_ STIT711 142 1965 I CAP Yeo I Yeo middot-1shySTl61A Ill 1964 I USFOS Yeo Yeo - D bulloshy

- r-

STISIH 137 1964 I CAP y D y -_ STl31A 140 1951 I CAP Yeo STl34dlU I Yeo

T21Sll 61 1969 4 CAP No rar-middot I Yeo

~-rmiddot-middot SS209A 95 1967 16 CAP Yeo D -SSU4R

11 UI A 41 -in D SS 2741AIM CAP bull

_ _ -shy - -shy- -shy

CALIBRATION

bull PHASEIAPPROACH

bull CALIBRATION ISSUES

bull PHASE II WORK SCOPE

bull SOFlWARE AVAILABILITY

bull EFFECT OF ADDITIONAL PLATFORMS

bull WEIGHTING APPROACH

bull EXAMPLE APPLICATION middot

(l=Ofce Mode] Wind

~

fcurrent l -middot

~SW ~leldl4

ultimate resistance force I first

design load Illgt

i _

deflection

-

bull

4

Historical Hurrlcane Dariiage and Survival of Platforms

Components Required for Calibration (Petrauskas 1992)

PRIOR Distribution

PLATFORM EXPERIENCE

BAYESIAN Survivals and Failures

UPDATING

Waves Winds and Currents bull

POSTERIOR Distribution

- A_middotJA ~ Survival Data Failure Data

Result Result

Load Resistance

middotfl

-

Bayesian Updaing middotGeneral Description (Petrauskas 1992)

L Imiddot

WHERE TO INTRODUCE BIAS FACTOR (B

On Resistance (Rmue - B (RPREDICreD I

middotOn Loads (Smue middot - B (SPRE01cTEo

On Safety Margin (RSmue - B (RSPREDICTED

CALIBRATION METHODOLOGY

gt

r

FIELD OBSERVATIONS

Observed Behavior

survivals Damages

Failures

r

bull

CAPACITY ANALYSIS

Predicted Load and Resistance Predicted Failure Modes

Survival Damage Failure cases

RELIABILITY ANALYSIS

- Probability of Failure ~

Ukellhood Failure

BAYESIAN UPDATING

Prior Distnbution of B Combined Likelihood Function

Posterior Dlstnbution of B

BIAS FACTOR

Mean Value cov bull

Represents Modeling Uncertainties

Prior Distribution of B Likelihood Functionmiddot Observing Success case STISIK and Posterior Distribution of B

2 00

uo

160

140

120

e 100 L $ uoIshy

0 60 ~ ~ 040

020

0 00

0 00 0 50

_ I 00

b

I 50 2 00

-shy lk [bl STISIK)

--shy Pnor-f(b)

--shy Pos1enor r(bl ISIK)

Prior or B Mean bull 10 COV bull 030

Posterlor of Bbull Meanbull 136COV 0 16

2 50

Likelihood Function and Posterior Distribution Example Platform STISIK

- -

~

- Prior and Posterior Distributions or B Failure cases (ST177B STlSlH ST130A T21) Damage cases (T23 T25 ST161A) Success cases (STlSlK ST130Q ST134W WD90A MC311 MC397)

400

350

300

2 so 0-

I

bull bull

bull bull gt

- bullbull ~ bull bull I bull

_ bull bull bullbull I bull bull bullII V -

bull bull

I bull ~ bull

bullbull bullI bull bullI bull bull bull

middot -~

bull__ ~ middotI middotmiddot

L 2 00

e ~ I SO

-I 00

- 0 so

000 gt

0 00 020 0 40 060 0 80 I 00 120 140 160 110 200

b -

-

shy

Prior-r

- - - - Posterior fLlll failure cases

bull bull bull bull bull bull bull Posterior flall damage cases

- bull bull - bull middot Postenor flall success cases

Posterior flall success fulure damage cases

-

Posterior of B Mean= 119 COV 0101

Posterior Distribution or Blas Factor (8) bull Different Categories and All Cases Combined

shy

l

PHASE I - KEY CONCLUSIONS

bull GLOBAL CORRECTION FACTOR

bull 15 TO 20 CONSERVATISM MEAN)

bull UNEXPECTED SURVIVALS HAVE GREATEST IMPACT

bull DAMAGED CASES MORE INFORMATIVE

bull CALIBRATION ISSUES PHASE II

bull

EFFECT OF IMPROVED CAPACITY ANALYSIS RECIPE

bull APPLICABILITY TO MULTIPLE FAILURE MODES

bull NO FOUNDATION DAMAGE CASE INCLUDED

bull IMPROVED REPRESENTATION OF FULL ANDREW DISTRIBUTION

-bull NUMBER OF PLATFORMS CONSIDERED

I 1 I

I t ) bull ~

PHASE II WORK SCOPE (CALIBRATION)

bull

bull ESTABLISH MULTIPLE BIAS FACTORS

Use Nine Base Case Platforms

middot - Capacity Analysis for Revised recipe

Foundation Blas Factor from Foundation JIP

bull DEVELOP WEIGHTING APPROACH

bull UPDATE MULTIPLE BIAS FACTORS

bullEffect of Platforms Not In the Nine Base Cases

bull EXAMPLE APPLICATION OF BIAS FACTORS I

bullbull

PHASE I PHASE II

CAPACITYbull ANALYSISRu bull

---------------------1----middot

Jc1 - iJbullraquoJ dbull

FttlS frt[f~l ratdhJJ

-

LIKELIHOOD FUNCTION

DEVELOPMENT

lk Jb Jfall1N) bull Jb)

111 Ill Isurvtvoq bull 1 - bl

-middot -------------------shy ~---------------------middot-middot

r111ir8 Cb1 111 Cbl BAYESIAN

UPDATING

~cove

R1R2 bull middot---------------------~----

Jblbull P(g cOJ

1 bull JbRI (MBSI I bull 1

~middot

llcJllJfalluro) bull P(gcOJ Ill 111 b21fallun) bull P ((g1 CO) U 1112 c OD

middot--------------------shy ----------------------shy bull

r111111r8 Cbl 111 Ill i-1nro1

deg ~1b2)laquottfrr (b1b21-lnfo) re Ill bull 1-11middot b2) bull lodegamp

middot

i1 covamp i2 cov~i

-

shy

--

Likelihood Function vs b All Success Case STISI K STl30Q STl34W WD90A MC311 MC397

I 0000

0 9000

OIOOO

0 7000

~ 0amp000

ta osoOO

~ 0 4000

0 3000

0 2000

0 1000

bull 0 0000

bull

I

middot

middot bull I

Ibull

-sibull ~ -middot-middot-middot-middot -middot-

----7----+-----~ I

Ibull bull bull I

rbull

--- Jk ISIK[blsuccess)

- bull - bull - lk 130Q[blsuccess) bull

lk 134W[blsuccess)

Jk-WD90A(blsuccc~I

- - - lk MC397[blsucccss)

--ltgt-- lk middotMC31 l[blsucccss)

--- Jk [bl 6 out of6succcsHascs)

05 I 5 2 250

b

Likelihood Functionsmiddot Success Cases - Phbullsc L

shy

bullbull bull bull bull bull bull -------------------~

bull bull bullbull bullbull bullbull bullbull bullbull bullbull bull

bull bullbull

bullbullbullbull

bullbull

bullbullbullbullbullbullbull

bullbullbullbullbullbull

bullbullbullbullbullbullbull

z 0

t z ) LL

c 0 0c i ~ i

J

~ sect ~--------------~ e l5

I

I

SOFTWARE

bull PF PROGRAM

bull CALREL

bull PF +CALREL

bull RELACS

bull

gt i

l

SENSITIVITY STUDIES

POSTERIOR DISTRIBUTION OF B

bull EFFECT OF CHANGE IN DISTRIBUTION OF PRIOR OF B

bull EFFECT OF REMOVAL OF ONE PLATFORM FROM EACH CATEGORY

bull EFFECT OF REMOVAL OF THREE DAMAGE CASES

bull EFFECT OF ADDITIONAL 13 SIMILAR PLATFORMS

-I

Prior and Posterior Distributions of B Variation In COV of PriorofB fromOl to04 -- -

4500

-4000

-

- -

3500 --middot - bull bull - bull bull Prior-r ICOV =02

3000

- - - - Prior-r ICOV =03 -

bull -~ e 2soo -

bull bull bull bull bull bull bull Pnor-r ICOV =o 4t frJ ~ 2000 Posterior rlCOV =0 2

bull bull Postenor rlCOV =0 3 bull bull 1500 bull middot bull - shy -middot- Posterior rlCOV =04

middot~ I 000 - bull

-middotmiddot ~ ~middotmiddot M -~

- middot 0 500

wbullbullbull _ I middot~ middotmiddotmiddotmiddotmiddot middotmiddotmiddotmiddotmiddot - -~ ~ __ middot_ _ bull -shy0 000

0 000 0200 0 400 0 600 0 800 IOOO 1200 1400 1600 1800 2000 -

- - b

-

=shy

Posterior Distribution of Bias Factor (B) bull Effect of Variation In COV of Prior of B

bull

bullPrior and Posterior Distributions of B Evaluating Effect of Removal of One Platform from Each Category from the Base Case of 13 Platforms

400

~ ~ 350

j IA i

3 00 bull I bull I ---Prior-r I bull l

250 1 I 1

I bull bull ---Posterior r lall 13 casesshy- ~E 2 00 bull bull bull bull bull bull bull Posterior r without STISIK

- ~ 1 bull - bull I - bull bull - bull bull Posterior r Iwithout T2I I 50 I

J-tt~i - bull - bull - Posterior r Iwithout T2S

tOO [-

I I ~ ~ I 0 50

0 00

___ r J i

l~T

~--

0 00 0 20 0 40 0 60 080 I 00 120 140 160 110 2 00

b

zshy

Posterior Distribution of Bias Factor (8) bull Effect of Removal of middotPlatform from Any Category

bullPrior and Posterior Distributions or B Evaluating Effect or Removal or 3 Damage Cases from the Base Case or 13 Platrorms

4

35

3

25

e L 2 ~ bulle shy

I 15

bull I

05

0

0 02 04 06 I 2 I 4 16 IS 2

I I

I

I

I Ibull

J -tmiddot L--

bullI

~

I

fl

II ~

I I v I

middot bull bull

bull L

~ ll- I-~lt

OS I

b

I

bull

1middot bullbullbullbull

~ __

bull

- - - - Prior-r

- - - - Postenor flall failure cases

Postenor flall success cases

--- Posterior flall 13 cases

---Posterior fI success failure cases only

IPosterior or B Mean =t11 COV =0121

-

_I

Posterir Distribution or Bias Factor (B) bull Effect or Removal or 3-Damage Cases

-

bullPrior and Posterior Distributions or B Evaluating Effect or Additional 13 Similar Observations as

Used in Andrew JIP The POsterior or B from Andrew-JIP Considered as New Prior or B

- 500

450

400

350

bull bull

bull bull bull bull J bull

bull bull f ~ I

I ~ Ibull bull

bullbull bull

l bull I -bullbull bull bull

I

I bull bull

bull I I~~bull bull bull

bull ~

I bull bull

~ ~_ ~~~

- - New Prior r bull

Posterior 114 new failure cases ~ 300- bull bull bull bull bull bull bull Posterior f 13 new damage casesl 150 ~

- bull bull - bull bull Posterior 116 new success cases i 2 00

Postenor 1113 new success failure -150 damage cases

I 00

0 50 -

0 00 0 00 0 20 0 40 0 60 0 80 I 00 120 I 40 160 110 2 00

Posterior ol B Mean= 120 COV= 0071 b

shy

shy

_rshy

Posterior Distribution of Bias Factor (B) bull Effect of Additional 13-Similar Observations

1

1 I

WEIGHTING APPROACH bull

bull ASSESSMENT OF PLATFORMS DATA

Select Platforms Most Mappable to the Analyzed Platforms

Unlikely SurvivalDamage Cases Identification

Based on Cut-Off Criteria Agemiddot Hs etc

bull INFORMATION REQUIRED FOR COMPARISON

bull CATEGORIZATION OF ADDITIONALmiddot PLATFORMS

bull UPDATING OF CALIBRATION RESULTS

--

OVERVIEW OF PLATFORMS SURVIVEDDAMAGED -~

Operator -

- Platforms

Survived CS)

Platforms

Damaged CD)

Platforms Used In

Phase IJIP -

JI~ ~Bill~IRBDll -

Amoco -

Chevron -

Exxon

Mob II

Phllllps

Shell

Unocal -

30

143

72 22

7

39 25shy -

1

12

bull

1 bull

1

1(S)+1 (D)

3 (S) +3(D)

2 (S)

bull shy

bull

bull

bull

Total Partlclnants Platforms 338 - 15 10C6S+4D) -

Other Operators

310

13

3 (D)

Total Platforms 648 28 13C6S+7D) -

Summary of Andrew JIP bull Phase I Platforms

Plairshy w v- N-ber Lea-le 1 shy car Hind_ o-w N-bull Deplh fl fl IJpe la bull n Yeadon Jadtet lbelledtf liefalit -- _ -OD

IL H - -shy _ICllshy IL IL - - rQClto ~ _

K STUIK 137 1963 8 365 608S4 Cellardeck

-middot STl-shy 170 1964 4 -~ K 39 62269 Cellardeck - --ST134W 137 1911 4 43 608S4 Smvlnlshy- _ - -

WD90A 184 1964 8 K 267Sbull34 83 SOS31 SUboftllar deck SmvMI

MC311 343 1978 8 x S7 62 - MC397 468 1991 4 x shy S33bull6SO 65911 -shySmeSmvlvol -- gt -

r-- T23ST52 63 1969 4 ton- shyKshy 52594 S03 -T25SSl39 62 1969 4 Kshy SOS94 S063 - middot ton-shy__ -

bullbull a ton--shyST161A 118 1964 8 K - 347S 51495 Cellarclecti

r~

ST1771 142 1965 8 K 39 60167 CeDordeck failme

STISIH 137 1964 8 K 395 608S4 CeDardeck Failme

K ST130A 140 1958 a SS 6095 I lblv Pai1uno - T21ST72 61 1969 4 Kshy S4S94 4974 I lhlv Pai1uno -

- -

JE

INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON

bull WATER DEPTH

bull bull YEAR OF INSTALLATIONDESIGN

bull PLATFORM ORIENTATION

bull PLATFORM LATITUDE AND LONGITUDE

bull KEY SKETCHES OF PLATFORM STRUCTURAL FRAMING

Number of Legs Plies Leg Batter Number of Conductors bull Deck Elevations (Including sub-cellar deck If any) Jacket Framing Elevations Brace type In Vertical Frames Sizes of Primary Members (leg pile vertical frames braces primary horizontals) Joint Details (can gap etc) LegPiie Annulus Grouting Status

bull middotPOST ANDREW INSPECTION REPORT Significant Wave Height Hs (ft)

bull INFORMATION ON DAMAGES PRIOR TO ANDREW

bull RESULTS FROM IN-HOUSE STUDIES IF AVAILABLE

bull ~RIGINAL DESIGN CRITERIA INFORMATION IF AVAILABLE

bull SOIL SHEAR STRENGTH PROFILE IF AVAILABLE

Platforms Identified for Evaluating Therr Likelihood of Falling Under Unexpected Survival Category

OPlRATOR

middotADA

BLOClt

llmucnJllE

NAME

YEAll

INSTAUJD

WA1ER

DBPTll

DISTANCE

JlllOM SHOllE

(MILU)

Bl

(M)

AMOCO 111113

STl11

WDI075

A

bull D

St

17

120

1n

44

20

10 I

CllllVllQfl STlllO c 61 110 21 11 STIUO D Q 1110 21 11

STl134 p Q 137 29 11

ST0151 G 61 137 11

ST0151 I 3 121 11

ST0151 J Q 140 11

STl151 It 137 11

ST0151 PJl001 Q UT 11

ST0151 PROD-2 Q UT 11

ST017 A 3 140 35 11

WDIUT c 65 214 34 10

llXXON STI055 II 51 6T 14 I

STl15 A t3 34

WDI073

WD0073

Amiddotshy Q

1111 1111

22

22

I

I

WDI073 c 1n 22 I

- WDOCl93-shy II

65 1110 2T A 3 194 31 MOBU 111121

SS IOl7

A B

501

50

50

31

T

I

T 6

MtlllPllY SS114 B T 50 T T

PlllllIPS SS0149 A 5 55 33 T

SAMllDAN STln A 107 T T

SBllLL WDl103 A 65 223 2T WDl103 B 65 221 2T WD0104 c 65 221 2T WD0104 D 6T 20 2T

UNOCAL SS020I

SS 020I SSl20I

SS 0209

SS0209

II p

PURll

A

B

3

n 61

103

97

97

95

100

34

34

34

34 34

bullbull bullbullbull

- SS0209 D u 95 34 I

- SS 1209 SSl253

SS0253

G A

c

6T

2

100

165

1T5

34

54

54

bullbull I

I I

I I

SOUTH MARCH

~LAND

N

SOUTH PaTOI ENE ISWltD AREA

MMS Defined Limits ror Offshore Structures Requiring Inspections Following Andrew

t---Jbulln3 T25bull I bullTtl

bull1$1Htl1Abull bull1SIK1779bull

~

middot-DaTA AREA~T

CRANO rsLE

AREA bull1300

bullll4W

=shy

bullPath of Hurricane Andrew with Platforms Used In Calibration

Likelihood Function vs b All Success Case STISI K STIJOQ ST134W WD90A MCJll MCJ97

I 0000

0 9000

0 8000

07000

-Bo6000

tiiii 0 5000 -

04000

0 3000

0 2000

01000

0 0000

0 05 25

ri1rt middot-middot-middot-middot-middot-

I 5 2

b

---11 middot151K(blsu~ccsI

- bull - bull - fk 130Q(bfU~lCSS)

lie bull134W(blsuccc]

lk-WD90A[blsultCS]

- - - lk middotMC397(bfUtlCS]

--o-- lk-MCJll(bbuctcsJ

--- ll (bl 6 out of 6 Mlttcs case]

_

( Likelihood Functions middotSuccess Cases - p - 1 c I

CATEGORIZATION OF ADDITIONAL PLATFORMS

I I I

bull CATEGORIES

Unexpected Survivals

Expected Survivals

bull APPROACH

Identify Corresponding Platform(s) Evaluated In the JIP

Determine Max Wave Height at Platform Location

Evaluate Whether Wave Hit the Deck

middot - Evaluate Likelihood of R and S to Differ from the Base Cases

Review Joint Information

Assess Special Features

bull IDENTIFY NUMBER OF PLATFORMS IN EACH CATEGORY

I

UPDATING OF CALIBRATION RESULTS

bull INCLUDE EFFECT OF ADDITIONAL PLATFORMS ON bullBbull

Consider Increased 1Uncertalntles In R and S

Develop bullNeW Ukellhood Functions for Different Categories

middotLet bullxbull Number of Ukely Unexpected Survival Cases and y Number of Ukely Expected Survival Cases

Determine Combined Likelihood Function lk (b Iadditional platforms) middot

bull REVISED LIKELIHOOD FUNCTION

lk(b I all cases) = lk(b I 9 platforms) middot x lk(b Ix platforms) x lk(b I y platforms)

bull ESTABLISH REVISED BIAS FACTOR B

f8 (b) u f8 (b) lk (b I all cases)

II

t t ~

shy

APPLICATIONS OF BIAS FACTORS

ii APPLICABLE TO PARTICULAR LOAD AND RESISTANCE RECIPE

bull USE BY THE API COMMITTEES AND MMS

Criteria Revisions

Recommended Design Guidelines Revisions bull

bullbull REASSESSMENT STUDIES AND DECISION MAKING

middot

bull RISK ASSESSMENT STUDIES BY THE OPERATORS

Individual Platforms

Fleet of Platforms

middot IMR Program Implementation

bullbull NOT TO BE USED IN DESIGN

I I fl J

EXAMPLE USE OF BIAS FACTORS

bull PHASE I EXAMPLE APPLICATION

bull EVALUATION OF ANNUAL PROBABILITY OF FAILURE

Establish Annual Maximum Seastate Data

Determine Coefficients to Define Base Shear

Determine Ultimate Capacity of Platform

Determine Probablllty of Failure Versus bullbbull

-bull DETERMINE UPDATED PROBABILITY OF FAILURE

Pf - P [S gt R]

Pf I P [~ gt Rb I b] f 8 (b) db

Pr - J P [S gt Rb I b] f bull 8 (b) db

bull TOTAL PROBABILITIES OF FAILURE

Total Pf - 1 bull Il [1 middotPf] - l Pf bull

Total Pf - 1 bull Il [1middot Pf] - l Pfmiddot

Total Pr = 1 - n [1middot PfbullJ = l Pfshy

I I I

EXAMPLE APPLICATION bull PHASE I

PLATFORM ST 151 K

bull GULF OF MEXICO

bull WATER DEPTH = 137 FT

bull ANNUAL MAXIMUM HS MEDIAN 175 ft COV 031

bull ASSUMED STORM DURATION 3 hoursmiddot

bull ASSUMED NUMBER OF WAVES 800

bull MEDIAN HS FOR STORM HOURS 1 2 AND 3

bull MEDIAN CURRENT VALUE bull

bull DETERMINE ULTIMATE CAPACITY FOR THREE DIRECTIONS

bull DETERMINE ANNUAL FAILURE PROBABILITIES

Broadsidemiddot Diagonal End-on

0037 0025 0007

0071 0044 0018

0022 0016 0004

DETERMINE TOTAL ANNUAL PROBABILITY OF FAILURE

Pf - 007

Pf - 013

Pr - 004

bull

EXAMPLE USE OF BIAS FACTORS - PHASE II middot

bull DETERMINE PREDICTIVE PROBABILITY OF FAILURE

Case - 1

Pf _ S S Pf (b b2) fa1a2 (b b2) db db2

g bull -1 (b1 R I S) - 1

92 - (b2 R2 I S) - 1

Case -2

Pf s ~ [bJ f11 (b) db

Pf [bJ p [g lt 0)

g - bull 1 (b1 R1 I S) - 1

1

bull1) l

I

BENEFITS OF ANDREW JIP bull PHASE II

lt

bull IMPROVED LOAD AND RESISTANCE RECIPE

bull IMPROVED CALIBRATION PROCEDURE

bull MULTIPLE BIAS FACTORS

bull PROCEDURE TO INCLUDE EFFECT OF OTHER PLATFORMS

bull PROCEDURE TO USE MULTIPLE BIAS FACTORS

  • Structure Bookmarks
    • ANDREW PHASE II JOINT1 INDUSTRY PROJECT
    • PROGRESS MEETING NUMBER 1
    • ANDREW PHASE II JOINT INDUSTRY PROJECT PROGRESS MEETING NUMBER 1 NOVEMBER 16 AGENDA
    • ANDREW PHASE II UST OF PARTICIPANTS
    • ANDREW PHASE I OBJECTIVES AND RESULTS
    • ANDREW PHASE II PROJECT OBJECTIVES
    • SUBSEQUENT PROJECTS
    • ANDREW PHASE II PROJECT TEAM
    • ANDREW PHASE II UPDATE MODELING middotAND ANALYSIS RECIPE
    • -ANDREW PHASE II CONDUCTOR RESISTANCE
    • ANDREW PHASE Il WA VE FORCE PROFILE
    • ANDREW PHASE II WA VE FORCE PROFILEmiddot
    • ANDREW PHASE II BRACE MODELING
    • ANDREW PHASE II FOUNDATIONS
    • ANDREW PHASE II middot middot JOINT MODELING
    • ANDREW PHASE Il JOINT MODELING OPTIONS DISCUSSION
    • ANDREW PHASE II JOINT MODEJJNG OPTIONS ILLUSTRATIONS
    • ANDREW PHASE II JOINT MODELING PHASE II METHOD
    • ANDREW PHASE II ANDREW STORM CONDmONS UPDATED
    • Platform Selection bullbullbull1111111111111
    • BasisCriteria for Selection bullbullbull1111111111111
    • CALIBRATION
    • WHERE TO INTRODUCE BIAS FACTOR (B
    • PHASE I -KEY CONCLUSIONS
    • bull CALIBRATION ISSUES PHASE II
    • PHASE II WORK SCOPE (CALIBRATION)
    • SOFTWARE
    • SENSITIVITY STUDIES POSTERIOR DISTRIBUTION OF B
    • WEIGHTING APPROACH bull
    • INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON
    • CATEGORIZATION OF ADDITIONAL PLATFORMS
    • UPDATING OF CALIBRATION RESULTS
    • APPLICATIONS OF BIAS FACTORS
    • EXAMPLE USE OF BIAS FACTORS
    • EXAMPLE APPLICATION bull PHASE I PLATFORM ST 151 K
    • EXAMPLE USE OF BIAS FACTORS -PHASE II middot
    • BENEFITS OF ANDREW JIP bull PHASE II
Page 22:  · . ' ANDREW PHASE II PROJECT OBJECTIVES . ' • ·UPDATE MODELING AND ANALYSIS RECIPE· JOINTS ' ' ·BRACES WAVE PROFILE FOUNDATIONS . ' ' CONDUCTORS· • ANALVZE PLATFORMS USING

1middot1middot ______________ _ ________ __ _ __

0

200 400 600 800 1000 12000 Force at Elevatloii (kip)

200

a ll_

Wave Force at Major Elevations for Discrete Wave Heights

180

bullHmiddot40ft

Hmiddot495ft

H bull 51 ft

Hmiddot55ft_

H-59ft I

Andrew (H bull 60 86 ft) II

i I

_11---__

- middot- Hmiddot30ft

-

- bull -

middot bull bull bull middot

---Hmiddot53ft

bullbullbullbullHmiddot57ft

-----H-63ft

Elevation Above Mudllne 100

(ft)

180

140

120

80

60

40

20

Loading at Major Elevations

---H-49511

-deg-Hmiddot608611

0 200 400 600 800 1000 1200

Force It Elevatlon (kip)

--

Base Shear vs Wave Height ST151K

6000

5000 1---------Ult1mateCapacitybull 4890 kip (100 Improved Soll)-----------~--

4000a a lii ~ 3000

81 ampI

2000

1000

0 +-~~~--i~~~~~~~~~-+-~~~--lf--~~~~~~~~-+-~~~--l

30 35 40 45 50 55 60 65

Wave Height (ft)

bull middot

~

bull

II

2

g

~

It)

0

cCD E 0 E a c E euro~ CDshygtItgtmiddoto ~ti c -a Gic CD gt ~

-= c -Cll

I z

bullO 0i i ~ g 0

() CI -bull(q-ll) iuawow llu1wniiaAO

11

shy

116

114

112

g 110

bullc = 108

~ ~ 106 -sect 104

102

100

98

96

-

6555 60

Wave Height vs Center of Action ST151K

30 35 40 45 50

Wave Height (ft)

t

Comparison of Pushover Curves for ST 151K Variable Load Pattern and Monotonic Load Pattern (Andrew)

Soll Improved 100

--- --------- --- ----6000

5000

- 4000 8 - I

1 3000 Ul

81 m 2000

_J -shy

1000

---Variable Load

- - Andrew Load

0 f-~~~~~r--~~~~-t-~~~~~-t-~~~~~+-~~~~----t~~~~~-1

0 1 - 2 3 4 5 6

Deck Displacement (ft)

-middotshy

1 t I I ~

ANDREW PHASE II

BRACE MODELING

bull MARSHALL STRUTS USED FORPHASE I

bull PHASE II WILL ASSESS DIFFERENCE IN USING BUCKLING BEAMS AND MARSHALL STRUTS

middot bull COMPARE FOR 2D FRAME

bull COMPARE FOR PLATFORM 177 WHERE LOWER

BAY EXHIBITS LARGE BRACE MOMENTS

Project andrew Model beamframe Version 1 Mon Nov 14 145447 1994

~

middot Frame Test buckling beam-column model

-

Project andrew Model strutframe Version 1 Mon Nov 14 145754 1994

bull

Frame Test strut model

middot-middot _____________--_

_middotmiddotr_~ --ll--~-middot__

Buckling Beam bull

Strut (K=O 65)

Lateral Displacement

ANDREW PHASE II

FOUNDATIONS

bull ~LICIT NONLINEAR API FOUNDATION USED INiPHASE I

bull FOUNDATIONS

JIP SELECTED MINI-VANE DATA FOR

STRENGTH PROFILE

bull COMMENTS ON BENCHMARK COMPARISONS

- SIGNIFICANT VARIABILITY DUE TO FOUNDATION MODELING METHODS

II Ii

z

ctV 4x SPlO caissonfy-42s soil-api(lf-097step-114)

Inelastic Events Legend Elastic Strut Buckling

Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield

--------shyBeam Clmn Fully Plastic Fracture

~--~~~~~~~~~--~----~__i~~---~~~~__--~

Pro)ect apipile Model South Pelto 10 Version Wed May 18072235 1994 - -

1bullt

+

t

f i bullbullbull

bullbull bullbull

I I

5

~x SPlO Caissonfy=42s soil=api(lf=lOOstep=137)

Inelastic Events Legend

Elastic Strut Bucklilg

Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield

earn Clmn Full~ Plas1c Fracture

ANDREW PHASE II middot middot

JOINT MODELING

middotbull API RIGID-PLASTIC USED IN PHASE I

a ELASTIC-BRITTLE USED IN PHASE IB

amiddot MSL WILL PROVIDE NONLINEAR FORCE-DEFORMATION

AND MOMENT-DEFORMATION RELATIONS FOR PHASE II

a MODELING OPTIONS FOR PHASE II -

l

- UNCO~LEDmiddotONE DIMENSIONAL MEMBER END JOINT MODELING

J Ii

ANDREW PHASE Il

JOINT MODELING OPTIONS DISCUSSION

bull JD COUPLED INTERACTION SURFACE

- IDEAL

- DATA NOT AVAILABLE TO CALIBRATE

bull 2D COUPLED INTERACTION SURFACE

- SIGNIFICANT MODELING IMPROVEMENT

- DATAAVAILABLE

- IMPLEMENTATION PRACTICAL BUT NEEDS DEVELOPMENT AND CALIBRATION

bull 2D UNCOUPLED INTERACTION SURFACE

- SIGNIFICANT MODELING IMPROVEMENT

- DATA AVAILABLE

- IMPLEMENTATION PRACTICAL AND AVAILABLE

ANDREW PHASE II

JOINT MODEJJNG OPTIONS ILLUSTRATIONS

P

e

ANDREW PHASE II

JOINT MODELING PHASE II METHOD

e

I I I

ANDREW PHASE II

ANDREW STORM CONDmONS UPDATED

bull -PHASE I USED OCEANWEATHER HINDCAST-(Nov 1992)

bull middotmiddotPHASE II WILL USE THE REVISED HINDCAST (Nov 1994)

bull lt c lt I lt

2-2

l---4-~~~+-~~-+-~~~1--~~-+-~~--l--~T--+-~~--i24

f-tJVtMfiz~ 1Cfl (r STvb(

Plollol bullbull 5middotNOYmiddot9214bull1J Irbullbull Illbull [ANDREVl~NDREVCUSTbullMAXll 5-NOY-1992 10bull54 - 9208

HEIGHT lml

I~ MAXI

I L 1~middotamp SZ~- ~A I 13c

I J) ( r I X~0iWJ~0gtJ gt L gt gt I 1c I 120

-- N

~ deg

I I I I I I e C I ~=-41 122 -94 -92 -90 -88 -86 -84 -82 -80

-

Figure 7 contours of maximum hindcast significant wave height

_

~

~tl ~

NOltNttIJ 17lt ltNS S iJJ)y

Maximum Significant Wave Height - Hurricane Andrew 1992

30 ~

26 ~ I ~-~~~- == -shy -~ 35-_

24deg1--shy --------~- middot-middot-------middotmiddotmiddot----middot------middot-middot~middot- - - ~-----1---middot

22middot ~~~~~~~-~ -94 -92 -90 -88 -84deg-86 -82 -somiddot

c

-

Platform Selection bullbullbull1111111111111

0 BasisCriteria for Selection middot

0 Review of Phase I Platforms middot 0 Alternative Platforms

0 PMB Recommendations

0 Participants Comments

gt

L

l I

BasisCriteria for Selection bullbullbull1111111111111

0 Benefit for Calibration - Classification (UnexpectedExpected)

- Failure Mode (StructuresFoundations)

- Damaged StructuresKnown response

- Better Representation of Full ~opulation

0 Availability of Data - Existing Models

- Inspection Data

- Damage Data

PREDICTED BEHAVIOR

-

OBSERVED

BEHAVIOR

-SURVIVED DAMAGED FAILED

SURVIVED

ST134W WD90A

-MC311 -

MC397 -

--

-

-

ST151K ST130Q

-

-

-

DAMAGED - -

-

-

--

-

-

ST52 SS139 ST177B -

ST161A

-_

-

FAILED

-

-

-

ST151H ST130A ST72

Operator - Platfonns - Platfonns Survived Damagedamp

- Falled -

bull

-

-

Jle eBDIGIPllDbl -Amocoshy 30 1 Chevron -

143 12 Exxon 72 Mobll 22 1 Phllllps 7 bull

-Shell 39 Unocal 25 1

Total Partlclnts Platfonns 338 15

Other Operators bull- 310 13

-

Total Platfonns - -

648 28

fbml

Platforms Considered 6 7

tie of Total Platfonns 1 25

OVERVIEW OF PLATFORMS SURVIVEDDAMAGEDFAILED

-

shy

v

Platforms Evaluated in Phase I

-

-Platfonn Water Year Number Analysis Avallabillty of Availability Platform

- Name Depth rl of Model DetaDed rl ClassUlcatlon

Installation legs Inspection Soll Report Based on Information Phase I Calibration - -ft

- -- Suaill flalfnrm ~BSi -STISIK 137 1963 8 CAP Yes Yes Un eel Sarvlval

-STIJOO 170 1964 4 CAPmiddotshy Yes ST134data u ST134W 137 1981 4 CAP Yes Yesmiddot E

-

edSarvlval -

middotsurvival

WD90A 184 - 1964 8 CAP Yes E -

- Sarvlvill

MC3ll 343 1978 8 KARMA

MC397 468 1991 4 KARMA -

bull ~reglmmiddot- -

Yes Sure Survival

bull Yes - Sare Sanlval-

-T23STS2 63 1969 4 CAP

T2SSS139 62 1969 4 CAP -

Yes

Yes

Yes Ex

Yes Une

bull SurvivalLikely to DamalI

middot bull SanlvalLlkely to DamalI

ST161A 118 1964 8 USFOS

Eail11a pound1al(bullllm Clss ST177B 142 --1965 8 CAP

STISIH 137 1964 - 8 CAP C

STIJOA 140 19S8 8 CAP - -

Yes

Yes

ollamed

Yes

- Yes E

-

at about 1 mUe

Yes

ST134data

bull SafvlvalLlkelv to Dam- bull- -

E middotrahre

Emeried Failure -

Likely FaUare

T21ST72 61 1969 4 CAP Collamed No Likely FaUure

Platforms Identified for Evaluating Their Likelihood of Falling Under UnexpeCted Survival Category

CllEllATOa

AllB

llLOCll

Sl1lvcnJllE

NAME

YEAR

INSTAUJCD

WATER

IJBPJB

middotmiddotDISTANCE

ftOMSllOllE

(MIUIS)

Ill

(M)

AMOCO BllW

STllI

WDI075

A

bull D

4

n uo 17J

44

32

bull 20

10

bull CllEVllOH STtllO c I 1bull 21 11

STl130 D 0 161 21 11 STllJ4 I 0 137 D - 11

STIUI G I 137 32 11

STIUI I 0 121 32 11

STIUI

J 0 140 32 11

STIUI It 4 137 32 11

STllSI PROl)I 0 137 32 11

STllSI PROD-2 0 137 32 11

STll7 A 0

140 35 11

WDl117 c 65 214 34 10

DXON STshy E 51 14 bull STl165 A t3 bull 34 WD0073

WD0073

WD0073

A _ c

4

62

4

IA IA

17J

22

22

22

bullbullbull WD003-shy E 65 161 Z7

A 0 IN 31 MOBD Bl12I

ssn A bull

50

50

31

7

bull 7

MIJllPllY SS114 B 7 50 7 7

PlllUIPS SSl14t A 55 33 7

SAMEDAN ST17J A 4 117 7 bull 7

SJIELL WD0103 A 65 223 27 WDllOJ bull 65 221 27 WDll04 c 65 221 27 WDOI04 D 67 2 27

UNOCAL

SSl20I

SSl20I

bull SSl20I

SS0209

SS0209

SS0209

SS0209

SS 0253

SS 0253

E

I

PtJllE

A

bull D

G A

c

0

4

4

7J

61

0

0

69

103

97

97

9S

100

95

100

165

175

34

34

34

34

34

34

34

54

54

bullbullbullbullbullbullbullbullbull

Platforms Recommended for Phase II

llalr y_ w- Aot Av- _ I 1 Doplla Model nm bull bull _ bull I I H11 s18-1 Irht lbwl~ bullbullbull(-illtdlel OM lllMtD

II ---

-STISIIC 137 1963 I CAP Yeo Yeo I Yeo

STtshy -170 1964 4 CAP Yeo ST134dlU I Yeo

STl34W 137 Yeo Yeo -1981 4 CAP D Yeo

WD90A 114 1964 I CAP Yeo - D

MC311 30 1971 I KARMA Yeo -sre-

MC397 middot 461 1991 4 ICARMA Yeo Sireshy - - r- -

__ T13ST5l 63 1969 4 CAP Yeo Yeobull middot- ~n- D Yeo _ Tl5SSl39 62 1969 4 CAP Yeo Yeo I Yeo

_ STIT711 142 1965 I CAP Yeo I Yeo middot-1shySTl61A Ill 1964 I USFOS Yeo Yeo - D bulloshy

- r-

STISIH 137 1964 I CAP y D y -_ STl31A 140 1951 I CAP Yeo STl34dlU I Yeo

T21Sll 61 1969 4 CAP No rar-middot I Yeo

~-rmiddot-middot SS209A 95 1967 16 CAP Yeo D -SSU4R

11 UI A 41 -in D SS 2741AIM CAP bull

_ _ -shy - -shy- -shy

CALIBRATION

bull PHASEIAPPROACH

bull CALIBRATION ISSUES

bull PHASE II WORK SCOPE

bull SOFlWARE AVAILABILITY

bull EFFECT OF ADDITIONAL PLATFORMS

bull WEIGHTING APPROACH

bull EXAMPLE APPLICATION middot

(l=Ofce Mode] Wind

~

fcurrent l -middot

~SW ~leldl4

ultimate resistance force I first

design load Illgt

i _

deflection

-

bull

4

Historical Hurrlcane Dariiage and Survival of Platforms

Components Required for Calibration (Petrauskas 1992)

PRIOR Distribution

PLATFORM EXPERIENCE

BAYESIAN Survivals and Failures

UPDATING

Waves Winds and Currents bull

POSTERIOR Distribution

- A_middotJA ~ Survival Data Failure Data

Result Result

Load Resistance

middotfl

-

Bayesian Updaing middotGeneral Description (Petrauskas 1992)

L Imiddot

WHERE TO INTRODUCE BIAS FACTOR (B

On Resistance (Rmue - B (RPREDICreD I

middotOn Loads (Smue middot - B (SPRE01cTEo

On Safety Margin (RSmue - B (RSPREDICTED

CALIBRATION METHODOLOGY

gt

r

FIELD OBSERVATIONS

Observed Behavior

survivals Damages

Failures

r

bull

CAPACITY ANALYSIS

Predicted Load and Resistance Predicted Failure Modes

Survival Damage Failure cases

RELIABILITY ANALYSIS

- Probability of Failure ~

Ukellhood Failure

BAYESIAN UPDATING

Prior Distnbution of B Combined Likelihood Function

Posterior Dlstnbution of B

BIAS FACTOR

Mean Value cov bull

Represents Modeling Uncertainties

Prior Distribution of B Likelihood Functionmiddot Observing Success case STISIK and Posterior Distribution of B

2 00

uo

160

140

120

e 100 L $ uoIshy

0 60 ~ ~ 040

020

0 00

0 00 0 50

_ I 00

b

I 50 2 00

-shy lk [bl STISIK)

--shy Pnor-f(b)

--shy Pos1enor r(bl ISIK)

Prior or B Mean bull 10 COV bull 030

Posterlor of Bbull Meanbull 136COV 0 16

2 50

Likelihood Function and Posterior Distribution Example Platform STISIK

- -

~

- Prior and Posterior Distributions or B Failure cases (ST177B STlSlH ST130A T21) Damage cases (T23 T25 ST161A) Success cases (STlSlK ST130Q ST134W WD90A MC311 MC397)

400

350

300

2 so 0-

I

bull bull

bull bull gt

- bullbull ~ bull bull I bull

_ bull bull bullbull I bull bull bullII V -

bull bull

I bull ~ bull

bullbull bullI bull bullI bull bull bull

middot -~

bull__ ~ middotI middotmiddot

L 2 00

e ~ I SO

-I 00

- 0 so

000 gt

0 00 020 0 40 060 0 80 I 00 120 140 160 110 200

b -

-

shy

Prior-r

- - - - Posterior fLlll failure cases

bull bull bull bull bull bull bull Posterior flall damage cases

- bull bull - bull middot Postenor flall success cases

Posterior flall success fulure damage cases

-

Posterior of B Mean= 119 COV 0101

Posterior Distribution or Blas Factor (8) bull Different Categories and All Cases Combined

shy

l

PHASE I - KEY CONCLUSIONS

bull GLOBAL CORRECTION FACTOR

bull 15 TO 20 CONSERVATISM MEAN)

bull UNEXPECTED SURVIVALS HAVE GREATEST IMPACT

bull DAMAGED CASES MORE INFORMATIVE

bull CALIBRATION ISSUES PHASE II

bull

EFFECT OF IMPROVED CAPACITY ANALYSIS RECIPE

bull APPLICABILITY TO MULTIPLE FAILURE MODES

bull NO FOUNDATION DAMAGE CASE INCLUDED

bull IMPROVED REPRESENTATION OF FULL ANDREW DISTRIBUTION

-bull NUMBER OF PLATFORMS CONSIDERED

I 1 I

I t ) bull ~

PHASE II WORK SCOPE (CALIBRATION)

bull

bull ESTABLISH MULTIPLE BIAS FACTORS

Use Nine Base Case Platforms

middot - Capacity Analysis for Revised recipe

Foundation Blas Factor from Foundation JIP

bull DEVELOP WEIGHTING APPROACH

bull UPDATE MULTIPLE BIAS FACTORS

bullEffect of Platforms Not In the Nine Base Cases

bull EXAMPLE APPLICATION OF BIAS FACTORS I

bullbull

PHASE I PHASE II

CAPACITYbull ANALYSISRu bull

---------------------1----middot

Jc1 - iJbullraquoJ dbull

FttlS frt[f~l ratdhJJ

-

LIKELIHOOD FUNCTION

DEVELOPMENT

lk Jb Jfall1N) bull Jb)

111 Ill Isurvtvoq bull 1 - bl

-middot -------------------shy ~---------------------middot-middot

r111ir8 Cb1 111 Cbl BAYESIAN

UPDATING

~cove

R1R2 bull middot---------------------~----

Jblbull P(g cOJ

1 bull JbRI (MBSI I bull 1

~middot

llcJllJfalluro) bull P(gcOJ Ill 111 b21fallun) bull P ((g1 CO) U 1112 c OD

middot--------------------shy ----------------------shy bull

r111111r8 Cbl 111 Ill i-1nro1

deg ~1b2)laquottfrr (b1b21-lnfo) re Ill bull 1-11middot b2) bull lodegamp

middot

i1 covamp i2 cov~i

-

shy

--

Likelihood Function vs b All Success Case STISI K STl30Q STl34W WD90A MC311 MC397

I 0000

0 9000

OIOOO

0 7000

~ 0amp000

ta osoOO

~ 0 4000

0 3000

0 2000

0 1000

bull 0 0000

bull

I

middot

middot bull I

Ibull

-sibull ~ -middot-middot-middot-middot -middot-

----7----+-----~ I

Ibull bull bull I

rbull

--- Jk ISIK[blsuccess)

- bull - bull - lk 130Q[blsuccess) bull

lk 134W[blsuccess)

Jk-WD90A(blsuccc~I

- - - lk MC397[blsucccss)

--ltgt-- lk middotMC31 l[blsucccss)

--- Jk [bl 6 out of6succcsHascs)

05 I 5 2 250

b

Likelihood Functionsmiddot Success Cases - Phbullsc L

shy

bullbull bull bull bull bull bull -------------------~

bull bull bullbull bullbull bullbull bullbull bullbull bullbull bull

bull bullbull

bullbullbullbull

bullbull

bullbullbullbullbullbullbull

bullbullbullbullbullbull

bullbullbullbullbullbullbull

z 0

t z ) LL

c 0 0c i ~ i

J

~ sect ~--------------~ e l5

I

I

SOFTWARE

bull PF PROGRAM

bull CALREL

bull PF +CALREL

bull RELACS

bull

gt i

l

SENSITIVITY STUDIES

POSTERIOR DISTRIBUTION OF B

bull EFFECT OF CHANGE IN DISTRIBUTION OF PRIOR OF B

bull EFFECT OF REMOVAL OF ONE PLATFORM FROM EACH CATEGORY

bull EFFECT OF REMOVAL OF THREE DAMAGE CASES

bull EFFECT OF ADDITIONAL 13 SIMILAR PLATFORMS

-I

Prior and Posterior Distributions of B Variation In COV of PriorofB fromOl to04 -- -

4500

-4000

-

- -

3500 --middot - bull bull - bull bull Prior-r ICOV =02

3000

- - - - Prior-r ICOV =03 -

bull -~ e 2soo -

bull bull bull bull bull bull bull Pnor-r ICOV =o 4t frJ ~ 2000 Posterior rlCOV =0 2

bull bull Postenor rlCOV =0 3 bull bull 1500 bull middot bull - shy -middot- Posterior rlCOV =04

middot~ I 000 - bull

-middotmiddot ~ ~middotmiddot M -~

- middot 0 500

wbullbullbull _ I middot~ middotmiddotmiddotmiddotmiddot middotmiddotmiddotmiddotmiddot - -~ ~ __ middot_ _ bull -shy0 000

0 000 0200 0 400 0 600 0 800 IOOO 1200 1400 1600 1800 2000 -

- - b

-

=shy

Posterior Distribution of Bias Factor (B) bull Effect of Variation In COV of Prior of B

bull

bullPrior and Posterior Distributions of B Evaluating Effect of Removal of One Platform from Each Category from the Base Case of 13 Platforms

400

~ ~ 350

j IA i

3 00 bull I bull I ---Prior-r I bull l

250 1 I 1

I bull bull ---Posterior r lall 13 casesshy- ~E 2 00 bull bull bull bull bull bull bull Posterior r without STISIK

- ~ 1 bull - bull I - bull bull - bull bull Posterior r Iwithout T2I I 50 I

J-tt~i - bull - bull - Posterior r Iwithout T2S

tOO [-

I I ~ ~ I 0 50

0 00

___ r J i

l~T

~--

0 00 0 20 0 40 0 60 080 I 00 120 140 160 110 2 00

b

zshy

Posterior Distribution of Bias Factor (8) bull Effect of Removal of middotPlatform from Any Category

bullPrior and Posterior Distributions or B Evaluating Effect or Removal or 3 Damage Cases from the Base Case or 13 Platrorms

4

35

3

25

e L 2 ~ bulle shy

I 15

bull I

05

0

0 02 04 06 I 2 I 4 16 IS 2

I I

I

I

I Ibull

J -tmiddot L--

bullI

~

I

fl

II ~

I I v I

middot bull bull

bull L

~ ll- I-~lt

OS I

b

I

bull

1middot bullbullbullbull

~ __

bull

- - - - Prior-r

- - - - Postenor flall failure cases

Postenor flall success cases

--- Posterior flall 13 cases

---Posterior fI success failure cases only

IPosterior or B Mean =t11 COV =0121

-

_I

Posterir Distribution or Bias Factor (B) bull Effect or Removal or 3-Damage Cases

-

bullPrior and Posterior Distributions or B Evaluating Effect or Additional 13 Similar Observations as

Used in Andrew JIP The POsterior or B from Andrew-JIP Considered as New Prior or B

- 500

450

400

350

bull bull

bull bull bull bull J bull

bull bull f ~ I

I ~ Ibull bull

bullbull bull

l bull I -bullbull bull bull

I

I bull bull

bull I I~~bull bull bull

bull ~

I bull bull

~ ~_ ~~~

- - New Prior r bull

Posterior 114 new failure cases ~ 300- bull bull bull bull bull bull bull Posterior f 13 new damage casesl 150 ~

- bull bull - bull bull Posterior 116 new success cases i 2 00

Postenor 1113 new success failure -150 damage cases

I 00

0 50 -

0 00 0 00 0 20 0 40 0 60 0 80 I 00 120 I 40 160 110 2 00

Posterior ol B Mean= 120 COV= 0071 b

shy

shy

_rshy

Posterior Distribution of Bias Factor (B) bull Effect of Additional 13-Similar Observations

1

1 I

WEIGHTING APPROACH bull

bull ASSESSMENT OF PLATFORMS DATA

Select Platforms Most Mappable to the Analyzed Platforms

Unlikely SurvivalDamage Cases Identification

Based on Cut-Off Criteria Agemiddot Hs etc

bull INFORMATION REQUIRED FOR COMPARISON

bull CATEGORIZATION OF ADDITIONALmiddot PLATFORMS

bull UPDATING OF CALIBRATION RESULTS

--

OVERVIEW OF PLATFORMS SURVIVEDDAMAGED -~

Operator -

- Platforms

Survived CS)

Platforms

Damaged CD)

Platforms Used In

Phase IJIP -

JI~ ~Bill~IRBDll -

Amoco -

Chevron -

Exxon

Mob II

Phllllps

Shell

Unocal -

30

143

72 22

7

39 25shy -

1

12

bull

1 bull

1

1(S)+1 (D)

3 (S) +3(D)

2 (S)

bull shy

bull

bull

bull

Total Partlclnants Platforms 338 - 15 10C6S+4D) -

Other Operators

310

13

3 (D)

Total Platforms 648 28 13C6S+7D) -

Summary of Andrew JIP bull Phase I Platforms

Plairshy w v- N-ber Lea-le 1 shy car Hind_ o-w N-bull Deplh fl fl IJpe la bull n Yeadon Jadtet lbelledtf liefalit -- _ -OD

IL H - -shy _ICllshy IL IL - - rQClto ~ _

K STUIK 137 1963 8 365 608S4 Cellardeck

-middot STl-shy 170 1964 4 -~ K 39 62269 Cellardeck - --ST134W 137 1911 4 43 608S4 Smvlnlshy- _ - -

WD90A 184 1964 8 K 267Sbull34 83 SOS31 SUboftllar deck SmvMI

MC311 343 1978 8 x S7 62 - MC397 468 1991 4 x shy S33bull6SO 65911 -shySmeSmvlvol -- gt -

r-- T23ST52 63 1969 4 ton- shyKshy 52594 S03 -T25SSl39 62 1969 4 Kshy SOS94 S063 - middot ton-shy__ -

bullbull a ton--shyST161A 118 1964 8 K - 347S 51495 Cellarclecti

r~

ST1771 142 1965 8 K 39 60167 CeDordeck failme

STISIH 137 1964 8 K 395 608S4 CeDardeck Failme

K ST130A 140 1958 a SS 6095 I lblv Pai1uno - T21ST72 61 1969 4 Kshy S4S94 4974 I lhlv Pai1uno -

- -

JE

INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON

bull WATER DEPTH

bull bull YEAR OF INSTALLATIONDESIGN

bull PLATFORM ORIENTATION

bull PLATFORM LATITUDE AND LONGITUDE

bull KEY SKETCHES OF PLATFORM STRUCTURAL FRAMING

Number of Legs Plies Leg Batter Number of Conductors bull Deck Elevations (Including sub-cellar deck If any) Jacket Framing Elevations Brace type In Vertical Frames Sizes of Primary Members (leg pile vertical frames braces primary horizontals) Joint Details (can gap etc) LegPiie Annulus Grouting Status

bull middotPOST ANDREW INSPECTION REPORT Significant Wave Height Hs (ft)

bull INFORMATION ON DAMAGES PRIOR TO ANDREW

bull RESULTS FROM IN-HOUSE STUDIES IF AVAILABLE

bull ~RIGINAL DESIGN CRITERIA INFORMATION IF AVAILABLE

bull SOIL SHEAR STRENGTH PROFILE IF AVAILABLE

Platforms Identified for Evaluating Therr Likelihood of Falling Under Unexpected Survival Category

OPlRATOR

middotADA

BLOClt

llmucnJllE

NAME

YEAll

INSTAUJD

WA1ER

DBPTll

DISTANCE

JlllOM SHOllE

(MILU)

Bl

(M)

AMOCO 111113

STl11

WDI075

A

bull D

St

17

120

1n

44

20

10 I

CllllVllQfl STlllO c 61 110 21 11 STIUO D Q 1110 21 11

STl134 p Q 137 29 11

ST0151 G 61 137 11

ST0151 I 3 121 11

ST0151 J Q 140 11

STl151 It 137 11

ST0151 PJl001 Q UT 11

ST0151 PROD-2 Q UT 11

ST017 A 3 140 35 11

WDIUT c 65 214 34 10

llXXON STI055 II 51 6T 14 I

STl15 A t3 34

WDI073

WD0073

Amiddotshy Q

1111 1111

22

22

I

I

WDI073 c 1n 22 I

- WDOCl93-shy II

65 1110 2T A 3 194 31 MOBU 111121

SS IOl7

A B

501

50

50

31

T

I

T 6

MtlllPllY SS114 B T 50 T T

PlllllIPS SS0149 A 5 55 33 T

SAMllDAN STln A 107 T T

SBllLL WDl103 A 65 223 2T WDl103 B 65 221 2T WD0104 c 65 221 2T WD0104 D 6T 20 2T

UNOCAL SS020I

SS 020I SSl20I

SS 0209

SS0209

II p

PURll

A

B

3

n 61

103

97

97

95

100

34

34

34

34 34

bullbull bullbullbull

- SS0209 D u 95 34 I

- SS 1209 SSl253

SS0253

G A

c

6T

2

100

165

1T5

34

54

54

bullbull I

I I

I I

SOUTH MARCH

~LAND

N

SOUTH PaTOI ENE ISWltD AREA

MMS Defined Limits ror Offshore Structures Requiring Inspections Following Andrew

t---Jbulln3 T25bull I bullTtl

bull1$1Htl1Abull bull1SIK1779bull

~

middot-DaTA AREA~T

CRANO rsLE

AREA bull1300

bullll4W

=shy

bullPath of Hurricane Andrew with Platforms Used In Calibration

Likelihood Function vs b All Success Case STISI K STIJOQ ST134W WD90A MCJll MCJ97

I 0000

0 9000

0 8000

07000

-Bo6000

tiiii 0 5000 -

04000

0 3000

0 2000

01000

0 0000

0 05 25

ri1rt middot-middot-middot-middot-middot-

I 5 2

b

---11 middot151K(blsu~ccsI

- bull - bull - fk 130Q(bfU~lCSS)

lie bull134W(blsuccc]

lk-WD90A[blsultCS]

- - - lk middotMC397(bfUtlCS]

--o-- lk-MCJll(bbuctcsJ

--- ll (bl 6 out of 6 Mlttcs case]

_

( Likelihood Functions middotSuccess Cases - p - 1 c I

CATEGORIZATION OF ADDITIONAL PLATFORMS

I I I

bull CATEGORIES

Unexpected Survivals

Expected Survivals

bull APPROACH

Identify Corresponding Platform(s) Evaluated In the JIP

Determine Max Wave Height at Platform Location

Evaluate Whether Wave Hit the Deck

middot - Evaluate Likelihood of R and S to Differ from the Base Cases

Review Joint Information

Assess Special Features

bull IDENTIFY NUMBER OF PLATFORMS IN EACH CATEGORY

I

UPDATING OF CALIBRATION RESULTS

bull INCLUDE EFFECT OF ADDITIONAL PLATFORMS ON bullBbull

Consider Increased 1Uncertalntles In R and S

Develop bullNeW Ukellhood Functions for Different Categories

middotLet bullxbull Number of Ukely Unexpected Survival Cases and y Number of Ukely Expected Survival Cases

Determine Combined Likelihood Function lk (b Iadditional platforms) middot

bull REVISED LIKELIHOOD FUNCTION

lk(b I all cases) = lk(b I 9 platforms) middot x lk(b Ix platforms) x lk(b I y platforms)

bull ESTABLISH REVISED BIAS FACTOR B

f8 (b) u f8 (b) lk (b I all cases)

II

t t ~

shy

APPLICATIONS OF BIAS FACTORS

ii APPLICABLE TO PARTICULAR LOAD AND RESISTANCE RECIPE

bull USE BY THE API COMMITTEES AND MMS

Criteria Revisions

Recommended Design Guidelines Revisions bull

bullbull REASSESSMENT STUDIES AND DECISION MAKING

middot

bull RISK ASSESSMENT STUDIES BY THE OPERATORS

Individual Platforms

Fleet of Platforms

middot IMR Program Implementation

bullbull NOT TO BE USED IN DESIGN

I I fl J

EXAMPLE USE OF BIAS FACTORS

bull PHASE I EXAMPLE APPLICATION

bull EVALUATION OF ANNUAL PROBABILITY OF FAILURE

Establish Annual Maximum Seastate Data

Determine Coefficients to Define Base Shear

Determine Ultimate Capacity of Platform

Determine Probablllty of Failure Versus bullbbull

-bull DETERMINE UPDATED PROBABILITY OF FAILURE

Pf - P [S gt R]

Pf I P [~ gt Rb I b] f 8 (b) db

Pr - J P [S gt Rb I b] f bull 8 (b) db

bull TOTAL PROBABILITIES OF FAILURE

Total Pf - 1 bull Il [1 middotPf] - l Pf bull

Total Pf - 1 bull Il [1middot Pf] - l Pfmiddot

Total Pr = 1 - n [1middot PfbullJ = l Pfshy

I I I

EXAMPLE APPLICATION bull PHASE I

PLATFORM ST 151 K

bull GULF OF MEXICO

bull WATER DEPTH = 137 FT

bull ANNUAL MAXIMUM HS MEDIAN 175 ft COV 031

bull ASSUMED STORM DURATION 3 hoursmiddot

bull ASSUMED NUMBER OF WAVES 800

bull MEDIAN HS FOR STORM HOURS 1 2 AND 3

bull MEDIAN CURRENT VALUE bull

bull DETERMINE ULTIMATE CAPACITY FOR THREE DIRECTIONS

bull DETERMINE ANNUAL FAILURE PROBABILITIES

Broadsidemiddot Diagonal End-on

0037 0025 0007

0071 0044 0018

0022 0016 0004

DETERMINE TOTAL ANNUAL PROBABILITY OF FAILURE

Pf - 007

Pf - 013

Pr - 004

bull

EXAMPLE USE OF BIAS FACTORS - PHASE II middot

bull DETERMINE PREDICTIVE PROBABILITY OF FAILURE

Case - 1

Pf _ S S Pf (b b2) fa1a2 (b b2) db db2

g bull -1 (b1 R I S) - 1

92 - (b2 R2 I S) - 1

Case -2

Pf s ~ [bJ f11 (b) db

Pf [bJ p [g lt 0)

g - bull 1 (b1 R1 I S) - 1

1

bull1) l

I

BENEFITS OF ANDREW JIP bull PHASE II

lt

bull IMPROVED LOAD AND RESISTANCE RECIPE

bull IMPROVED CALIBRATION PROCEDURE

bull MULTIPLE BIAS FACTORS

bull PROCEDURE TO INCLUDE EFFECT OF OTHER PLATFORMS

bull PROCEDURE TO USE MULTIPLE BIAS FACTORS

  • Structure Bookmarks
    • ANDREW PHASE II JOINT1 INDUSTRY PROJECT
    • PROGRESS MEETING NUMBER 1
    • ANDREW PHASE II JOINT INDUSTRY PROJECT PROGRESS MEETING NUMBER 1 NOVEMBER 16 AGENDA
    • ANDREW PHASE II UST OF PARTICIPANTS
    • ANDREW PHASE I OBJECTIVES AND RESULTS
    • ANDREW PHASE II PROJECT OBJECTIVES
    • SUBSEQUENT PROJECTS
    • ANDREW PHASE II PROJECT TEAM
    • ANDREW PHASE II UPDATE MODELING middotAND ANALYSIS RECIPE
    • -ANDREW PHASE II CONDUCTOR RESISTANCE
    • ANDREW PHASE Il WA VE FORCE PROFILE
    • ANDREW PHASE II WA VE FORCE PROFILEmiddot
    • ANDREW PHASE II BRACE MODELING
    • ANDREW PHASE II FOUNDATIONS
    • ANDREW PHASE II middot middot JOINT MODELING
    • ANDREW PHASE Il JOINT MODELING OPTIONS DISCUSSION
    • ANDREW PHASE II JOINT MODEJJNG OPTIONS ILLUSTRATIONS
    • ANDREW PHASE II JOINT MODELING PHASE II METHOD
    • ANDREW PHASE II ANDREW STORM CONDmONS UPDATED
    • Platform Selection bullbullbull1111111111111
    • BasisCriteria for Selection bullbullbull1111111111111
    • CALIBRATION
    • WHERE TO INTRODUCE BIAS FACTOR (B
    • PHASE I -KEY CONCLUSIONS
    • bull CALIBRATION ISSUES PHASE II
    • PHASE II WORK SCOPE (CALIBRATION)
    • SOFTWARE
    • SENSITIVITY STUDIES POSTERIOR DISTRIBUTION OF B
    • WEIGHTING APPROACH bull
    • INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON
    • CATEGORIZATION OF ADDITIONAL PLATFORMS
    • UPDATING OF CALIBRATION RESULTS
    • APPLICATIONS OF BIAS FACTORS
    • EXAMPLE USE OF BIAS FACTORS
    • EXAMPLE APPLICATION bull PHASE I PLATFORM ST 151 K
    • EXAMPLE USE OF BIAS FACTORS -PHASE II middot
    • BENEFITS OF ANDREW JIP bull PHASE II
Page 23:  · . ' ANDREW PHASE II PROJECT OBJECTIVES . ' • ·UPDATE MODELING AND ANALYSIS RECIPE· JOINTS ' ' ·BRACES WAVE PROFILE FOUNDATIONS . ' ' CONDUCTORS· • ANALVZE PLATFORMS USING

Elevation Above Mudllne 100

(ft)

180

140

120

80

60

40

20

Loading at Major Elevations

---H-49511

-deg-Hmiddot608611

0 200 400 600 800 1000 1200

Force It Elevatlon (kip)

--

Base Shear vs Wave Height ST151K

6000

5000 1---------Ult1mateCapacitybull 4890 kip (100 Improved Soll)-----------~--

4000a a lii ~ 3000

81 ampI

2000

1000

0 +-~~~--i~~~~~~~~~-+-~~~--lf--~~~~~~~~-+-~~~--l

30 35 40 45 50 55 60 65

Wave Height (ft)

bull middot

~

bull

II

2

g

~

It)

0

cCD E 0 E a c E euro~ CDshygtItgtmiddoto ~ti c -a Gic CD gt ~

-= c -Cll

I z

bullO 0i i ~ g 0

() CI -bull(q-ll) iuawow llu1wniiaAO

11

shy

116

114

112

g 110

bullc = 108

~ ~ 106 -sect 104

102

100

98

96

-

6555 60

Wave Height vs Center of Action ST151K

30 35 40 45 50

Wave Height (ft)

t

Comparison of Pushover Curves for ST 151K Variable Load Pattern and Monotonic Load Pattern (Andrew)

Soll Improved 100

--- --------- --- ----6000

5000

- 4000 8 - I

1 3000 Ul

81 m 2000

_J -shy

1000

---Variable Load

- - Andrew Load

0 f-~~~~~r--~~~~-t-~~~~~-t-~~~~~+-~~~~----t~~~~~-1

0 1 - 2 3 4 5 6

Deck Displacement (ft)

-middotshy

1 t I I ~

ANDREW PHASE II

BRACE MODELING

bull MARSHALL STRUTS USED FORPHASE I

bull PHASE II WILL ASSESS DIFFERENCE IN USING BUCKLING BEAMS AND MARSHALL STRUTS

middot bull COMPARE FOR 2D FRAME

bull COMPARE FOR PLATFORM 177 WHERE LOWER

BAY EXHIBITS LARGE BRACE MOMENTS

Project andrew Model beamframe Version 1 Mon Nov 14 145447 1994

~

middot Frame Test buckling beam-column model

-

Project andrew Model strutframe Version 1 Mon Nov 14 145754 1994

bull

Frame Test strut model

middot-middot _____________--_

_middotmiddotr_~ --ll--~-middot__

Buckling Beam bull

Strut (K=O 65)

Lateral Displacement

ANDREW PHASE II

FOUNDATIONS

bull ~LICIT NONLINEAR API FOUNDATION USED INiPHASE I

bull FOUNDATIONS

JIP SELECTED MINI-VANE DATA FOR

STRENGTH PROFILE

bull COMMENTS ON BENCHMARK COMPARISONS

- SIGNIFICANT VARIABILITY DUE TO FOUNDATION MODELING METHODS

II Ii

z

ctV 4x SPlO caissonfy-42s soil-api(lf-097step-114)

Inelastic Events Legend Elastic Strut Buckling

Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield

--------shyBeam Clmn Fully Plastic Fracture

~--~~~~~~~~~--~----~__i~~---~~~~__--~

Pro)ect apipile Model South Pelto 10 Version Wed May 18072235 1994 - -

1bullt

+

t

f i bullbullbull

bullbull bullbull

I I

5

~x SPlO Caissonfy=42s soil=api(lf=lOOstep=137)

Inelastic Events Legend

Elastic Strut Bucklilg

Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield

earn Clmn Full~ Plas1c Fracture

ANDREW PHASE II middot middot

JOINT MODELING

middotbull API RIGID-PLASTIC USED IN PHASE I

a ELASTIC-BRITTLE USED IN PHASE IB

amiddot MSL WILL PROVIDE NONLINEAR FORCE-DEFORMATION

AND MOMENT-DEFORMATION RELATIONS FOR PHASE II

a MODELING OPTIONS FOR PHASE II -

l

- UNCO~LEDmiddotONE DIMENSIONAL MEMBER END JOINT MODELING

J Ii

ANDREW PHASE Il

JOINT MODELING OPTIONS DISCUSSION

bull JD COUPLED INTERACTION SURFACE

- IDEAL

- DATA NOT AVAILABLE TO CALIBRATE

bull 2D COUPLED INTERACTION SURFACE

- SIGNIFICANT MODELING IMPROVEMENT

- DATAAVAILABLE

- IMPLEMENTATION PRACTICAL BUT NEEDS DEVELOPMENT AND CALIBRATION

bull 2D UNCOUPLED INTERACTION SURFACE

- SIGNIFICANT MODELING IMPROVEMENT

- DATA AVAILABLE

- IMPLEMENTATION PRACTICAL AND AVAILABLE

ANDREW PHASE II

JOINT MODEJJNG OPTIONS ILLUSTRATIONS

P

e

ANDREW PHASE II

JOINT MODELING PHASE II METHOD

e

I I I

ANDREW PHASE II

ANDREW STORM CONDmONS UPDATED

bull -PHASE I USED OCEANWEATHER HINDCAST-(Nov 1992)

bull middotmiddotPHASE II WILL USE THE REVISED HINDCAST (Nov 1994)

bull lt c lt I lt

2-2

l---4-~~~+-~~-+-~~~1--~~-+-~~--l--~T--+-~~--i24

f-tJVtMfiz~ 1Cfl (r STvb(

Plollol bullbull 5middotNOYmiddot9214bull1J Irbullbull Illbull [ANDREVl~NDREVCUSTbullMAXll 5-NOY-1992 10bull54 - 9208

HEIGHT lml

I~ MAXI

I L 1~middotamp SZ~- ~A I 13c

I J) ( r I X~0iWJ~0gtJ gt L gt gt I 1c I 120

-- N

~ deg

I I I I I I e C I ~=-41 122 -94 -92 -90 -88 -86 -84 -82 -80

-

Figure 7 contours of maximum hindcast significant wave height

_

~

~tl ~

NOltNttIJ 17lt ltNS S iJJ)y

Maximum Significant Wave Height - Hurricane Andrew 1992

30 ~

26 ~ I ~-~~~- == -shy -~ 35-_

24deg1--shy --------~- middot-middot-------middotmiddotmiddot----middot------middot-middot~middot- - - ~-----1---middot

22middot ~~~~~~~-~ -94 -92 -90 -88 -84deg-86 -82 -somiddot

c

-

Platform Selection bullbullbull1111111111111

0 BasisCriteria for Selection middot

0 Review of Phase I Platforms middot 0 Alternative Platforms

0 PMB Recommendations

0 Participants Comments

gt

L

l I

BasisCriteria for Selection bullbullbull1111111111111

0 Benefit for Calibration - Classification (UnexpectedExpected)

- Failure Mode (StructuresFoundations)

- Damaged StructuresKnown response

- Better Representation of Full ~opulation

0 Availability of Data - Existing Models

- Inspection Data

- Damage Data

PREDICTED BEHAVIOR

-

OBSERVED

BEHAVIOR

-SURVIVED DAMAGED FAILED

SURVIVED

ST134W WD90A

-MC311 -

MC397 -

--

-

-

ST151K ST130Q

-

-

-

DAMAGED - -

-

-

--

-

-

ST52 SS139 ST177B -

ST161A

-_

-

FAILED

-

-

-

ST151H ST130A ST72

Operator - Platfonns - Platfonns Survived Damagedamp

- Falled -

bull

-

-

Jle eBDIGIPllDbl -Amocoshy 30 1 Chevron -

143 12 Exxon 72 Mobll 22 1 Phllllps 7 bull

-Shell 39 Unocal 25 1

Total Partlclnts Platfonns 338 15

Other Operators bull- 310 13

-

Total Platfonns - -

648 28

fbml

Platforms Considered 6 7

tie of Total Platfonns 1 25

OVERVIEW OF PLATFORMS SURVIVEDDAMAGEDFAILED

-

shy

v

Platforms Evaluated in Phase I

-

-Platfonn Water Year Number Analysis Avallabillty of Availability Platform

- Name Depth rl of Model DetaDed rl ClassUlcatlon

Installation legs Inspection Soll Report Based on Information Phase I Calibration - -ft

- -- Suaill flalfnrm ~BSi -STISIK 137 1963 8 CAP Yes Yes Un eel Sarvlval

-STIJOO 170 1964 4 CAPmiddotshy Yes ST134data u ST134W 137 1981 4 CAP Yes Yesmiddot E

-

edSarvlval -

middotsurvival

WD90A 184 - 1964 8 CAP Yes E -

- Sarvlvill

MC3ll 343 1978 8 KARMA

MC397 468 1991 4 KARMA -

bull ~reglmmiddot- -

Yes Sure Survival

bull Yes - Sare Sanlval-

-T23STS2 63 1969 4 CAP

T2SSS139 62 1969 4 CAP -

Yes

Yes

Yes Ex

Yes Une

bull SurvivalLikely to DamalI

middot bull SanlvalLlkely to DamalI

ST161A 118 1964 8 USFOS

Eail11a pound1al(bullllm Clss ST177B 142 --1965 8 CAP

STISIH 137 1964 - 8 CAP C

STIJOA 140 19S8 8 CAP - -

Yes

Yes

ollamed

Yes

- Yes E

-

at about 1 mUe

Yes

ST134data

bull SafvlvalLlkelv to Dam- bull- -

E middotrahre

Emeried Failure -

Likely FaUare

T21ST72 61 1969 4 CAP Collamed No Likely FaUure

Platforms Identified for Evaluating Their Likelihood of Falling Under UnexpeCted Survival Category

CllEllATOa

AllB

llLOCll

Sl1lvcnJllE

NAME

YEAR

INSTAUJCD

WATER

IJBPJB

middotmiddotDISTANCE

ftOMSllOllE

(MIUIS)

Ill

(M)

AMOCO BllW

STllI

WDI075

A

bull D

4

n uo 17J

44

32

bull 20

10

bull CllEVllOH STtllO c I 1bull 21 11

STl130 D 0 161 21 11 STllJ4 I 0 137 D - 11

STIUI G I 137 32 11

STIUI I 0 121 32 11

STIUI

J 0 140 32 11

STIUI It 4 137 32 11

STllSI PROl)I 0 137 32 11

STllSI PROD-2 0 137 32 11

STll7 A 0

140 35 11

WDl117 c 65 214 34 10

DXON STshy E 51 14 bull STl165 A t3 bull 34 WD0073

WD0073

WD0073

A _ c

4

62

4

IA IA

17J

22

22

22

bullbullbull WD003-shy E 65 161 Z7

A 0 IN 31 MOBD Bl12I

ssn A bull

50

50

31

7

bull 7

MIJllPllY SS114 B 7 50 7 7

PlllUIPS SSl14t A 55 33 7

SAMEDAN ST17J A 4 117 7 bull 7

SJIELL WD0103 A 65 223 27 WDllOJ bull 65 221 27 WDll04 c 65 221 27 WDOI04 D 67 2 27

UNOCAL

SSl20I

SSl20I

bull SSl20I

SS0209

SS0209

SS0209

SS0209

SS 0253

SS 0253

E

I

PtJllE

A

bull D

G A

c

0

4

4

7J

61

0

0

69

103

97

97

9S

100

95

100

165

175

34

34

34

34

34

34

34

54

54

bullbullbullbullbullbullbullbullbull

Platforms Recommended for Phase II

llalr y_ w- Aot Av- _ I 1 Doplla Model nm bull bull _ bull I I H11 s18-1 Irht lbwl~ bullbullbull(-illtdlel OM lllMtD

II ---

-STISIIC 137 1963 I CAP Yeo Yeo I Yeo

STtshy -170 1964 4 CAP Yeo ST134dlU I Yeo

STl34W 137 Yeo Yeo -1981 4 CAP D Yeo

WD90A 114 1964 I CAP Yeo - D

MC311 30 1971 I KARMA Yeo -sre-

MC397 middot 461 1991 4 ICARMA Yeo Sireshy - - r- -

__ T13ST5l 63 1969 4 CAP Yeo Yeobull middot- ~n- D Yeo _ Tl5SSl39 62 1969 4 CAP Yeo Yeo I Yeo

_ STIT711 142 1965 I CAP Yeo I Yeo middot-1shySTl61A Ill 1964 I USFOS Yeo Yeo - D bulloshy

- r-

STISIH 137 1964 I CAP y D y -_ STl31A 140 1951 I CAP Yeo STl34dlU I Yeo

T21Sll 61 1969 4 CAP No rar-middot I Yeo

~-rmiddot-middot SS209A 95 1967 16 CAP Yeo D -SSU4R

11 UI A 41 -in D SS 2741AIM CAP bull

_ _ -shy - -shy- -shy

CALIBRATION

bull PHASEIAPPROACH

bull CALIBRATION ISSUES

bull PHASE II WORK SCOPE

bull SOFlWARE AVAILABILITY

bull EFFECT OF ADDITIONAL PLATFORMS

bull WEIGHTING APPROACH

bull EXAMPLE APPLICATION middot

(l=Ofce Mode] Wind

~

fcurrent l -middot

~SW ~leldl4

ultimate resistance force I first

design load Illgt

i _

deflection

-

bull

4

Historical Hurrlcane Dariiage and Survival of Platforms

Components Required for Calibration (Petrauskas 1992)

PRIOR Distribution

PLATFORM EXPERIENCE

BAYESIAN Survivals and Failures

UPDATING

Waves Winds and Currents bull

POSTERIOR Distribution

- A_middotJA ~ Survival Data Failure Data

Result Result

Load Resistance

middotfl

-

Bayesian Updaing middotGeneral Description (Petrauskas 1992)

L Imiddot

WHERE TO INTRODUCE BIAS FACTOR (B

On Resistance (Rmue - B (RPREDICreD I

middotOn Loads (Smue middot - B (SPRE01cTEo

On Safety Margin (RSmue - B (RSPREDICTED

CALIBRATION METHODOLOGY

gt

r

FIELD OBSERVATIONS

Observed Behavior

survivals Damages

Failures

r

bull

CAPACITY ANALYSIS

Predicted Load and Resistance Predicted Failure Modes

Survival Damage Failure cases

RELIABILITY ANALYSIS

- Probability of Failure ~

Ukellhood Failure

BAYESIAN UPDATING

Prior Distnbution of B Combined Likelihood Function

Posterior Dlstnbution of B

BIAS FACTOR

Mean Value cov bull

Represents Modeling Uncertainties

Prior Distribution of B Likelihood Functionmiddot Observing Success case STISIK and Posterior Distribution of B

2 00

uo

160

140

120

e 100 L $ uoIshy

0 60 ~ ~ 040

020

0 00

0 00 0 50

_ I 00

b

I 50 2 00

-shy lk [bl STISIK)

--shy Pnor-f(b)

--shy Pos1enor r(bl ISIK)

Prior or B Mean bull 10 COV bull 030

Posterlor of Bbull Meanbull 136COV 0 16

2 50

Likelihood Function and Posterior Distribution Example Platform STISIK

- -

~

- Prior and Posterior Distributions or B Failure cases (ST177B STlSlH ST130A T21) Damage cases (T23 T25 ST161A) Success cases (STlSlK ST130Q ST134W WD90A MC311 MC397)

400

350

300

2 so 0-

I

bull bull

bull bull gt

- bullbull ~ bull bull I bull

_ bull bull bullbull I bull bull bullII V -

bull bull

I bull ~ bull

bullbull bullI bull bullI bull bull bull

middot -~

bull__ ~ middotI middotmiddot

L 2 00

e ~ I SO

-I 00

- 0 so

000 gt

0 00 020 0 40 060 0 80 I 00 120 140 160 110 200

b -

-

shy

Prior-r

- - - - Posterior fLlll failure cases

bull bull bull bull bull bull bull Posterior flall damage cases

- bull bull - bull middot Postenor flall success cases

Posterior flall success fulure damage cases

-

Posterior of B Mean= 119 COV 0101

Posterior Distribution or Blas Factor (8) bull Different Categories and All Cases Combined

shy

l

PHASE I - KEY CONCLUSIONS

bull GLOBAL CORRECTION FACTOR

bull 15 TO 20 CONSERVATISM MEAN)

bull UNEXPECTED SURVIVALS HAVE GREATEST IMPACT

bull DAMAGED CASES MORE INFORMATIVE

bull CALIBRATION ISSUES PHASE II

bull

EFFECT OF IMPROVED CAPACITY ANALYSIS RECIPE

bull APPLICABILITY TO MULTIPLE FAILURE MODES

bull NO FOUNDATION DAMAGE CASE INCLUDED

bull IMPROVED REPRESENTATION OF FULL ANDREW DISTRIBUTION

-bull NUMBER OF PLATFORMS CONSIDERED

I 1 I

I t ) bull ~

PHASE II WORK SCOPE (CALIBRATION)

bull

bull ESTABLISH MULTIPLE BIAS FACTORS

Use Nine Base Case Platforms

middot - Capacity Analysis for Revised recipe

Foundation Blas Factor from Foundation JIP

bull DEVELOP WEIGHTING APPROACH

bull UPDATE MULTIPLE BIAS FACTORS

bullEffect of Platforms Not In the Nine Base Cases

bull EXAMPLE APPLICATION OF BIAS FACTORS I

bullbull

PHASE I PHASE II

CAPACITYbull ANALYSISRu bull

---------------------1----middot

Jc1 - iJbullraquoJ dbull

FttlS frt[f~l ratdhJJ

-

LIKELIHOOD FUNCTION

DEVELOPMENT

lk Jb Jfall1N) bull Jb)

111 Ill Isurvtvoq bull 1 - bl

-middot -------------------shy ~---------------------middot-middot

r111ir8 Cb1 111 Cbl BAYESIAN

UPDATING

~cove

R1R2 bull middot---------------------~----

Jblbull P(g cOJ

1 bull JbRI (MBSI I bull 1

~middot

llcJllJfalluro) bull P(gcOJ Ill 111 b21fallun) bull P ((g1 CO) U 1112 c OD

middot--------------------shy ----------------------shy bull

r111111r8 Cbl 111 Ill i-1nro1

deg ~1b2)laquottfrr (b1b21-lnfo) re Ill bull 1-11middot b2) bull lodegamp

middot

i1 covamp i2 cov~i

-

shy

--

Likelihood Function vs b All Success Case STISI K STl30Q STl34W WD90A MC311 MC397

I 0000

0 9000

OIOOO

0 7000

~ 0amp000

ta osoOO

~ 0 4000

0 3000

0 2000

0 1000

bull 0 0000

bull

I

middot

middot bull I

Ibull

-sibull ~ -middot-middot-middot-middot -middot-

----7----+-----~ I

Ibull bull bull I

rbull

--- Jk ISIK[blsuccess)

- bull - bull - lk 130Q[blsuccess) bull

lk 134W[blsuccess)

Jk-WD90A(blsuccc~I

- - - lk MC397[blsucccss)

--ltgt-- lk middotMC31 l[blsucccss)

--- Jk [bl 6 out of6succcsHascs)

05 I 5 2 250

b

Likelihood Functionsmiddot Success Cases - Phbullsc L

shy

bullbull bull bull bull bull bull -------------------~

bull bull bullbull bullbull bullbull bullbull bullbull bullbull bull

bull bullbull

bullbullbullbull

bullbull

bullbullbullbullbullbullbull

bullbullbullbullbullbull

bullbullbullbullbullbullbull

z 0

t z ) LL

c 0 0c i ~ i

J

~ sect ~--------------~ e l5

I

I

SOFTWARE

bull PF PROGRAM

bull CALREL

bull PF +CALREL

bull RELACS

bull

gt i

l

SENSITIVITY STUDIES

POSTERIOR DISTRIBUTION OF B

bull EFFECT OF CHANGE IN DISTRIBUTION OF PRIOR OF B

bull EFFECT OF REMOVAL OF ONE PLATFORM FROM EACH CATEGORY

bull EFFECT OF REMOVAL OF THREE DAMAGE CASES

bull EFFECT OF ADDITIONAL 13 SIMILAR PLATFORMS

-I

Prior and Posterior Distributions of B Variation In COV of PriorofB fromOl to04 -- -

4500

-4000

-

- -

3500 --middot - bull bull - bull bull Prior-r ICOV =02

3000

- - - - Prior-r ICOV =03 -

bull -~ e 2soo -

bull bull bull bull bull bull bull Pnor-r ICOV =o 4t frJ ~ 2000 Posterior rlCOV =0 2

bull bull Postenor rlCOV =0 3 bull bull 1500 bull middot bull - shy -middot- Posterior rlCOV =04

middot~ I 000 - bull

-middotmiddot ~ ~middotmiddot M -~

- middot 0 500

wbullbullbull _ I middot~ middotmiddotmiddotmiddotmiddot middotmiddotmiddotmiddotmiddot - -~ ~ __ middot_ _ bull -shy0 000

0 000 0200 0 400 0 600 0 800 IOOO 1200 1400 1600 1800 2000 -

- - b

-

=shy

Posterior Distribution of Bias Factor (B) bull Effect of Variation In COV of Prior of B

bull

bullPrior and Posterior Distributions of B Evaluating Effect of Removal of One Platform from Each Category from the Base Case of 13 Platforms

400

~ ~ 350

j IA i

3 00 bull I bull I ---Prior-r I bull l

250 1 I 1

I bull bull ---Posterior r lall 13 casesshy- ~E 2 00 bull bull bull bull bull bull bull Posterior r without STISIK

- ~ 1 bull - bull I - bull bull - bull bull Posterior r Iwithout T2I I 50 I

J-tt~i - bull - bull - Posterior r Iwithout T2S

tOO [-

I I ~ ~ I 0 50

0 00

___ r J i

l~T

~--

0 00 0 20 0 40 0 60 080 I 00 120 140 160 110 2 00

b

zshy

Posterior Distribution of Bias Factor (8) bull Effect of Removal of middotPlatform from Any Category

bullPrior and Posterior Distributions or B Evaluating Effect or Removal or 3 Damage Cases from the Base Case or 13 Platrorms

4

35

3

25

e L 2 ~ bulle shy

I 15

bull I

05

0

0 02 04 06 I 2 I 4 16 IS 2

I I

I

I

I Ibull

J -tmiddot L--

bullI

~

I

fl

II ~

I I v I

middot bull bull

bull L

~ ll- I-~lt

OS I

b

I

bull

1middot bullbullbullbull

~ __

bull

- - - - Prior-r

- - - - Postenor flall failure cases

Postenor flall success cases

--- Posterior flall 13 cases

---Posterior fI success failure cases only

IPosterior or B Mean =t11 COV =0121

-

_I

Posterir Distribution or Bias Factor (B) bull Effect or Removal or 3-Damage Cases

-

bullPrior and Posterior Distributions or B Evaluating Effect or Additional 13 Similar Observations as

Used in Andrew JIP The POsterior or B from Andrew-JIP Considered as New Prior or B

- 500

450

400

350

bull bull

bull bull bull bull J bull

bull bull f ~ I

I ~ Ibull bull

bullbull bull

l bull I -bullbull bull bull

I

I bull bull

bull I I~~bull bull bull

bull ~

I bull bull

~ ~_ ~~~

- - New Prior r bull

Posterior 114 new failure cases ~ 300- bull bull bull bull bull bull bull Posterior f 13 new damage casesl 150 ~

- bull bull - bull bull Posterior 116 new success cases i 2 00

Postenor 1113 new success failure -150 damage cases

I 00

0 50 -

0 00 0 00 0 20 0 40 0 60 0 80 I 00 120 I 40 160 110 2 00

Posterior ol B Mean= 120 COV= 0071 b

shy

shy

_rshy

Posterior Distribution of Bias Factor (B) bull Effect of Additional 13-Similar Observations

1

1 I

WEIGHTING APPROACH bull

bull ASSESSMENT OF PLATFORMS DATA

Select Platforms Most Mappable to the Analyzed Platforms

Unlikely SurvivalDamage Cases Identification

Based on Cut-Off Criteria Agemiddot Hs etc

bull INFORMATION REQUIRED FOR COMPARISON

bull CATEGORIZATION OF ADDITIONALmiddot PLATFORMS

bull UPDATING OF CALIBRATION RESULTS

--

OVERVIEW OF PLATFORMS SURVIVEDDAMAGED -~

Operator -

- Platforms

Survived CS)

Platforms

Damaged CD)

Platforms Used In

Phase IJIP -

JI~ ~Bill~IRBDll -

Amoco -

Chevron -

Exxon

Mob II

Phllllps

Shell

Unocal -

30

143

72 22

7

39 25shy -

1

12

bull

1 bull

1

1(S)+1 (D)

3 (S) +3(D)

2 (S)

bull shy

bull

bull

bull

Total Partlclnants Platforms 338 - 15 10C6S+4D) -

Other Operators

310

13

3 (D)

Total Platforms 648 28 13C6S+7D) -

Summary of Andrew JIP bull Phase I Platforms

Plairshy w v- N-ber Lea-le 1 shy car Hind_ o-w N-bull Deplh fl fl IJpe la bull n Yeadon Jadtet lbelledtf liefalit -- _ -OD

IL H - -shy _ICllshy IL IL - - rQClto ~ _

K STUIK 137 1963 8 365 608S4 Cellardeck

-middot STl-shy 170 1964 4 -~ K 39 62269 Cellardeck - --ST134W 137 1911 4 43 608S4 Smvlnlshy- _ - -

WD90A 184 1964 8 K 267Sbull34 83 SOS31 SUboftllar deck SmvMI

MC311 343 1978 8 x S7 62 - MC397 468 1991 4 x shy S33bull6SO 65911 -shySmeSmvlvol -- gt -

r-- T23ST52 63 1969 4 ton- shyKshy 52594 S03 -T25SSl39 62 1969 4 Kshy SOS94 S063 - middot ton-shy__ -

bullbull a ton--shyST161A 118 1964 8 K - 347S 51495 Cellarclecti

r~

ST1771 142 1965 8 K 39 60167 CeDordeck failme

STISIH 137 1964 8 K 395 608S4 CeDardeck Failme

K ST130A 140 1958 a SS 6095 I lblv Pai1uno - T21ST72 61 1969 4 Kshy S4S94 4974 I lhlv Pai1uno -

- -

JE

INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON

bull WATER DEPTH

bull bull YEAR OF INSTALLATIONDESIGN

bull PLATFORM ORIENTATION

bull PLATFORM LATITUDE AND LONGITUDE

bull KEY SKETCHES OF PLATFORM STRUCTURAL FRAMING

Number of Legs Plies Leg Batter Number of Conductors bull Deck Elevations (Including sub-cellar deck If any) Jacket Framing Elevations Brace type In Vertical Frames Sizes of Primary Members (leg pile vertical frames braces primary horizontals) Joint Details (can gap etc) LegPiie Annulus Grouting Status

bull middotPOST ANDREW INSPECTION REPORT Significant Wave Height Hs (ft)

bull INFORMATION ON DAMAGES PRIOR TO ANDREW

bull RESULTS FROM IN-HOUSE STUDIES IF AVAILABLE

bull ~RIGINAL DESIGN CRITERIA INFORMATION IF AVAILABLE

bull SOIL SHEAR STRENGTH PROFILE IF AVAILABLE

Platforms Identified for Evaluating Therr Likelihood of Falling Under Unexpected Survival Category

OPlRATOR

middotADA

BLOClt

llmucnJllE

NAME

YEAll

INSTAUJD

WA1ER

DBPTll

DISTANCE

JlllOM SHOllE

(MILU)

Bl

(M)

AMOCO 111113

STl11

WDI075

A

bull D

St

17

120

1n

44

20

10 I

CllllVllQfl STlllO c 61 110 21 11 STIUO D Q 1110 21 11

STl134 p Q 137 29 11

ST0151 G 61 137 11

ST0151 I 3 121 11

ST0151 J Q 140 11

STl151 It 137 11

ST0151 PJl001 Q UT 11

ST0151 PROD-2 Q UT 11

ST017 A 3 140 35 11

WDIUT c 65 214 34 10

llXXON STI055 II 51 6T 14 I

STl15 A t3 34

WDI073

WD0073

Amiddotshy Q

1111 1111

22

22

I

I

WDI073 c 1n 22 I

- WDOCl93-shy II

65 1110 2T A 3 194 31 MOBU 111121

SS IOl7

A B

501

50

50

31

T

I

T 6

MtlllPllY SS114 B T 50 T T

PlllllIPS SS0149 A 5 55 33 T

SAMllDAN STln A 107 T T

SBllLL WDl103 A 65 223 2T WDl103 B 65 221 2T WD0104 c 65 221 2T WD0104 D 6T 20 2T

UNOCAL SS020I

SS 020I SSl20I

SS 0209

SS0209

II p

PURll

A

B

3

n 61

103

97

97

95

100

34

34

34

34 34

bullbull bullbullbull

- SS0209 D u 95 34 I

- SS 1209 SSl253

SS0253

G A

c

6T

2

100

165

1T5

34

54

54

bullbull I

I I

I I

SOUTH MARCH

~LAND

N

SOUTH PaTOI ENE ISWltD AREA

MMS Defined Limits ror Offshore Structures Requiring Inspections Following Andrew

t---Jbulln3 T25bull I bullTtl

bull1$1Htl1Abull bull1SIK1779bull

~

middot-DaTA AREA~T

CRANO rsLE

AREA bull1300

bullll4W

=shy

bullPath of Hurricane Andrew with Platforms Used In Calibration

Likelihood Function vs b All Success Case STISI K STIJOQ ST134W WD90A MCJll MCJ97

I 0000

0 9000

0 8000

07000

-Bo6000

tiiii 0 5000 -

04000

0 3000

0 2000

01000

0 0000

0 05 25

ri1rt middot-middot-middot-middot-middot-

I 5 2

b

---11 middot151K(blsu~ccsI

- bull - bull - fk 130Q(bfU~lCSS)

lie bull134W(blsuccc]

lk-WD90A[blsultCS]

- - - lk middotMC397(bfUtlCS]

--o-- lk-MCJll(bbuctcsJ

--- ll (bl 6 out of 6 Mlttcs case]

_

( Likelihood Functions middotSuccess Cases - p - 1 c I

CATEGORIZATION OF ADDITIONAL PLATFORMS

I I I

bull CATEGORIES

Unexpected Survivals

Expected Survivals

bull APPROACH

Identify Corresponding Platform(s) Evaluated In the JIP

Determine Max Wave Height at Platform Location

Evaluate Whether Wave Hit the Deck

middot - Evaluate Likelihood of R and S to Differ from the Base Cases

Review Joint Information

Assess Special Features

bull IDENTIFY NUMBER OF PLATFORMS IN EACH CATEGORY

I

UPDATING OF CALIBRATION RESULTS

bull INCLUDE EFFECT OF ADDITIONAL PLATFORMS ON bullBbull

Consider Increased 1Uncertalntles In R and S

Develop bullNeW Ukellhood Functions for Different Categories

middotLet bullxbull Number of Ukely Unexpected Survival Cases and y Number of Ukely Expected Survival Cases

Determine Combined Likelihood Function lk (b Iadditional platforms) middot

bull REVISED LIKELIHOOD FUNCTION

lk(b I all cases) = lk(b I 9 platforms) middot x lk(b Ix platforms) x lk(b I y platforms)

bull ESTABLISH REVISED BIAS FACTOR B

f8 (b) u f8 (b) lk (b I all cases)

II

t t ~

shy

APPLICATIONS OF BIAS FACTORS

ii APPLICABLE TO PARTICULAR LOAD AND RESISTANCE RECIPE

bull USE BY THE API COMMITTEES AND MMS

Criteria Revisions

Recommended Design Guidelines Revisions bull

bullbull REASSESSMENT STUDIES AND DECISION MAKING

middot

bull RISK ASSESSMENT STUDIES BY THE OPERATORS

Individual Platforms

Fleet of Platforms

middot IMR Program Implementation

bullbull NOT TO BE USED IN DESIGN

I I fl J

EXAMPLE USE OF BIAS FACTORS

bull PHASE I EXAMPLE APPLICATION

bull EVALUATION OF ANNUAL PROBABILITY OF FAILURE

Establish Annual Maximum Seastate Data

Determine Coefficients to Define Base Shear

Determine Ultimate Capacity of Platform

Determine Probablllty of Failure Versus bullbbull

-bull DETERMINE UPDATED PROBABILITY OF FAILURE

Pf - P [S gt R]

Pf I P [~ gt Rb I b] f 8 (b) db

Pr - J P [S gt Rb I b] f bull 8 (b) db

bull TOTAL PROBABILITIES OF FAILURE

Total Pf - 1 bull Il [1 middotPf] - l Pf bull

Total Pf - 1 bull Il [1middot Pf] - l Pfmiddot

Total Pr = 1 - n [1middot PfbullJ = l Pfshy

I I I

EXAMPLE APPLICATION bull PHASE I

PLATFORM ST 151 K

bull GULF OF MEXICO

bull WATER DEPTH = 137 FT

bull ANNUAL MAXIMUM HS MEDIAN 175 ft COV 031

bull ASSUMED STORM DURATION 3 hoursmiddot

bull ASSUMED NUMBER OF WAVES 800

bull MEDIAN HS FOR STORM HOURS 1 2 AND 3

bull MEDIAN CURRENT VALUE bull

bull DETERMINE ULTIMATE CAPACITY FOR THREE DIRECTIONS

bull DETERMINE ANNUAL FAILURE PROBABILITIES

Broadsidemiddot Diagonal End-on

0037 0025 0007

0071 0044 0018

0022 0016 0004

DETERMINE TOTAL ANNUAL PROBABILITY OF FAILURE

Pf - 007

Pf - 013

Pr - 004

bull

EXAMPLE USE OF BIAS FACTORS - PHASE II middot

bull DETERMINE PREDICTIVE PROBABILITY OF FAILURE

Case - 1

Pf _ S S Pf (b b2) fa1a2 (b b2) db db2

g bull -1 (b1 R I S) - 1

92 - (b2 R2 I S) - 1

Case -2

Pf s ~ [bJ f11 (b) db

Pf [bJ p [g lt 0)

g - bull 1 (b1 R1 I S) - 1

1

bull1) l

I

BENEFITS OF ANDREW JIP bull PHASE II

lt

bull IMPROVED LOAD AND RESISTANCE RECIPE

bull IMPROVED CALIBRATION PROCEDURE

bull MULTIPLE BIAS FACTORS

bull PROCEDURE TO INCLUDE EFFECT OF OTHER PLATFORMS

bull PROCEDURE TO USE MULTIPLE BIAS FACTORS

  • Structure Bookmarks
    • ANDREW PHASE II JOINT1 INDUSTRY PROJECT
    • PROGRESS MEETING NUMBER 1
    • ANDREW PHASE II JOINT INDUSTRY PROJECT PROGRESS MEETING NUMBER 1 NOVEMBER 16 AGENDA
    • ANDREW PHASE II UST OF PARTICIPANTS
    • ANDREW PHASE I OBJECTIVES AND RESULTS
    • ANDREW PHASE II PROJECT OBJECTIVES
    • SUBSEQUENT PROJECTS
    • ANDREW PHASE II PROJECT TEAM
    • ANDREW PHASE II UPDATE MODELING middotAND ANALYSIS RECIPE
    • -ANDREW PHASE II CONDUCTOR RESISTANCE
    • ANDREW PHASE Il WA VE FORCE PROFILE
    • ANDREW PHASE II WA VE FORCE PROFILEmiddot
    • ANDREW PHASE II BRACE MODELING
    • ANDREW PHASE II FOUNDATIONS
    • ANDREW PHASE II middot middot JOINT MODELING
    • ANDREW PHASE Il JOINT MODELING OPTIONS DISCUSSION
    • ANDREW PHASE II JOINT MODEJJNG OPTIONS ILLUSTRATIONS
    • ANDREW PHASE II JOINT MODELING PHASE II METHOD
    • ANDREW PHASE II ANDREW STORM CONDmONS UPDATED
    • Platform Selection bullbullbull1111111111111
    • BasisCriteria for Selection bullbullbull1111111111111
    • CALIBRATION
    • WHERE TO INTRODUCE BIAS FACTOR (B
    • PHASE I -KEY CONCLUSIONS
    • bull CALIBRATION ISSUES PHASE II
    • PHASE II WORK SCOPE (CALIBRATION)
    • SOFTWARE
    • SENSITIVITY STUDIES POSTERIOR DISTRIBUTION OF B
    • WEIGHTING APPROACH bull
    • INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON
    • CATEGORIZATION OF ADDITIONAL PLATFORMS
    • UPDATING OF CALIBRATION RESULTS
    • APPLICATIONS OF BIAS FACTORS
    • EXAMPLE USE OF BIAS FACTORS
    • EXAMPLE APPLICATION bull PHASE I PLATFORM ST 151 K
    • EXAMPLE USE OF BIAS FACTORS -PHASE II middot
    • BENEFITS OF ANDREW JIP bull PHASE II
Page 24:  · . ' ANDREW PHASE II PROJECT OBJECTIVES . ' • ·UPDATE MODELING AND ANALYSIS RECIPE· JOINTS ' ' ·BRACES WAVE PROFILE FOUNDATIONS . ' ' CONDUCTORS· • ANALVZE PLATFORMS USING

--

Base Shear vs Wave Height ST151K

6000

5000 1---------Ult1mateCapacitybull 4890 kip (100 Improved Soll)-----------~--

4000a a lii ~ 3000

81 ampI

2000

1000

0 +-~~~--i~~~~~~~~~-+-~~~--lf--~~~~~~~~-+-~~~--l

30 35 40 45 50 55 60 65

Wave Height (ft)

bull middot

~

bull

II

2

g

~

It)

0

cCD E 0 E a c E euro~ CDshygtItgtmiddoto ~ti c -a Gic CD gt ~

-= c -Cll

I z

bullO 0i i ~ g 0

() CI -bull(q-ll) iuawow llu1wniiaAO

11

shy

116

114

112

g 110

bullc = 108

~ ~ 106 -sect 104

102

100

98

96

-

6555 60

Wave Height vs Center of Action ST151K

30 35 40 45 50

Wave Height (ft)

t

Comparison of Pushover Curves for ST 151K Variable Load Pattern and Monotonic Load Pattern (Andrew)

Soll Improved 100

--- --------- --- ----6000

5000

- 4000 8 - I

1 3000 Ul

81 m 2000

_J -shy

1000

---Variable Load

- - Andrew Load

0 f-~~~~~r--~~~~-t-~~~~~-t-~~~~~+-~~~~----t~~~~~-1

0 1 - 2 3 4 5 6

Deck Displacement (ft)

-middotshy

1 t I I ~

ANDREW PHASE II

BRACE MODELING

bull MARSHALL STRUTS USED FORPHASE I

bull PHASE II WILL ASSESS DIFFERENCE IN USING BUCKLING BEAMS AND MARSHALL STRUTS

middot bull COMPARE FOR 2D FRAME

bull COMPARE FOR PLATFORM 177 WHERE LOWER

BAY EXHIBITS LARGE BRACE MOMENTS

Project andrew Model beamframe Version 1 Mon Nov 14 145447 1994

~

middot Frame Test buckling beam-column model

-

Project andrew Model strutframe Version 1 Mon Nov 14 145754 1994

bull

Frame Test strut model

middot-middot _____________--_

_middotmiddotr_~ --ll--~-middot__

Buckling Beam bull

Strut (K=O 65)

Lateral Displacement

ANDREW PHASE II

FOUNDATIONS

bull ~LICIT NONLINEAR API FOUNDATION USED INiPHASE I

bull FOUNDATIONS

JIP SELECTED MINI-VANE DATA FOR

STRENGTH PROFILE

bull COMMENTS ON BENCHMARK COMPARISONS

- SIGNIFICANT VARIABILITY DUE TO FOUNDATION MODELING METHODS

II Ii

z

ctV 4x SPlO caissonfy-42s soil-api(lf-097step-114)

Inelastic Events Legend Elastic Strut Buckling

Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield

--------shyBeam Clmn Fully Plastic Fracture

~--~~~~~~~~~--~----~__i~~---~~~~__--~

Pro)ect apipile Model South Pelto 10 Version Wed May 18072235 1994 - -

1bullt

+

t

f i bullbullbull

bullbull bullbull

I I

5

~x SPlO Caissonfy=42s soil=api(lf=lOOstep=137)

Inelastic Events Legend

Elastic Strut Bucklilg

Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield

earn Clmn Full~ Plas1c Fracture

ANDREW PHASE II middot middot

JOINT MODELING

middotbull API RIGID-PLASTIC USED IN PHASE I

a ELASTIC-BRITTLE USED IN PHASE IB

amiddot MSL WILL PROVIDE NONLINEAR FORCE-DEFORMATION

AND MOMENT-DEFORMATION RELATIONS FOR PHASE II

a MODELING OPTIONS FOR PHASE II -

l

- UNCO~LEDmiddotONE DIMENSIONAL MEMBER END JOINT MODELING

J Ii

ANDREW PHASE Il

JOINT MODELING OPTIONS DISCUSSION

bull JD COUPLED INTERACTION SURFACE

- IDEAL

- DATA NOT AVAILABLE TO CALIBRATE

bull 2D COUPLED INTERACTION SURFACE

- SIGNIFICANT MODELING IMPROVEMENT

- DATAAVAILABLE

- IMPLEMENTATION PRACTICAL BUT NEEDS DEVELOPMENT AND CALIBRATION

bull 2D UNCOUPLED INTERACTION SURFACE

- SIGNIFICANT MODELING IMPROVEMENT

- DATA AVAILABLE

- IMPLEMENTATION PRACTICAL AND AVAILABLE

ANDREW PHASE II

JOINT MODEJJNG OPTIONS ILLUSTRATIONS

P

e

ANDREW PHASE II

JOINT MODELING PHASE II METHOD

e

I I I

ANDREW PHASE II

ANDREW STORM CONDmONS UPDATED

bull -PHASE I USED OCEANWEATHER HINDCAST-(Nov 1992)

bull middotmiddotPHASE II WILL USE THE REVISED HINDCAST (Nov 1994)

bull lt c lt I lt

2-2

l---4-~~~+-~~-+-~~~1--~~-+-~~--l--~T--+-~~--i24

f-tJVtMfiz~ 1Cfl (r STvb(

Plollol bullbull 5middotNOYmiddot9214bull1J Irbullbull Illbull [ANDREVl~NDREVCUSTbullMAXll 5-NOY-1992 10bull54 - 9208

HEIGHT lml

I~ MAXI

I L 1~middotamp SZ~- ~A I 13c

I J) ( r I X~0iWJ~0gtJ gt L gt gt I 1c I 120

-- N

~ deg

I I I I I I e C I ~=-41 122 -94 -92 -90 -88 -86 -84 -82 -80

-

Figure 7 contours of maximum hindcast significant wave height

_

~

~tl ~

NOltNttIJ 17lt ltNS S iJJ)y

Maximum Significant Wave Height - Hurricane Andrew 1992

30 ~

26 ~ I ~-~~~- == -shy -~ 35-_

24deg1--shy --------~- middot-middot-------middotmiddotmiddot----middot------middot-middot~middot- - - ~-----1---middot

22middot ~~~~~~~-~ -94 -92 -90 -88 -84deg-86 -82 -somiddot

c

-

Platform Selection bullbullbull1111111111111

0 BasisCriteria for Selection middot

0 Review of Phase I Platforms middot 0 Alternative Platforms

0 PMB Recommendations

0 Participants Comments

gt

L

l I

BasisCriteria for Selection bullbullbull1111111111111

0 Benefit for Calibration - Classification (UnexpectedExpected)

- Failure Mode (StructuresFoundations)

- Damaged StructuresKnown response

- Better Representation of Full ~opulation

0 Availability of Data - Existing Models

- Inspection Data

- Damage Data

PREDICTED BEHAVIOR

-

OBSERVED

BEHAVIOR

-SURVIVED DAMAGED FAILED

SURVIVED

ST134W WD90A

-MC311 -

MC397 -

--

-

-

ST151K ST130Q

-

-

-

DAMAGED - -

-

-

--

-

-

ST52 SS139 ST177B -

ST161A

-_

-

FAILED

-

-

-

ST151H ST130A ST72

Operator - Platfonns - Platfonns Survived Damagedamp

- Falled -

bull

-

-

Jle eBDIGIPllDbl -Amocoshy 30 1 Chevron -

143 12 Exxon 72 Mobll 22 1 Phllllps 7 bull

-Shell 39 Unocal 25 1

Total Partlclnts Platfonns 338 15

Other Operators bull- 310 13

-

Total Platfonns - -

648 28

fbml

Platforms Considered 6 7

tie of Total Platfonns 1 25

OVERVIEW OF PLATFORMS SURVIVEDDAMAGEDFAILED

-

shy

v

Platforms Evaluated in Phase I

-

-Platfonn Water Year Number Analysis Avallabillty of Availability Platform

- Name Depth rl of Model DetaDed rl ClassUlcatlon

Installation legs Inspection Soll Report Based on Information Phase I Calibration - -ft

- -- Suaill flalfnrm ~BSi -STISIK 137 1963 8 CAP Yes Yes Un eel Sarvlval

-STIJOO 170 1964 4 CAPmiddotshy Yes ST134data u ST134W 137 1981 4 CAP Yes Yesmiddot E

-

edSarvlval -

middotsurvival

WD90A 184 - 1964 8 CAP Yes E -

- Sarvlvill

MC3ll 343 1978 8 KARMA

MC397 468 1991 4 KARMA -

bull ~reglmmiddot- -

Yes Sure Survival

bull Yes - Sare Sanlval-

-T23STS2 63 1969 4 CAP

T2SSS139 62 1969 4 CAP -

Yes

Yes

Yes Ex

Yes Une

bull SurvivalLikely to DamalI

middot bull SanlvalLlkely to DamalI

ST161A 118 1964 8 USFOS

Eail11a pound1al(bullllm Clss ST177B 142 --1965 8 CAP

STISIH 137 1964 - 8 CAP C

STIJOA 140 19S8 8 CAP - -

Yes

Yes

ollamed

Yes

- Yes E

-

at about 1 mUe

Yes

ST134data

bull SafvlvalLlkelv to Dam- bull- -

E middotrahre

Emeried Failure -

Likely FaUare

T21ST72 61 1969 4 CAP Collamed No Likely FaUure

Platforms Identified for Evaluating Their Likelihood of Falling Under UnexpeCted Survival Category

CllEllATOa

AllB

llLOCll

Sl1lvcnJllE

NAME

YEAR

INSTAUJCD

WATER

IJBPJB

middotmiddotDISTANCE

ftOMSllOllE

(MIUIS)

Ill

(M)

AMOCO BllW

STllI

WDI075

A

bull D

4

n uo 17J

44

32

bull 20

10

bull CllEVllOH STtllO c I 1bull 21 11

STl130 D 0 161 21 11 STllJ4 I 0 137 D - 11

STIUI G I 137 32 11

STIUI I 0 121 32 11

STIUI

J 0 140 32 11

STIUI It 4 137 32 11

STllSI PROl)I 0 137 32 11

STllSI PROD-2 0 137 32 11

STll7 A 0

140 35 11

WDl117 c 65 214 34 10

DXON STshy E 51 14 bull STl165 A t3 bull 34 WD0073

WD0073

WD0073

A _ c

4

62

4

IA IA

17J

22

22

22

bullbullbull WD003-shy E 65 161 Z7

A 0 IN 31 MOBD Bl12I

ssn A bull

50

50

31

7

bull 7

MIJllPllY SS114 B 7 50 7 7

PlllUIPS SSl14t A 55 33 7

SAMEDAN ST17J A 4 117 7 bull 7

SJIELL WD0103 A 65 223 27 WDllOJ bull 65 221 27 WDll04 c 65 221 27 WDOI04 D 67 2 27

UNOCAL

SSl20I

SSl20I

bull SSl20I

SS0209

SS0209

SS0209

SS0209

SS 0253

SS 0253

E

I

PtJllE

A

bull D

G A

c

0

4

4

7J

61

0

0

69

103

97

97

9S

100

95

100

165

175

34

34

34

34

34

34

34

54

54

bullbullbullbullbullbullbullbullbull

Platforms Recommended for Phase II

llalr y_ w- Aot Av- _ I 1 Doplla Model nm bull bull _ bull I I H11 s18-1 Irht lbwl~ bullbullbull(-illtdlel OM lllMtD

II ---

-STISIIC 137 1963 I CAP Yeo Yeo I Yeo

STtshy -170 1964 4 CAP Yeo ST134dlU I Yeo

STl34W 137 Yeo Yeo -1981 4 CAP D Yeo

WD90A 114 1964 I CAP Yeo - D

MC311 30 1971 I KARMA Yeo -sre-

MC397 middot 461 1991 4 ICARMA Yeo Sireshy - - r- -

__ T13ST5l 63 1969 4 CAP Yeo Yeobull middot- ~n- D Yeo _ Tl5SSl39 62 1969 4 CAP Yeo Yeo I Yeo

_ STIT711 142 1965 I CAP Yeo I Yeo middot-1shySTl61A Ill 1964 I USFOS Yeo Yeo - D bulloshy

- r-

STISIH 137 1964 I CAP y D y -_ STl31A 140 1951 I CAP Yeo STl34dlU I Yeo

T21Sll 61 1969 4 CAP No rar-middot I Yeo

~-rmiddot-middot SS209A 95 1967 16 CAP Yeo D -SSU4R

11 UI A 41 -in D SS 2741AIM CAP bull

_ _ -shy - -shy- -shy

CALIBRATION

bull PHASEIAPPROACH

bull CALIBRATION ISSUES

bull PHASE II WORK SCOPE

bull SOFlWARE AVAILABILITY

bull EFFECT OF ADDITIONAL PLATFORMS

bull WEIGHTING APPROACH

bull EXAMPLE APPLICATION middot

(l=Ofce Mode] Wind

~

fcurrent l -middot

~SW ~leldl4

ultimate resistance force I first

design load Illgt

i _

deflection

-

bull

4

Historical Hurrlcane Dariiage and Survival of Platforms

Components Required for Calibration (Petrauskas 1992)

PRIOR Distribution

PLATFORM EXPERIENCE

BAYESIAN Survivals and Failures

UPDATING

Waves Winds and Currents bull

POSTERIOR Distribution

- A_middotJA ~ Survival Data Failure Data

Result Result

Load Resistance

middotfl

-

Bayesian Updaing middotGeneral Description (Petrauskas 1992)

L Imiddot

WHERE TO INTRODUCE BIAS FACTOR (B

On Resistance (Rmue - B (RPREDICreD I

middotOn Loads (Smue middot - B (SPRE01cTEo

On Safety Margin (RSmue - B (RSPREDICTED

CALIBRATION METHODOLOGY

gt

r

FIELD OBSERVATIONS

Observed Behavior

survivals Damages

Failures

r

bull

CAPACITY ANALYSIS

Predicted Load and Resistance Predicted Failure Modes

Survival Damage Failure cases

RELIABILITY ANALYSIS

- Probability of Failure ~

Ukellhood Failure

BAYESIAN UPDATING

Prior Distnbution of B Combined Likelihood Function

Posterior Dlstnbution of B

BIAS FACTOR

Mean Value cov bull

Represents Modeling Uncertainties

Prior Distribution of B Likelihood Functionmiddot Observing Success case STISIK and Posterior Distribution of B

2 00

uo

160

140

120

e 100 L $ uoIshy

0 60 ~ ~ 040

020

0 00

0 00 0 50

_ I 00

b

I 50 2 00

-shy lk [bl STISIK)

--shy Pnor-f(b)

--shy Pos1enor r(bl ISIK)

Prior or B Mean bull 10 COV bull 030

Posterlor of Bbull Meanbull 136COV 0 16

2 50

Likelihood Function and Posterior Distribution Example Platform STISIK

- -

~

- Prior and Posterior Distributions or B Failure cases (ST177B STlSlH ST130A T21) Damage cases (T23 T25 ST161A) Success cases (STlSlK ST130Q ST134W WD90A MC311 MC397)

400

350

300

2 so 0-

I

bull bull

bull bull gt

- bullbull ~ bull bull I bull

_ bull bull bullbull I bull bull bullII V -

bull bull

I bull ~ bull

bullbull bullI bull bullI bull bull bull

middot -~

bull__ ~ middotI middotmiddot

L 2 00

e ~ I SO

-I 00

- 0 so

000 gt

0 00 020 0 40 060 0 80 I 00 120 140 160 110 200

b -

-

shy

Prior-r

- - - - Posterior fLlll failure cases

bull bull bull bull bull bull bull Posterior flall damage cases

- bull bull - bull middot Postenor flall success cases

Posterior flall success fulure damage cases

-

Posterior of B Mean= 119 COV 0101

Posterior Distribution or Blas Factor (8) bull Different Categories and All Cases Combined

shy

l

PHASE I - KEY CONCLUSIONS

bull GLOBAL CORRECTION FACTOR

bull 15 TO 20 CONSERVATISM MEAN)

bull UNEXPECTED SURVIVALS HAVE GREATEST IMPACT

bull DAMAGED CASES MORE INFORMATIVE

bull CALIBRATION ISSUES PHASE II

bull

EFFECT OF IMPROVED CAPACITY ANALYSIS RECIPE

bull APPLICABILITY TO MULTIPLE FAILURE MODES

bull NO FOUNDATION DAMAGE CASE INCLUDED

bull IMPROVED REPRESENTATION OF FULL ANDREW DISTRIBUTION

-bull NUMBER OF PLATFORMS CONSIDERED

I 1 I

I t ) bull ~

PHASE II WORK SCOPE (CALIBRATION)

bull

bull ESTABLISH MULTIPLE BIAS FACTORS

Use Nine Base Case Platforms

middot - Capacity Analysis for Revised recipe

Foundation Blas Factor from Foundation JIP

bull DEVELOP WEIGHTING APPROACH

bull UPDATE MULTIPLE BIAS FACTORS

bullEffect of Platforms Not In the Nine Base Cases

bull EXAMPLE APPLICATION OF BIAS FACTORS I

bullbull

PHASE I PHASE II

CAPACITYbull ANALYSISRu bull

---------------------1----middot

Jc1 - iJbullraquoJ dbull

FttlS frt[f~l ratdhJJ

-

LIKELIHOOD FUNCTION

DEVELOPMENT

lk Jb Jfall1N) bull Jb)

111 Ill Isurvtvoq bull 1 - bl

-middot -------------------shy ~---------------------middot-middot

r111ir8 Cb1 111 Cbl BAYESIAN

UPDATING

~cove

R1R2 bull middot---------------------~----

Jblbull P(g cOJ

1 bull JbRI (MBSI I bull 1

~middot

llcJllJfalluro) bull P(gcOJ Ill 111 b21fallun) bull P ((g1 CO) U 1112 c OD

middot--------------------shy ----------------------shy bull

r111111r8 Cbl 111 Ill i-1nro1

deg ~1b2)laquottfrr (b1b21-lnfo) re Ill bull 1-11middot b2) bull lodegamp

middot

i1 covamp i2 cov~i

-

shy

--

Likelihood Function vs b All Success Case STISI K STl30Q STl34W WD90A MC311 MC397

I 0000

0 9000

OIOOO

0 7000

~ 0amp000

ta osoOO

~ 0 4000

0 3000

0 2000

0 1000

bull 0 0000

bull

I

middot

middot bull I

Ibull

-sibull ~ -middot-middot-middot-middot -middot-

----7----+-----~ I

Ibull bull bull I

rbull

--- Jk ISIK[blsuccess)

- bull - bull - lk 130Q[blsuccess) bull

lk 134W[blsuccess)

Jk-WD90A(blsuccc~I

- - - lk MC397[blsucccss)

--ltgt-- lk middotMC31 l[blsucccss)

--- Jk [bl 6 out of6succcsHascs)

05 I 5 2 250

b

Likelihood Functionsmiddot Success Cases - Phbullsc L

shy

bullbull bull bull bull bull bull -------------------~

bull bull bullbull bullbull bullbull bullbull bullbull bullbull bull

bull bullbull

bullbullbullbull

bullbull

bullbullbullbullbullbullbull

bullbullbullbullbullbull

bullbullbullbullbullbullbull

z 0

t z ) LL

c 0 0c i ~ i

J

~ sect ~--------------~ e l5

I

I

SOFTWARE

bull PF PROGRAM

bull CALREL

bull PF +CALREL

bull RELACS

bull

gt i

l

SENSITIVITY STUDIES

POSTERIOR DISTRIBUTION OF B

bull EFFECT OF CHANGE IN DISTRIBUTION OF PRIOR OF B

bull EFFECT OF REMOVAL OF ONE PLATFORM FROM EACH CATEGORY

bull EFFECT OF REMOVAL OF THREE DAMAGE CASES

bull EFFECT OF ADDITIONAL 13 SIMILAR PLATFORMS

-I

Prior and Posterior Distributions of B Variation In COV of PriorofB fromOl to04 -- -

4500

-4000

-

- -

3500 --middot - bull bull - bull bull Prior-r ICOV =02

3000

- - - - Prior-r ICOV =03 -

bull -~ e 2soo -

bull bull bull bull bull bull bull Pnor-r ICOV =o 4t frJ ~ 2000 Posterior rlCOV =0 2

bull bull Postenor rlCOV =0 3 bull bull 1500 bull middot bull - shy -middot- Posterior rlCOV =04

middot~ I 000 - bull

-middotmiddot ~ ~middotmiddot M -~

- middot 0 500

wbullbullbull _ I middot~ middotmiddotmiddotmiddotmiddot middotmiddotmiddotmiddotmiddot - -~ ~ __ middot_ _ bull -shy0 000

0 000 0200 0 400 0 600 0 800 IOOO 1200 1400 1600 1800 2000 -

- - b

-

=shy

Posterior Distribution of Bias Factor (B) bull Effect of Variation In COV of Prior of B

bull

bullPrior and Posterior Distributions of B Evaluating Effect of Removal of One Platform from Each Category from the Base Case of 13 Platforms

400

~ ~ 350

j IA i

3 00 bull I bull I ---Prior-r I bull l

250 1 I 1

I bull bull ---Posterior r lall 13 casesshy- ~E 2 00 bull bull bull bull bull bull bull Posterior r without STISIK

- ~ 1 bull - bull I - bull bull - bull bull Posterior r Iwithout T2I I 50 I

J-tt~i - bull - bull - Posterior r Iwithout T2S

tOO [-

I I ~ ~ I 0 50

0 00

___ r J i

l~T

~--

0 00 0 20 0 40 0 60 080 I 00 120 140 160 110 2 00

b

zshy

Posterior Distribution of Bias Factor (8) bull Effect of Removal of middotPlatform from Any Category

bullPrior and Posterior Distributions or B Evaluating Effect or Removal or 3 Damage Cases from the Base Case or 13 Platrorms

4

35

3

25

e L 2 ~ bulle shy

I 15

bull I

05

0

0 02 04 06 I 2 I 4 16 IS 2

I I

I

I

I Ibull

J -tmiddot L--

bullI

~

I

fl

II ~

I I v I

middot bull bull

bull L

~ ll- I-~lt

OS I

b

I

bull

1middot bullbullbullbull

~ __

bull

- - - - Prior-r

- - - - Postenor flall failure cases

Postenor flall success cases

--- Posterior flall 13 cases

---Posterior fI success failure cases only

IPosterior or B Mean =t11 COV =0121

-

_I

Posterir Distribution or Bias Factor (B) bull Effect or Removal or 3-Damage Cases

-

bullPrior and Posterior Distributions or B Evaluating Effect or Additional 13 Similar Observations as

Used in Andrew JIP The POsterior or B from Andrew-JIP Considered as New Prior or B

- 500

450

400

350

bull bull

bull bull bull bull J bull

bull bull f ~ I

I ~ Ibull bull

bullbull bull

l bull I -bullbull bull bull

I

I bull bull

bull I I~~bull bull bull

bull ~

I bull bull

~ ~_ ~~~

- - New Prior r bull

Posterior 114 new failure cases ~ 300- bull bull bull bull bull bull bull Posterior f 13 new damage casesl 150 ~

- bull bull - bull bull Posterior 116 new success cases i 2 00

Postenor 1113 new success failure -150 damage cases

I 00

0 50 -

0 00 0 00 0 20 0 40 0 60 0 80 I 00 120 I 40 160 110 2 00

Posterior ol B Mean= 120 COV= 0071 b

shy

shy

_rshy

Posterior Distribution of Bias Factor (B) bull Effect of Additional 13-Similar Observations

1

1 I

WEIGHTING APPROACH bull

bull ASSESSMENT OF PLATFORMS DATA

Select Platforms Most Mappable to the Analyzed Platforms

Unlikely SurvivalDamage Cases Identification

Based on Cut-Off Criteria Agemiddot Hs etc

bull INFORMATION REQUIRED FOR COMPARISON

bull CATEGORIZATION OF ADDITIONALmiddot PLATFORMS

bull UPDATING OF CALIBRATION RESULTS

--

OVERVIEW OF PLATFORMS SURVIVEDDAMAGED -~

Operator -

- Platforms

Survived CS)

Platforms

Damaged CD)

Platforms Used In

Phase IJIP -

JI~ ~Bill~IRBDll -

Amoco -

Chevron -

Exxon

Mob II

Phllllps

Shell

Unocal -

30

143

72 22

7

39 25shy -

1

12

bull

1 bull

1

1(S)+1 (D)

3 (S) +3(D)

2 (S)

bull shy

bull

bull

bull

Total Partlclnants Platforms 338 - 15 10C6S+4D) -

Other Operators

310

13

3 (D)

Total Platforms 648 28 13C6S+7D) -

Summary of Andrew JIP bull Phase I Platforms

Plairshy w v- N-ber Lea-le 1 shy car Hind_ o-w N-bull Deplh fl fl IJpe la bull n Yeadon Jadtet lbelledtf liefalit -- _ -OD

IL H - -shy _ICllshy IL IL - - rQClto ~ _

K STUIK 137 1963 8 365 608S4 Cellardeck

-middot STl-shy 170 1964 4 -~ K 39 62269 Cellardeck - --ST134W 137 1911 4 43 608S4 Smvlnlshy- _ - -

WD90A 184 1964 8 K 267Sbull34 83 SOS31 SUboftllar deck SmvMI

MC311 343 1978 8 x S7 62 - MC397 468 1991 4 x shy S33bull6SO 65911 -shySmeSmvlvol -- gt -

r-- T23ST52 63 1969 4 ton- shyKshy 52594 S03 -T25SSl39 62 1969 4 Kshy SOS94 S063 - middot ton-shy__ -

bullbull a ton--shyST161A 118 1964 8 K - 347S 51495 Cellarclecti

r~

ST1771 142 1965 8 K 39 60167 CeDordeck failme

STISIH 137 1964 8 K 395 608S4 CeDardeck Failme

K ST130A 140 1958 a SS 6095 I lblv Pai1uno - T21ST72 61 1969 4 Kshy S4S94 4974 I lhlv Pai1uno -

- -

JE

INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON

bull WATER DEPTH

bull bull YEAR OF INSTALLATIONDESIGN

bull PLATFORM ORIENTATION

bull PLATFORM LATITUDE AND LONGITUDE

bull KEY SKETCHES OF PLATFORM STRUCTURAL FRAMING

Number of Legs Plies Leg Batter Number of Conductors bull Deck Elevations (Including sub-cellar deck If any) Jacket Framing Elevations Brace type In Vertical Frames Sizes of Primary Members (leg pile vertical frames braces primary horizontals) Joint Details (can gap etc) LegPiie Annulus Grouting Status

bull middotPOST ANDREW INSPECTION REPORT Significant Wave Height Hs (ft)

bull INFORMATION ON DAMAGES PRIOR TO ANDREW

bull RESULTS FROM IN-HOUSE STUDIES IF AVAILABLE

bull ~RIGINAL DESIGN CRITERIA INFORMATION IF AVAILABLE

bull SOIL SHEAR STRENGTH PROFILE IF AVAILABLE

Platforms Identified for Evaluating Therr Likelihood of Falling Under Unexpected Survival Category

OPlRATOR

middotADA

BLOClt

llmucnJllE

NAME

YEAll

INSTAUJD

WA1ER

DBPTll

DISTANCE

JlllOM SHOllE

(MILU)

Bl

(M)

AMOCO 111113

STl11

WDI075

A

bull D

St

17

120

1n

44

20

10 I

CllllVllQfl STlllO c 61 110 21 11 STIUO D Q 1110 21 11

STl134 p Q 137 29 11

ST0151 G 61 137 11

ST0151 I 3 121 11

ST0151 J Q 140 11

STl151 It 137 11

ST0151 PJl001 Q UT 11

ST0151 PROD-2 Q UT 11

ST017 A 3 140 35 11

WDIUT c 65 214 34 10

llXXON STI055 II 51 6T 14 I

STl15 A t3 34

WDI073

WD0073

Amiddotshy Q

1111 1111

22

22

I

I

WDI073 c 1n 22 I

- WDOCl93-shy II

65 1110 2T A 3 194 31 MOBU 111121

SS IOl7

A B

501

50

50

31

T

I

T 6

MtlllPllY SS114 B T 50 T T

PlllllIPS SS0149 A 5 55 33 T

SAMllDAN STln A 107 T T

SBllLL WDl103 A 65 223 2T WDl103 B 65 221 2T WD0104 c 65 221 2T WD0104 D 6T 20 2T

UNOCAL SS020I

SS 020I SSl20I

SS 0209

SS0209

II p

PURll

A

B

3

n 61

103

97

97

95

100

34

34

34

34 34

bullbull bullbullbull

- SS0209 D u 95 34 I

- SS 1209 SSl253

SS0253

G A

c

6T

2

100

165

1T5

34

54

54

bullbull I

I I

I I

SOUTH MARCH

~LAND

N

SOUTH PaTOI ENE ISWltD AREA

MMS Defined Limits ror Offshore Structures Requiring Inspections Following Andrew

t---Jbulln3 T25bull I bullTtl

bull1$1Htl1Abull bull1SIK1779bull

~

middot-DaTA AREA~T

CRANO rsLE

AREA bull1300

bullll4W

=shy

bullPath of Hurricane Andrew with Platforms Used In Calibration

Likelihood Function vs b All Success Case STISI K STIJOQ ST134W WD90A MCJll MCJ97

I 0000

0 9000

0 8000

07000

-Bo6000

tiiii 0 5000 -

04000

0 3000

0 2000

01000

0 0000

0 05 25

ri1rt middot-middot-middot-middot-middot-

I 5 2

b

---11 middot151K(blsu~ccsI

- bull - bull - fk 130Q(bfU~lCSS)

lie bull134W(blsuccc]

lk-WD90A[blsultCS]

- - - lk middotMC397(bfUtlCS]

--o-- lk-MCJll(bbuctcsJ

--- ll (bl 6 out of 6 Mlttcs case]

_

( Likelihood Functions middotSuccess Cases - p - 1 c I

CATEGORIZATION OF ADDITIONAL PLATFORMS

I I I

bull CATEGORIES

Unexpected Survivals

Expected Survivals

bull APPROACH

Identify Corresponding Platform(s) Evaluated In the JIP

Determine Max Wave Height at Platform Location

Evaluate Whether Wave Hit the Deck

middot - Evaluate Likelihood of R and S to Differ from the Base Cases

Review Joint Information

Assess Special Features

bull IDENTIFY NUMBER OF PLATFORMS IN EACH CATEGORY

I

UPDATING OF CALIBRATION RESULTS

bull INCLUDE EFFECT OF ADDITIONAL PLATFORMS ON bullBbull

Consider Increased 1Uncertalntles In R and S

Develop bullNeW Ukellhood Functions for Different Categories

middotLet bullxbull Number of Ukely Unexpected Survival Cases and y Number of Ukely Expected Survival Cases

Determine Combined Likelihood Function lk (b Iadditional platforms) middot

bull REVISED LIKELIHOOD FUNCTION

lk(b I all cases) = lk(b I 9 platforms) middot x lk(b Ix platforms) x lk(b I y platforms)

bull ESTABLISH REVISED BIAS FACTOR B

f8 (b) u f8 (b) lk (b I all cases)

II

t t ~

shy

APPLICATIONS OF BIAS FACTORS

ii APPLICABLE TO PARTICULAR LOAD AND RESISTANCE RECIPE

bull USE BY THE API COMMITTEES AND MMS

Criteria Revisions

Recommended Design Guidelines Revisions bull

bullbull REASSESSMENT STUDIES AND DECISION MAKING

middot

bull RISK ASSESSMENT STUDIES BY THE OPERATORS

Individual Platforms

Fleet of Platforms

middot IMR Program Implementation

bullbull NOT TO BE USED IN DESIGN

I I fl J

EXAMPLE USE OF BIAS FACTORS

bull PHASE I EXAMPLE APPLICATION

bull EVALUATION OF ANNUAL PROBABILITY OF FAILURE

Establish Annual Maximum Seastate Data

Determine Coefficients to Define Base Shear

Determine Ultimate Capacity of Platform

Determine Probablllty of Failure Versus bullbbull

-bull DETERMINE UPDATED PROBABILITY OF FAILURE

Pf - P [S gt R]

Pf I P [~ gt Rb I b] f 8 (b) db

Pr - J P [S gt Rb I b] f bull 8 (b) db

bull TOTAL PROBABILITIES OF FAILURE

Total Pf - 1 bull Il [1 middotPf] - l Pf bull

Total Pf - 1 bull Il [1middot Pf] - l Pfmiddot

Total Pr = 1 - n [1middot PfbullJ = l Pfshy

I I I

EXAMPLE APPLICATION bull PHASE I

PLATFORM ST 151 K

bull GULF OF MEXICO

bull WATER DEPTH = 137 FT

bull ANNUAL MAXIMUM HS MEDIAN 175 ft COV 031

bull ASSUMED STORM DURATION 3 hoursmiddot

bull ASSUMED NUMBER OF WAVES 800

bull MEDIAN HS FOR STORM HOURS 1 2 AND 3

bull MEDIAN CURRENT VALUE bull

bull DETERMINE ULTIMATE CAPACITY FOR THREE DIRECTIONS

bull DETERMINE ANNUAL FAILURE PROBABILITIES

Broadsidemiddot Diagonal End-on

0037 0025 0007

0071 0044 0018

0022 0016 0004

DETERMINE TOTAL ANNUAL PROBABILITY OF FAILURE

Pf - 007

Pf - 013

Pr - 004

bull

EXAMPLE USE OF BIAS FACTORS - PHASE II middot

bull DETERMINE PREDICTIVE PROBABILITY OF FAILURE

Case - 1

Pf _ S S Pf (b b2) fa1a2 (b b2) db db2

g bull -1 (b1 R I S) - 1

92 - (b2 R2 I S) - 1

Case -2

Pf s ~ [bJ f11 (b) db

Pf [bJ p [g lt 0)

g - bull 1 (b1 R1 I S) - 1

1

bull1) l

I

BENEFITS OF ANDREW JIP bull PHASE II

lt

bull IMPROVED LOAD AND RESISTANCE RECIPE

bull IMPROVED CALIBRATION PROCEDURE

bull MULTIPLE BIAS FACTORS

bull PROCEDURE TO INCLUDE EFFECT OF OTHER PLATFORMS

bull PROCEDURE TO USE MULTIPLE BIAS FACTORS

  • Structure Bookmarks
    • ANDREW PHASE II JOINT1 INDUSTRY PROJECT
    • PROGRESS MEETING NUMBER 1
    • ANDREW PHASE II JOINT INDUSTRY PROJECT PROGRESS MEETING NUMBER 1 NOVEMBER 16 AGENDA
    • ANDREW PHASE II UST OF PARTICIPANTS
    • ANDREW PHASE I OBJECTIVES AND RESULTS
    • ANDREW PHASE II PROJECT OBJECTIVES
    • SUBSEQUENT PROJECTS
    • ANDREW PHASE II PROJECT TEAM
    • ANDREW PHASE II UPDATE MODELING middotAND ANALYSIS RECIPE
    • -ANDREW PHASE II CONDUCTOR RESISTANCE
    • ANDREW PHASE Il WA VE FORCE PROFILE
    • ANDREW PHASE II WA VE FORCE PROFILEmiddot
    • ANDREW PHASE II BRACE MODELING
    • ANDREW PHASE II FOUNDATIONS
    • ANDREW PHASE II middot middot JOINT MODELING
    • ANDREW PHASE Il JOINT MODELING OPTIONS DISCUSSION
    • ANDREW PHASE II JOINT MODEJJNG OPTIONS ILLUSTRATIONS
    • ANDREW PHASE II JOINT MODELING PHASE II METHOD
    • ANDREW PHASE II ANDREW STORM CONDmONS UPDATED
    • Platform Selection bullbullbull1111111111111
    • BasisCriteria for Selection bullbullbull1111111111111
    • CALIBRATION
    • WHERE TO INTRODUCE BIAS FACTOR (B
    • PHASE I -KEY CONCLUSIONS
    • bull CALIBRATION ISSUES PHASE II
    • PHASE II WORK SCOPE (CALIBRATION)
    • SOFTWARE
    • SENSITIVITY STUDIES POSTERIOR DISTRIBUTION OF B
    • WEIGHTING APPROACH bull
    • INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON
    • CATEGORIZATION OF ADDITIONAL PLATFORMS
    • UPDATING OF CALIBRATION RESULTS
    • APPLICATIONS OF BIAS FACTORS
    • EXAMPLE USE OF BIAS FACTORS
    • EXAMPLE APPLICATION bull PHASE I PLATFORM ST 151 K
    • EXAMPLE USE OF BIAS FACTORS -PHASE II middot
    • BENEFITS OF ANDREW JIP bull PHASE II
Page 25:  · . ' ANDREW PHASE II PROJECT OBJECTIVES . ' • ·UPDATE MODELING AND ANALYSIS RECIPE· JOINTS ' ' ·BRACES WAVE PROFILE FOUNDATIONS . ' ' CONDUCTORS· • ANALVZE PLATFORMS USING

II

2

g

~

It)

0

cCD E 0 E a c E euro~ CDshygtItgtmiddoto ~ti c -a Gic CD gt ~

-= c -Cll

I z

bullO 0i i ~ g 0

() CI -bull(q-ll) iuawow llu1wniiaAO

11

shy

116

114

112

g 110

bullc = 108

~ ~ 106 -sect 104

102

100

98

96

-

6555 60

Wave Height vs Center of Action ST151K

30 35 40 45 50

Wave Height (ft)

t

Comparison of Pushover Curves for ST 151K Variable Load Pattern and Monotonic Load Pattern (Andrew)

Soll Improved 100

--- --------- --- ----6000

5000

- 4000 8 - I

1 3000 Ul

81 m 2000

_J -shy

1000

---Variable Load

- - Andrew Load

0 f-~~~~~r--~~~~-t-~~~~~-t-~~~~~+-~~~~----t~~~~~-1

0 1 - 2 3 4 5 6

Deck Displacement (ft)

-middotshy

1 t I I ~

ANDREW PHASE II

BRACE MODELING

bull MARSHALL STRUTS USED FORPHASE I

bull PHASE II WILL ASSESS DIFFERENCE IN USING BUCKLING BEAMS AND MARSHALL STRUTS

middot bull COMPARE FOR 2D FRAME

bull COMPARE FOR PLATFORM 177 WHERE LOWER

BAY EXHIBITS LARGE BRACE MOMENTS

Project andrew Model beamframe Version 1 Mon Nov 14 145447 1994

~

middot Frame Test buckling beam-column model

-

Project andrew Model strutframe Version 1 Mon Nov 14 145754 1994

bull

Frame Test strut model

middot-middot _____________--_

_middotmiddotr_~ --ll--~-middot__

Buckling Beam bull

Strut (K=O 65)

Lateral Displacement

ANDREW PHASE II

FOUNDATIONS

bull ~LICIT NONLINEAR API FOUNDATION USED INiPHASE I

bull FOUNDATIONS

JIP SELECTED MINI-VANE DATA FOR

STRENGTH PROFILE

bull COMMENTS ON BENCHMARK COMPARISONS

- SIGNIFICANT VARIABILITY DUE TO FOUNDATION MODELING METHODS

II Ii

z

ctV 4x SPlO caissonfy-42s soil-api(lf-097step-114)

Inelastic Events Legend Elastic Strut Buckling

Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield

--------shyBeam Clmn Fully Plastic Fracture

~--~~~~~~~~~--~----~__i~~---~~~~__--~

Pro)ect apipile Model South Pelto 10 Version Wed May 18072235 1994 - -

1bullt

+

t

f i bullbullbull

bullbull bullbull

I I

5

~x SPlO Caissonfy=42s soil=api(lf=lOOstep=137)

Inelastic Events Legend

Elastic Strut Bucklilg

Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield

earn Clmn Full~ Plas1c Fracture

ANDREW PHASE II middot middot

JOINT MODELING

middotbull API RIGID-PLASTIC USED IN PHASE I

a ELASTIC-BRITTLE USED IN PHASE IB

amiddot MSL WILL PROVIDE NONLINEAR FORCE-DEFORMATION

AND MOMENT-DEFORMATION RELATIONS FOR PHASE II

a MODELING OPTIONS FOR PHASE II -

l

- UNCO~LEDmiddotONE DIMENSIONAL MEMBER END JOINT MODELING

J Ii

ANDREW PHASE Il

JOINT MODELING OPTIONS DISCUSSION

bull JD COUPLED INTERACTION SURFACE

- IDEAL

- DATA NOT AVAILABLE TO CALIBRATE

bull 2D COUPLED INTERACTION SURFACE

- SIGNIFICANT MODELING IMPROVEMENT

- DATAAVAILABLE

- IMPLEMENTATION PRACTICAL BUT NEEDS DEVELOPMENT AND CALIBRATION

bull 2D UNCOUPLED INTERACTION SURFACE

- SIGNIFICANT MODELING IMPROVEMENT

- DATA AVAILABLE

- IMPLEMENTATION PRACTICAL AND AVAILABLE

ANDREW PHASE II

JOINT MODEJJNG OPTIONS ILLUSTRATIONS

P

e

ANDREW PHASE II

JOINT MODELING PHASE II METHOD

e

I I I

ANDREW PHASE II

ANDREW STORM CONDmONS UPDATED

bull -PHASE I USED OCEANWEATHER HINDCAST-(Nov 1992)

bull middotmiddotPHASE II WILL USE THE REVISED HINDCAST (Nov 1994)

bull lt c lt I lt

2-2

l---4-~~~+-~~-+-~~~1--~~-+-~~--l--~T--+-~~--i24

f-tJVtMfiz~ 1Cfl (r STvb(

Plollol bullbull 5middotNOYmiddot9214bull1J Irbullbull Illbull [ANDREVl~NDREVCUSTbullMAXll 5-NOY-1992 10bull54 - 9208

HEIGHT lml

I~ MAXI

I L 1~middotamp SZ~- ~A I 13c

I J) ( r I X~0iWJ~0gtJ gt L gt gt I 1c I 120

-- N

~ deg

I I I I I I e C I ~=-41 122 -94 -92 -90 -88 -86 -84 -82 -80

-

Figure 7 contours of maximum hindcast significant wave height

_

~

~tl ~

NOltNttIJ 17lt ltNS S iJJ)y

Maximum Significant Wave Height - Hurricane Andrew 1992

30 ~

26 ~ I ~-~~~- == -shy -~ 35-_

24deg1--shy --------~- middot-middot-------middotmiddotmiddot----middot------middot-middot~middot- - - ~-----1---middot

22middot ~~~~~~~-~ -94 -92 -90 -88 -84deg-86 -82 -somiddot

c

-

Platform Selection bullbullbull1111111111111

0 BasisCriteria for Selection middot

0 Review of Phase I Platforms middot 0 Alternative Platforms

0 PMB Recommendations

0 Participants Comments

gt

L

l I

BasisCriteria for Selection bullbullbull1111111111111

0 Benefit for Calibration - Classification (UnexpectedExpected)

- Failure Mode (StructuresFoundations)

- Damaged StructuresKnown response

- Better Representation of Full ~opulation

0 Availability of Data - Existing Models

- Inspection Data

- Damage Data

PREDICTED BEHAVIOR

-

OBSERVED

BEHAVIOR

-SURVIVED DAMAGED FAILED

SURVIVED

ST134W WD90A

-MC311 -

MC397 -

--

-

-

ST151K ST130Q

-

-

-

DAMAGED - -

-

-

--

-

-

ST52 SS139 ST177B -

ST161A

-_

-

FAILED

-

-

-

ST151H ST130A ST72

Operator - Platfonns - Platfonns Survived Damagedamp

- Falled -

bull

-

-

Jle eBDIGIPllDbl -Amocoshy 30 1 Chevron -

143 12 Exxon 72 Mobll 22 1 Phllllps 7 bull

-Shell 39 Unocal 25 1

Total Partlclnts Platfonns 338 15

Other Operators bull- 310 13

-

Total Platfonns - -

648 28

fbml

Platforms Considered 6 7

tie of Total Platfonns 1 25

OVERVIEW OF PLATFORMS SURVIVEDDAMAGEDFAILED

-

shy

v

Platforms Evaluated in Phase I

-

-Platfonn Water Year Number Analysis Avallabillty of Availability Platform

- Name Depth rl of Model DetaDed rl ClassUlcatlon

Installation legs Inspection Soll Report Based on Information Phase I Calibration - -ft

- -- Suaill flalfnrm ~BSi -STISIK 137 1963 8 CAP Yes Yes Un eel Sarvlval

-STIJOO 170 1964 4 CAPmiddotshy Yes ST134data u ST134W 137 1981 4 CAP Yes Yesmiddot E

-

edSarvlval -

middotsurvival

WD90A 184 - 1964 8 CAP Yes E -

- Sarvlvill

MC3ll 343 1978 8 KARMA

MC397 468 1991 4 KARMA -

bull ~reglmmiddot- -

Yes Sure Survival

bull Yes - Sare Sanlval-

-T23STS2 63 1969 4 CAP

T2SSS139 62 1969 4 CAP -

Yes

Yes

Yes Ex

Yes Une

bull SurvivalLikely to DamalI

middot bull SanlvalLlkely to DamalI

ST161A 118 1964 8 USFOS

Eail11a pound1al(bullllm Clss ST177B 142 --1965 8 CAP

STISIH 137 1964 - 8 CAP C

STIJOA 140 19S8 8 CAP - -

Yes

Yes

ollamed

Yes

- Yes E

-

at about 1 mUe

Yes

ST134data

bull SafvlvalLlkelv to Dam- bull- -

E middotrahre

Emeried Failure -

Likely FaUare

T21ST72 61 1969 4 CAP Collamed No Likely FaUure

Platforms Identified for Evaluating Their Likelihood of Falling Under UnexpeCted Survival Category

CllEllATOa

AllB

llLOCll

Sl1lvcnJllE

NAME

YEAR

INSTAUJCD

WATER

IJBPJB

middotmiddotDISTANCE

ftOMSllOllE

(MIUIS)

Ill

(M)

AMOCO BllW

STllI

WDI075

A

bull D

4

n uo 17J

44

32

bull 20

10

bull CllEVllOH STtllO c I 1bull 21 11

STl130 D 0 161 21 11 STllJ4 I 0 137 D - 11

STIUI G I 137 32 11

STIUI I 0 121 32 11

STIUI

J 0 140 32 11

STIUI It 4 137 32 11

STllSI PROl)I 0 137 32 11

STllSI PROD-2 0 137 32 11

STll7 A 0

140 35 11

WDl117 c 65 214 34 10

DXON STshy E 51 14 bull STl165 A t3 bull 34 WD0073

WD0073

WD0073

A _ c

4

62

4

IA IA

17J

22

22

22

bullbullbull WD003-shy E 65 161 Z7

A 0 IN 31 MOBD Bl12I

ssn A bull

50

50

31

7

bull 7

MIJllPllY SS114 B 7 50 7 7

PlllUIPS SSl14t A 55 33 7

SAMEDAN ST17J A 4 117 7 bull 7

SJIELL WD0103 A 65 223 27 WDllOJ bull 65 221 27 WDll04 c 65 221 27 WDOI04 D 67 2 27

UNOCAL

SSl20I

SSl20I

bull SSl20I

SS0209

SS0209

SS0209

SS0209

SS 0253

SS 0253

E

I

PtJllE

A

bull D

G A

c

0

4

4

7J

61

0

0

69

103

97

97

9S

100

95

100

165

175

34

34

34

34

34

34

34

54

54

bullbullbullbullbullbullbullbullbull

Platforms Recommended for Phase II

llalr y_ w- Aot Av- _ I 1 Doplla Model nm bull bull _ bull I I H11 s18-1 Irht lbwl~ bullbullbull(-illtdlel OM lllMtD

II ---

-STISIIC 137 1963 I CAP Yeo Yeo I Yeo

STtshy -170 1964 4 CAP Yeo ST134dlU I Yeo

STl34W 137 Yeo Yeo -1981 4 CAP D Yeo

WD90A 114 1964 I CAP Yeo - D

MC311 30 1971 I KARMA Yeo -sre-

MC397 middot 461 1991 4 ICARMA Yeo Sireshy - - r- -

__ T13ST5l 63 1969 4 CAP Yeo Yeobull middot- ~n- D Yeo _ Tl5SSl39 62 1969 4 CAP Yeo Yeo I Yeo

_ STIT711 142 1965 I CAP Yeo I Yeo middot-1shySTl61A Ill 1964 I USFOS Yeo Yeo - D bulloshy

- r-

STISIH 137 1964 I CAP y D y -_ STl31A 140 1951 I CAP Yeo STl34dlU I Yeo

T21Sll 61 1969 4 CAP No rar-middot I Yeo

~-rmiddot-middot SS209A 95 1967 16 CAP Yeo D -SSU4R

11 UI A 41 -in D SS 2741AIM CAP bull

_ _ -shy - -shy- -shy

CALIBRATION

bull PHASEIAPPROACH

bull CALIBRATION ISSUES

bull PHASE II WORK SCOPE

bull SOFlWARE AVAILABILITY

bull EFFECT OF ADDITIONAL PLATFORMS

bull WEIGHTING APPROACH

bull EXAMPLE APPLICATION middot

(l=Ofce Mode] Wind

~

fcurrent l -middot

~SW ~leldl4

ultimate resistance force I first

design load Illgt

i _

deflection

-

bull

4

Historical Hurrlcane Dariiage and Survival of Platforms

Components Required for Calibration (Petrauskas 1992)

PRIOR Distribution

PLATFORM EXPERIENCE

BAYESIAN Survivals and Failures

UPDATING

Waves Winds and Currents bull

POSTERIOR Distribution

- A_middotJA ~ Survival Data Failure Data

Result Result

Load Resistance

middotfl

-

Bayesian Updaing middotGeneral Description (Petrauskas 1992)

L Imiddot

WHERE TO INTRODUCE BIAS FACTOR (B

On Resistance (Rmue - B (RPREDICreD I

middotOn Loads (Smue middot - B (SPRE01cTEo

On Safety Margin (RSmue - B (RSPREDICTED

CALIBRATION METHODOLOGY

gt

r

FIELD OBSERVATIONS

Observed Behavior

survivals Damages

Failures

r

bull

CAPACITY ANALYSIS

Predicted Load and Resistance Predicted Failure Modes

Survival Damage Failure cases

RELIABILITY ANALYSIS

- Probability of Failure ~

Ukellhood Failure

BAYESIAN UPDATING

Prior Distnbution of B Combined Likelihood Function

Posterior Dlstnbution of B

BIAS FACTOR

Mean Value cov bull

Represents Modeling Uncertainties

Prior Distribution of B Likelihood Functionmiddot Observing Success case STISIK and Posterior Distribution of B

2 00

uo

160

140

120

e 100 L $ uoIshy

0 60 ~ ~ 040

020

0 00

0 00 0 50

_ I 00

b

I 50 2 00

-shy lk [bl STISIK)

--shy Pnor-f(b)

--shy Pos1enor r(bl ISIK)

Prior or B Mean bull 10 COV bull 030

Posterlor of Bbull Meanbull 136COV 0 16

2 50

Likelihood Function and Posterior Distribution Example Platform STISIK

- -

~

- Prior and Posterior Distributions or B Failure cases (ST177B STlSlH ST130A T21) Damage cases (T23 T25 ST161A) Success cases (STlSlK ST130Q ST134W WD90A MC311 MC397)

400

350

300

2 so 0-

I

bull bull

bull bull gt

- bullbull ~ bull bull I bull

_ bull bull bullbull I bull bull bullII V -

bull bull

I bull ~ bull

bullbull bullI bull bullI bull bull bull

middot -~

bull__ ~ middotI middotmiddot

L 2 00

e ~ I SO

-I 00

- 0 so

000 gt

0 00 020 0 40 060 0 80 I 00 120 140 160 110 200

b -

-

shy

Prior-r

- - - - Posterior fLlll failure cases

bull bull bull bull bull bull bull Posterior flall damage cases

- bull bull - bull middot Postenor flall success cases

Posterior flall success fulure damage cases

-

Posterior of B Mean= 119 COV 0101

Posterior Distribution or Blas Factor (8) bull Different Categories and All Cases Combined

shy

l

PHASE I - KEY CONCLUSIONS

bull GLOBAL CORRECTION FACTOR

bull 15 TO 20 CONSERVATISM MEAN)

bull UNEXPECTED SURVIVALS HAVE GREATEST IMPACT

bull DAMAGED CASES MORE INFORMATIVE

bull CALIBRATION ISSUES PHASE II

bull

EFFECT OF IMPROVED CAPACITY ANALYSIS RECIPE

bull APPLICABILITY TO MULTIPLE FAILURE MODES

bull NO FOUNDATION DAMAGE CASE INCLUDED

bull IMPROVED REPRESENTATION OF FULL ANDREW DISTRIBUTION

-bull NUMBER OF PLATFORMS CONSIDERED

I 1 I

I t ) bull ~

PHASE II WORK SCOPE (CALIBRATION)

bull

bull ESTABLISH MULTIPLE BIAS FACTORS

Use Nine Base Case Platforms

middot - Capacity Analysis for Revised recipe

Foundation Blas Factor from Foundation JIP

bull DEVELOP WEIGHTING APPROACH

bull UPDATE MULTIPLE BIAS FACTORS

bullEffect of Platforms Not In the Nine Base Cases

bull EXAMPLE APPLICATION OF BIAS FACTORS I

bullbull

PHASE I PHASE II

CAPACITYbull ANALYSISRu bull

---------------------1----middot

Jc1 - iJbullraquoJ dbull

FttlS frt[f~l ratdhJJ

-

LIKELIHOOD FUNCTION

DEVELOPMENT

lk Jb Jfall1N) bull Jb)

111 Ill Isurvtvoq bull 1 - bl

-middot -------------------shy ~---------------------middot-middot

r111ir8 Cb1 111 Cbl BAYESIAN

UPDATING

~cove

R1R2 bull middot---------------------~----

Jblbull P(g cOJ

1 bull JbRI (MBSI I bull 1

~middot

llcJllJfalluro) bull P(gcOJ Ill 111 b21fallun) bull P ((g1 CO) U 1112 c OD

middot--------------------shy ----------------------shy bull

r111111r8 Cbl 111 Ill i-1nro1

deg ~1b2)laquottfrr (b1b21-lnfo) re Ill bull 1-11middot b2) bull lodegamp

middot

i1 covamp i2 cov~i

-

shy

--

Likelihood Function vs b All Success Case STISI K STl30Q STl34W WD90A MC311 MC397

I 0000

0 9000

OIOOO

0 7000

~ 0amp000

ta osoOO

~ 0 4000

0 3000

0 2000

0 1000

bull 0 0000

bull

I

middot

middot bull I

Ibull

-sibull ~ -middot-middot-middot-middot -middot-

----7----+-----~ I

Ibull bull bull I

rbull

--- Jk ISIK[blsuccess)

- bull - bull - lk 130Q[blsuccess) bull

lk 134W[blsuccess)

Jk-WD90A(blsuccc~I

- - - lk MC397[blsucccss)

--ltgt-- lk middotMC31 l[blsucccss)

--- Jk [bl 6 out of6succcsHascs)

05 I 5 2 250

b

Likelihood Functionsmiddot Success Cases - Phbullsc L

shy

bullbull bull bull bull bull bull -------------------~

bull bull bullbull bullbull bullbull bullbull bullbull bullbull bull

bull bullbull

bullbullbullbull

bullbull

bullbullbullbullbullbullbull

bullbullbullbullbullbull

bullbullbullbullbullbullbull

z 0

t z ) LL

c 0 0c i ~ i

J

~ sect ~--------------~ e l5

I

I

SOFTWARE

bull PF PROGRAM

bull CALREL

bull PF +CALREL

bull RELACS

bull

gt i

l

SENSITIVITY STUDIES

POSTERIOR DISTRIBUTION OF B

bull EFFECT OF CHANGE IN DISTRIBUTION OF PRIOR OF B

bull EFFECT OF REMOVAL OF ONE PLATFORM FROM EACH CATEGORY

bull EFFECT OF REMOVAL OF THREE DAMAGE CASES

bull EFFECT OF ADDITIONAL 13 SIMILAR PLATFORMS

-I

Prior and Posterior Distributions of B Variation In COV of PriorofB fromOl to04 -- -

4500

-4000

-

- -

3500 --middot - bull bull - bull bull Prior-r ICOV =02

3000

- - - - Prior-r ICOV =03 -

bull -~ e 2soo -

bull bull bull bull bull bull bull Pnor-r ICOV =o 4t frJ ~ 2000 Posterior rlCOV =0 2

bull bull Postenor rlCOV =0 3 bull bull 1500 bull middot bull - shy -middot- Posterior rlCOV =04

middot~ I 000 - bull

-middotmiddot ~ ~middotmiddot M -~

- middot 0 500

wbullbullbull _ I middot~ middotmiddotmiddotmiddotmiddot middotmiddotmiddotmiddotmiddot - -~ ~ __ middot_ _ bull -shy0 000

0 000 0200 0 400 0 600 0 800 IOOO 1200 1400 1600 1800 2000 -

- - b

-

=shy

Posterior Distribution of Bias Factor (B) bull Effect of Variation In COV of Prior of B

bull

bullPrior and Posterior Distributions of B Evaluating Effect of Removal of One Platform from Each Category from the Base Case of 13 Platforms

400

~ ~ 350

j IA i

3 00 bull I bull I ---Prior-r I bull l

250 1 I 1

I bull bull ---Posterior r lall 13 casesshy- ~E 2 00 bull bull bull bull bull bull bull Posterior r without STISIK

- ~ 1 bull - bull I - bull bull - bull bull Posterior r Iwithout T2I I 50 I

J-tt~i - bull - bull - Posterior r Iwithout T2S

tOO [-

I I ~ ~ I 0 50

0 00

___ r J i

l~T

~--

0 00 0 20 0 40 0 60 080 I 00 120 140 160 110 2 00

b

zshy

Posterior Distribution of Bias Factor (8) bull Effect of Removal of middotPlatform from Any Category

bullPrior and Posterior Distributions or B Evaluating Effect or Removal or 3 Damage Cases from the Base Case or 13 Platrorms

4

35

3

25

e L 2 ~ bulle shy

I 15

bull I

05

0

0 02 04 06 I 2 I 4 16 IS 2

I I

I

I

I Ibull

J -tmiddot L--

bullI

~

I

fl

II ~

I I v I

middot bull bull

bull L

~ ll- I-~lt

OS I

b

I

bull

1middot bullbullbullbull

~ __

bull

- - - - Prior-r

- - - - Postenor flall failure cases

Postenor flall success cases

--- Posterior flall 13 cases

---Posterior fI success failure cases only

IPosterior or B Mean =t11 COV =0121

-

_I

Posterir Distribution or Bias Factor (B) bull Effect or Removal or 3-Damage Cases

-

bullPrior and Posterior Distributions or B Evaluating Effect or Additional 13 Similar Observations as

Used in Andrew JIP The POsterior or B from Andrew-JIP Considered as New Prior or B

- 500

450

400

350

bull bull

bull bull bull bull J bull

bull bull f ~ I

I ~ Ibull bull

bullbull bull

l bull I -bullbull bull bull

I

I bull bull

bull I I~~bull bull bull

bull ~

I bull bull

~ ~_ ~~~

- - New Prior r bull

Posterior 114 new failure cases ~ 300- bull bull bull bull bull bull bull Posterior f 13 new damage casesl 150 ~

- bull bull - bull bull Posterior 116 new success cases i 2 00

Postenor 1113 new success failure -150 damage cases

I 00

0 50 -

0 00 0 00 0 20 0 40 0 60 0 80 I 00 120 I 40 160 110 2 00

Posterior ol B Mean= 120 COV= 0071 b

shy

shy

_rshy

Posterior Distribution of Bias Factor (B) bull Effect of Additional 13-Similar Observations

1

1 I

WEIGHTING APPROACH bull

bull ASSESSMENT OF PLATFORMS DATA

Select Platforms Most Mappable to the Analyzed Platforms

Unlikely SurvivalDamage Cases Identification

Based on Cut-Off Criteria Agemiddot Hs etc

bull INFORMATION REQUIRED FOR COMPARISON

bull CATEGORIZATION OF ADDITIONALmiddot PLATFORMS

bull UPDATING OF CALIBRATION RESULTS

--

OVERVIEW OF PLATFORMS SURVIVEDDAMAGED -~

Operator -

- Platforms

Survived CS)

Platforms

Damaged CD)

Platforms Used In

Phase IJIP -

JI~ ~Bill~IRBDll -

Amoco -

Chevron -

Exxon

Mob II

Phllllps

Shell

Unocal -

30

143

72 22

7

39 25shy -

1

12

bull

1 bull

1

1(S)+1 (D)

3 (S) +3(D)

2 (S)

bull shy

bull

bull

bull

Total Partlclnants Platforms 338 - 15 10C6S+4D) -

Other Operators

310

13

3 (D)

Total Platforms 648 28 13C6S+7D) -

Summary of Andrew JIP bull Phase I Platforms

Plairshy w v- N-ber Lea-le 1 shy car Hind_ o-w N-bull Deplh fl fl IJpe la bull n Yeadon Jadtet lbelledtf liefalit -- _ -OD

IL H - -shy _ICllshy IL IL - - rQClto ~ _

K STUIK 137 1963 8 365 608S4 Cellardeck

-middot STl-shy 170 1964 4 -~ K 39 62269 Cellardeck - --ST134W 137 1911 4 43 608S4 Smvlnlshy- _ - -

WD90A 184 1964 8 K 267Sbull34 83 SOS31 SUboftllar deck SmvMI

MC311 343 1978 8 x S7 62 - MC397 468 1991 4 x shy S33bull6SO 65911 -shySmeSmvlvol -- gt -

r-- T23ST52 63 1969 4 ton- shyKshy 52594 S03 -T25SSl39 62 1969 4 Kshy SOS94 S063 - middot ton-shy__ -

bullbull a ton--shyST161A 118 1964 8 K - 347S 51495 Cellarclecti

r~

ST1771 142 1965 8 K 39 60167 CeDordeck failme

STISIH 137 1964 8 K 395 608S4 CeDardeck Failme

K ST130A 140 1958 a SS 6095 I lblv Pai1uno - T21ST72 61 1969 4 Kshy S4S94 4974 I lhlv Pai1uno -

- -

JE

INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON

bull WATER DEPTH

bull bull YEAR OF INSTALLATIONDESIGN

bull PLATFORM ORIENTATION

bull PLATFORM LATITUDE AND LONGITUDE

bull KEY SKETCHES OF PLATFORM STRUCTURAL FRAMING

Number of Legs Plies Leg Batter Number of Conductors bull Deck Elevations (Including sub-cellar deck If any) Jacket Framing Elevations Brace type In Vertical Frames Sizes of Primary Members (leg pile vertical frames braces primary horizontals) Joint Details (can gap etc) LegPiie Annulus Grouting Status

bull middotPOST ANDREW INSPECTION REPORT Significant Wave Height Hs (ft)

bull INFORMATION ON DAMAGES PRIOR TO ANDREW

bull RESULTS FROM IN-HOUSE STUDIES IF AVAILABLE

bull ~RIGINAL DESIGN CRITERIA INFORMATION IF AVAILABLE

bull SOIL SHEAR STRENGTH PROFILE IF AVAILABLE

Platforms Identified for Evaluating Therr Likelihood of Falling Under Unexpected Survival Category

OPlRATOR

middotADA

BLOClt

llmucnJllE

NAME

YEAll

INSTAUJD

WA1ER

DBPTll

DISTANCE

JlllOM SHOllE

(MILU)

Bl

(M)

AMOCO 111113

STl11

WDI075

A

bull D

St

17

120

1n

44

20

10 I

CllllVllQfl STlllO c 61 110 21 11 STIUO D Q 1110 21 11

STl134 p Q 137 29 11

ST0151 G 61 137 11

ST0151 I 3 121 11

ST0151 J Q 140 11

STl151 It 137 11

ST0151 PJl001 Q UT 11

ST0151 PROD-2 Q UT 11

ST017 A 3 140 35 11

WDIUT c 65 214 34 10

llXXON STI055 II 51 6T 14 I

STl15 A t3 34

WDI073

WD0073

Amiddotshy Q

1111 1111

22

22

I

I

WDI073 c 1n 22 I

- WDOCl93-shy II

65 1110 2T A 3 194 31 MOBU 111121

SS IOl7

A B

501

50

50

31

T

I

T 6

MtlllPllY SS114 B T 50 T T

PlllllIPS SS0149 A 5 55 33 T

SAMllDAN STln A 107 T T

SBllLL WDl103 A 65 223 2T WDl103 B 65 221 2T WD0104 c 65 221 2T WD0104 D 6T 20 2T

UNOCAL SS020I

SS 020I SSl20I

SS 0209

SS0209

II p

PURll

A

B

3

n 61

103

97

97

95

100

34

34

34

34 34

bullbull bullbullbull

- SS0209 D u 95 34 I

- SS 1209 SSl253

SS0253

G A

c

6T

2

100

165

1T5

34

54

54

bullbull I

I I

I I

SOUTH MARCH

~LAND

N

SOUTH PaTOI ENE ISWltD AREA

MMS Defined Limits ror Offshore Structures Requiring Inspections Following Andrew

t---Jbulln3 T25bull I bullTtl

bull1$1Htl1Abull bull1SIK1779bull

~

middot-DaTA AREA~T

CRANO rsLE

AREA bull1300

bullll4W

=shy

bullPath of Hurricane Andrew with Platforms Used In Calibration

Likelihood Function vs b All Success Case STISI K STIJOQ ST134W WD90A MCJll MCJ97

I 0000

0 9000

0 8000

07000

-Bo6000

tiiii 0 5000 -

04000

0 3000

0 2000

01000

0 0000

0 05 25

ri1rt middot-middot-middot-middot-middot-

I 5 2

b

---11 middot151K(blsu~ccsI

- bull - bull - fk 130Q(bfU~lCSS)

lie bull134W(blsuccc]

lk-WD90A[blsultCS]

- - - lk middotMC397(bfUtlCS]

--o-- lk-MCJll(bbuctcsJ

--- ll (bl 6 out of 6 Mlttcs case]

_

( Likelihood Functions middotSuccess Cases - p - 1 c I

CATEGORIZATION OF ADDITIONAL PLATFORMS

I I I

bull CATEGORIES

Unexpected Survivals

Expected Survivals

bull APPROACH

Identify Corresponding Platform(s) Evaluated In the JIP

Determine Max Wave Height at Platform Location

Evaluate Whether Wave Hit the Deck

middot - Evaluate Likelihood of R and S to Differ from the Base Cases

Review Joint Information

Assess Special Features

bull IDENTIFY NUMBER OF PLATFORMS IN EACH CATEGORY

I

UPDATING OF CALIBRATION RESULTS

bull INCLUDE EFFECT OF ADDITIONAL PLATFORMS ON bullBbull

Consider Increased 1Uncertalntles In R and S

Develop bullNeW Ukellhood Functions for Different Categories

middotLet bullxbull Number of Ukely Unexpected Survival Cases and y Number of Ukely Expected Survival Cases

Determine Combined Likelihood Function lk (b Iadditional platforms) middot

bull REVISED LIKELIHOOD FUNCTION

lk(b I all cases) = lk(b I 9 platforms) middot x lk(b Ix platforms) x lk(b I y platforms)

bull ESTABLISH REVISED BIAS FACTOR B

f8 (b) u f8 (b) lk (b I all cases)

II

t t ~

shy

APPLICATIONS OF BIAS FACTORS

ii APPLICABLE TO PARTICULAR LOAD AND RESISTANCE RECIPE

bull USE BY THE API COMMITTEES AND MMS

Criteria Revisions

Recommended Design Guidelines Revisions bull

bullbull REASSESSMENT STUDIES AND DECISION MAKING

middot

bull RISK ASSESSMENT STUDIES BY THE OPERATORS

Individual Platforms

Fleet of Platforms

middot IMR Program Implementation

bullbull NOT TO BE USED IN DESIGN

I I fl J

EXAMPLE USE OF BIAS FACTORS

bull PHASE I EXAMPLE APPLICATION

bull EVALUATION OF ANNUAL PROBABILITY OF FAILURE

Establish Annual Maximum Seastate Data

Determine Coefficients to Define Base Shear

Determine Ultimate Capacity of Platform

Determine Probablllty of Failure Versus bullbbull

-bull DETERMINE UPDATED PROBABILITY OF FAILURE

Pf - P [S gt R]

Pf I P [~ gt Rb I b] f 8 (b) db

Pr - J P [S gt Rb I b] f bull 8 (b) db

bull TOTAL PROBABILITIES OF FAILURE

Total Pf - 1 bull Il [1 middotPf] - l Pf bull

Total Pf - 1 bull Il [1middot Pf] - l Pfmiddot

Total Pr = 1 - n [1middot PfbullJ = l Pfshy

I I I

EXAMPLE APPLICATION bull PHASE I

PLATFORM ST 151 K

bull GULF OF MEXICO

bull WATER DEPTH = 137 FT

bull ANNUAL MAXIMUM HS MEDIAN 175 ft COV 031

bull ASSUMED STORM DURATION 3 hoursmiddot

bull ASSUMED NUMBER OF WAVES 800

bull MEDIAN HS FOR STORM HOURS 1 2 AND 3

bull MEDIAN CURRENT VALUE bull

bull DETERMINE ULTIMATE CAPACITY FOR THREE DIRECTIONS

bull DETERMINE ANNUAL FAILURE PROBABILITIES

Broadsidemiddot Diagonal End-on

0037 0025 0007

0071 0044 0018

0022 0016 0004

DETERMINE TOTAL ANNUAL PROBABILITY OF FAILURE

Pf - 007

Pf - 013

Pr - 004

bull

EXAMPLE USE OF BIAS FACTORS - PHASE II middot

bull DETERMINE PREDICTIVE PROBABILITY OF FAILURE

Case - 1

Pf _ S S Pf (b b2) fa1a2 (b b2) db db2

g bull -1 (b1 R I S) - 1

92 - (b2 R2 I S) - 1

Case -2

Pf s ~ [bJ f11 (b) db

Pf [bJ p [g lt 0)

g - bull 1 (b1 R1 I S) - 1

1

bull1) l

I

BENEFITS OF ANDREW JIP bull PHASE II

lt

bull IMPROVED LOAD AND RESISTANCE RECIPE

bull IMPROVED CALIBRATION PROCEDURE

bull MULTIPLE BIAS FACTORS

bull PROCEDURE TO INCLUDE EFFECT OF OTHER PLATFORMS

bull PROCEDURE TO USE MULTIPLE BIAS FACTORS

  • Structure Bookmarks
    • ANDREW PHASE II JOINT1 INDUSTRY PROJECT
    • PROGRESS MEETING NUMBER 1
    • ANDREW PHASE II JOINT INDUSTRY PROJECT PROGRESS MEETING NUMBER 1 NOVEMBER 16 AGENDA
    • ANDREW PHASE II UST OF PARTICIPANTS
    • ANDREW PHASE I OBJECTIVES AND RESULTS
    • ANDREW PHASE II PROJECT OBJECTIVES
    • SUBSEQUENT PROJECTS
    • ANDREW PHASE II PROJECT TEAM
    • ANDREW PHASE II UPDATE MODELING middotAND ANALYSIS RECIPE
    • -ANDREW PHASE II CONDUCTOR RESISTANCE
    • ANDREW PHASE Il WA VE FORCE PROFILE
    • ANDREW PHASE II WA VE FORCE PROFILEmiddot
    • ANDREW PHASE II BRACE MODELING
    • ANDREW PHASE II FOUNDATIONS
    • ANDREW PHASE II middot middot JOINT MODELING
    • ANDREW PHASE Il JOINT MODELING OPTIONS DISCUSSION
    • ANDREW PHASE II JOINT MODEJJNG OPTIONS ILLUSTRATIONS
    • ANDREW PHASE II JOINT MODELING PHASE II METHOD
    • ANDREW PHASE II ANDREW STORM CONDmONS UPDATED
    • Platform Selection bullbullbull1111111111111
    • BasisCriteria for Selection bullbullbull1111111111111
    • CALIBRATION
    • WHERE TO INTRODUCE BIAS FACTOR (B
    • PHASE I -KEY CONCLUSIONS
    • bull CALIBRATION ISSUES PHASE II
    • PHASE II WORK SCOPE (CALIBRATION)
    • SOFTWARE
    • SENSITIVITY STUDIES POSTERIOR DISTRIBUTION OF B
    • WEIGHTING APPROACH bull
    • INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON
    • CATEGORIZATION OF ADDITIONAL PLATFORMS
    • UPDATING OF CALIBRATION RESULTS
    • APPLICATIONS OF BIAS FACTORS
    • EXAMPLE USE OF BIAS FACTORS
    • EXAMPLE APPLICATION bull PHASE I PLATFORM ST 151 K
    • EXAMPLE USE OF BIAS FACTORS -PHASE II middot
    • BENEFITS OF ANDREW JIP bull PHASE II
Page 26:  · . ' ANDREW PHASE II PROJECT OBJECTIVES . ' • ·UPDATE MODELING AND ANALYSIS RECIPE· JOINTS ' ' ·BRACES WAVE PROFILE FOUNDATIONS . ' ' CONDUCTORS· • ANALVZE PLATFORMS USING

116

114

112

g 110

bullc = 108

~ ~ 106 -sect 104

102

100

98

96

-

6555 60

Wave Height vs Center of Action ST151K

30 35 40 45 50

Wave Height (ft)

t

Comparison of Pushover Curves for ST 151K Variable Load Pattern and Monotonic Load Pattern (Andrew)

Soll Improved 100

--- --------- --- ----6000

5000

- 4000 8 - I

1 3000 Ul

81 m 2000

_J -shy

1000

---Variable Load

- - Andrew Load

0 f-~~~~~r--~~~~-t-~~~~~-t-~~~~~+-~~~~----t~~~~~-1

0 1 - 2 3 4 5 6

Deck Displacement (ft)

-middotshy

1 t I I ~

ANDREW PHASE II

BRACE MODELING

bull MARSHALL STRUTS USED FORPHASE I

bull PHASE II WILL ASSESS DIFFERENCE IN USING BUCKLING BEAMS AND MARSHALL STRUTS

middot bull COMPARE FOR 2D FRAME

bull COMPARE FOR PLATFORM 177 WHERE LOWER

BAY EXHIBITS LARGE BRACE MOMENTS

Project andrew Model beamframe Version 1 Mon Nov 14 145447 1994

~

middot Frame Test buckling beam-column model

-

Project andrew Model strutframe Version 1 Mon Nov 14 145754 1994

bull

Frame Test strut model

middot-middot _____________--_

_middotmiddotr_~ --ll--~-middot__

Buckling Beam bull

Strut (K=O 65)

Lateral Displacement

ANDREW PHASE II

FOUNDATIONS

bull ~LICIT NONLINEAR API FOUNDATION USED INiPHASE I

bull FOUNDATIONS

JIP SELECTED MINI-VANE DATA FOR

STRENGTH PROFILE

bull COMMENTS ON BENCHMARK COMPARISONS

- SIGNIFICANT VARIABILITY DUE TO FOUNDATION MODELING METHODS

II Ii

z

ctV 4x SPlO caissonfy-42s soil-api(lf-097step-114)

Inelastic Events Legend Elastic Strut Buckling

Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield

--------shyBeam Clmn Fully Plastic Fracture

~--~~~~~~~~~--~----~__i~~---~~~~__--~

Pro)ect apipile Model South Pelto 10 Version Wed May 18072235 1994 - -

1bullt

+

t

f i bullbullbull

bullbull bullbull

I I

5

~x SPlO Caissonfy=42s soil=api(lf=lOOstep=137)

Inelastic Events Legend

Elastic Strut Bucklilg

Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield

earn Clmn Full~ Plas1c Fracture

ANDREW PHASE II middot middot

JOINT MODELING

middotbull API RIGID-PLASTIC USED IN PHASE I

a ELASTIC-BRITTLE USED IN PHASE IB

amiddot MSL WILL PROVIDE NONLINEAR FORCE-DEFORMATION

AND MOMENT-DEFORMATION RELATIONS FOR PHASE II

a MODELING OPTIONS FOR PHASE II -

l

- UNCO~LEDmiddotONE DIMENSIONAL MEMBER END JOINT MODELING

J Ii

ANDREW PHASE Il

JOINT MODELING OPTIONS DISCUSSION

bull JD COUPLED INTERACTION SURFACE

- IDEAL

- DATA NOT AVAILABLE TO CALIBRATE

bull 2D COUPLED INTERACTION SURFACE

- SIGNIFICANT MODELING IMPROVEMENT

- DATAAVAILABLE

- IMPLEMENTATION PRACTICAL BUT NEEDS DEVELOPMENT AND CALIBRATION

bull 2D UNCOUPLED INTERACTION SURFACE

- SIGNIFICANT MODELING IMPROVEMENT

- DATA AVAILABLE

- IMPLEMENTATION PRACTICAL AND AVAILABLE

ANDREW PHASE II

JOINT MODEJJNG OPTIONS ILLUSTRATIONS

P

e

ANDREW PHASE II

JOINT MODELING PHASE II METHOD

e

I I I

ANDREW PHASE II

ANDREW STORM CONDmONS UPDATED

bull -PHASE I USED OCEANWEATHER HINDCAST-(Nov 1992)

bull middotmiddotPHASE II WILL USE THE REVISED HINDCAST (Nov 1994)

bull lt c lt I lt

2-2

l---4-~~~+-~~-+-~~~1--~~-+-~~--l--~T--+-~~--i24

f-tJVtMfiz~ 1Cfl (r STvb(

Plollol bullbull 5middotNOYmiddot9214bull1J Irbullbull Illbull [ANDREVl~NDREVCUSTbullMAXll 5-NOY-1992 10bull54 - 9208

HEIGHT lml

I~ MAXI

I L 1~middotamp SZ~- ~A I 13c

I J) ( r I X~0iWJ~0gtJ gt L gt gt I 1c I 120

-- N

~ deg

I I I I I I e C I ~=-41 122 -94 -92 -90 -88 -86 -84 -82 -80

-

Figure 7 contours of maximum hindcast significant wave height

_

~

~tl ~

NOltNttIJ 17lt ltNS S iJJ)y

Maximum Significant Wave Height - Hurricane Andrew 1992

30 ~

26 ~ I ~-~~~- == -shy -~ 35-_

24deg1--shy --------~- middot-middot-------middotmiddotmiddot----middot------middot-middot~middot- - - ~-----1---middot

22middot ~~~~~~~-~ -94 -92 -90 -88 -84deg-86 -82 -somiddot

c

-

Platform Selection bullbullbull1111111111111

0 BasisCriteria for Selection middot

0 Review of Phase I Platforms middot 0 Alternative Platforms

0 PMB Recommendations

0 Participants Comments

gt

L

l I

BasisCriteria for Selection bullbullbull1111111111111

0 Benefit for Calibration - Classification (UnexpectedExpected)

- Failure Mode (StructuresFoundations)

- Damaged StructuresKnown response

- Better Representation of Full ~opulation

0 Availability of Data - Existing Models

- Inspection Data

- Damage Data

PREDICTED BEHAVIOR

-

OBSERVED

BEHAVIOR

-SURVIVED DAMAGED FAILED

SURVIVED

ST134W WD90A

-MC311 -

MC397 -

--

-

-

ST151K ST130Q

-

-

-

DAMAGED - -

-

-

--

-

-

ST52 SS139 ST177B -

ST161A

-_

-

FAILED

-

-

-

ST151H ST130A ST72

Operator - Platfonns - Platfonns Survived Damagedamp

- Falled -

bull

-

-

Jle eBDIGIPllDbl -Amocoshy 30 1 Chevron -

143 12 Exxon 72 Mobll 22 1 Phllllps 7 bull

-Shell 39 Unocal 25 1

Total Partlclnts Platfonns 338 15

Other Operators bull- 310 13

-

Total Platfonns - -

648 28

fbml

Platforms Considered 6 7

tie of Total Platfonns 1 25

OVERVIEW OF PLATFORMS SURVIVEDDAMAGEDFAILED

-

shy

v

Platforms Evaluated in Phase I

-

-Platfonn Water Year Number Analysis Avallabillty of Availability Platform

- Name Depth rl of Model DetaDed rl ClassUlcatlon

Installation legs Inspection Soll Report Based on Information Phase I Calibration - -ft

- -- Suaill flalfnrm ~BSi -STISIK 137 1963 8 CAP Yes Yes Un eel Sarvlval

-STIJOO 170 1964 4 CAPmiddotshy Yes ST134data u ST134W 137 1981 4 CAP Yes Yesmiddot E

-

edSarvlval -

middotsurvival

WD90A 184 - 1964 8 CAP Yes E -

- Sarvlvill

MC3ll 343 1978 8 KARMA

MC397 468 1991 4 KARMA -

bull ~reglmmiddot- -

Yes Sure Survival

bull Yes - Sare Sanlval-

-T23STS2 63 1969 4 CAP

T2SSS139 62 1969 4 CAP -

Yes

Yes

Yes Ex

Yes Une

bull SurvivalLikely to DamalI

middot bull SanlvalLlkely to DamalI

ST161A 118 1964 8 USFOS

Eail11a pound1al(bullllm Clss ST177B 142 --1965 8 CAP

STISIH 137 1964 - 8 CAP C

STIJOA 140 19S8 8 CAP - -

Yes

Yes

ollamed

Yes

- Yes E

-

at about 1 mUe

Yes

ST134data

bull SafvlvalLlkelv to Dam- bull- -

E middotrahre

Emeried Failure -

Likely FaUare

T21ST72 61 1969 4 CAP Collamed No Likely FaUure

Platforms Identified for Evaluating Their Likelihood of Falling Under UnexpeCted Survival Category

CllEllATOa

AllB

llLOCll

Sl1lvcnJllE

NAME

YEAR

INSTAUJCD

WATER

IJBPJB

middotmiddotDISTANCE

ftOMSllOllE

(MIUIS)

Ill

(M)

AMOCO BllW

STllI

WDI075

A

bull D

4

n uo 17J

44

32

bull 20

10

bull CllEVllOH STtllO c I 1bull 21 11

STl130 D 0 161 21 11 STllJ4 I 0 137 D - 11

STIUI G I 137 32 11

STIUI I 0 121 32 11

STIUI

J 0 140 32 11

STIUI It 4 137 32 11

STllSI PROl)I 0 137 32 11

STllSI PROD-2 0 137 32 11

STll7 A 0

140 35 11

WDl117 c 65 214 34 10

DXON STshy E 51 14 bull STl165 A t3 bull 34 WD0073

WD0073

WD0073

A _ c

4

62

4

IA IA

17J

22

22

22

bullbullbull WD003-shy E 65 161 Z7

A 0 IN 31 MOBD Bl12I

ssn A bull

50

50

31

7

bull 7

MIJllPllY SS114 B 7 50 7 7

PlllUIPS SSl14t A 55 33 7

SAMEDAN ST17J A 4 117 7 bull 7

SJIELL WD0103 A 65 223 27 WDllOJ bull 65 221 27 WDll04 c 65 221 27 WDOI04 D 67 2 27

UNOCAL

SSl20I

SSl20I

bull SSl20I

SS0209

SS0209

SS0209

SS0209

SS 0253

SS 0253

E

I

PtJllE

A

bull D

G A

c

0

4

4

7J

61

0

0

69

103

97

97

9S

100

95

100

165

175

34

34

34

34

34

34

34

54

54

bullbullbullbullbullbullbullbullbull

Platforms Recommended for Phase II

llalr y_ w- Aot Av- _ I 1 Doplla Model nm bull bull _ bull I I H11 s18-1 Irht lbwl~ bullbullbull(-illtdlel OM lllMtD

II ---

-STISIIC 137 1963 I CAP Yeo Yeo I Yeo

STtshy -170 1964 4 CAP Yeo ST134dlU I Yeo

STl34W 137 Yeo Yeo -1981 4 CAP D Yeo

WD90A 114 1964 I CAP Yeo - D

MC311 30 1971 I KARMA Yeo -sre-

MC397 middot 461 1991 4 ICARMA Yeo Sireshy - - r- -

__ T13ST5l 63 1969 4 CAP Yeo Yeobull middot- ~n- D Yeo _ Tl5SSl39 62 1969 4 CAP Yeo Yeo I Yeo

_ STIT711 142 1965 I CAP Yeo I Yeo middot-1shySTl61A Ill 1964 I USFOS Yeo Yeo - D bulloshy

- r-

STISIH 137 1964 I CAP y D y -_ STl31A 140 1951 I CAP Yeo STl34dlU I Yeo

T21Sll 61 1969 4 CAP No rar-middot I Yeo

~-rmiddot-middot SS209A 95 1967 16 CAP Yeo D -SSU4R

11 UI A 41 -in D SS 2741AIM CAP bull

_ _ -shy - -shy- -shy

CALIBRATION

bull PHASEIAPPROACH

bull CALIBRATION ISSUES

bull PHASE II WORK SCOPE

bull SOFlWARE AVAILABILITY

bull EFFECT OF ADDITIONAL PLATFORMS

bull WEIGHTING APPROACH

bull EXAMPLE APPLICATION middot

(l=Ofce Mode] Wind

~

fcurrent l -middot

~SW ~leldl4

ultimate resistance force I first

design load Illgt

i _

deflection

-

bull

4

Historical Hurrlcane Dariiage and Survival of Platforms

Components Required for Calibration (Petrauskas 1992)

PRIOR Distribution

PLATFORM EXPERIENCE

BAYESIAN Survivals and Failures

UPDATING

Waves Winds and Currents bull

POSTERIOR Distribution

- A_middotJA ~ Survival Data Failure Data

Result Result

Load Resistance

middotfl

-

Bayesian Updaing middotGeneral Description (Petrauskas 1992)

L Imiddot

WHERE TO INTRODUCE BIAS FACTOR (B

On Resistance (Rmue - B (RPREDICreD I

middotOn Loads (Smue middot - B (SPRE01cTEo

On Safety Margin (RSmue - B (RSPREDICTED

CALIBRATION METHODOLOGY

gt

r

FIELD OBSERVATIONS

Observed Behavior

survivals Damages

Failures

r

bull

CAPACITY ANALYSIS

Predicted Load and Resistance Predicted Failure Modes

Survival Damage Failure cases

RELIABILITY ANALYSIS

- Probability of Failure ~

Ukellhood Failure

BAYESIAN UPDATING

Prior Distnbution of B Combined Likelihood Function

Posterior Dlstnbution of B

BIAS FACTOR

Mean Value cov bull

Represents Modeling Uncertainties

Prior Distribution of B Likelihood Functionmiddot Observing Success case STISIK and Posterior Distribution of B

2 00

uo

160

140

120

e 100 L $ uoIshy

0 60 ~ ~ 040

020

0 00

0 00 0 50

_ I 00

b

I 50 2 00

-shy lk [bl STISIK)

--shy Pnor-f(b)

--shy Pos1enor r(bl ISIK)

Prior or B Mean bull 10 COV bull 030

Posterlor of Bbull Meanbull 136COV 0 16

2 50

Likelihood Function and Posterior Distribution Example Platform STISIK

- -

~

- Prior and Posterior Distributions or B Failure cases (ST177B STlSlH ST130A T21) Damage cases (T23 T25 ST161A) Success cases (STlSlK ST130Q ST134W WD90A MC311 MC397)

400

350

300

2 so 0-

I

bull bull

bull bull gt

- bullbull ~ bull bull I bull

_ bull bull bullbull I bull bull bullII V -

bull bull

I bull ~ bull

bullbull bullI bull bullI bull bull bull

middot -~

bull__ ~ middotI middotmiddot

L 2 00

e ~ I SO

-I 00

- 0 so

000 gt

0 00 020 0 40 060 0 80 I 00 120 140 160 110 200

b -

-

shy

Prior-r

- - - - Posterior fLlll failure cases

bull bull bull bull bull bull bull Posterior flall damage cases

- bull bull - bull middot Postenor flall success cases

Posterior flall success fulure damage cases

-

Posterior of B Mean= 119 COV 0101

Posterior Distribution or Blas Factor (8) bull Different Categories and All Cases Combined

shy

l

PHASE I - KEY CONCLUSIONS

bull GLOBAL CORRECTION FACTOR

bull 15 TO 20 CONSERVATISM MEAN)

bull UNEXPECTED SURVIVALS HAVE GREATEST IMPACT

bull DAMAGED CASES MORE INFORMATIVE

bull CALIBRATION ISSUES PHASE II

bull

EFFECT OF IMPROVED CAPACITY ANALYSIS RECIPE

bull APPLICABILITY TO MULTIPLE FAILURE MODES

bull NO FOUNDATION DAMAGE CASE INCLUDED

bull IMPROVED REPRESENTATION OF FULL ANDREW DISTRIBUTION

-bull NUMBER OF PLATFORMS CONSIDERED

I 1 I

I t ) bull ~

PHASE II WORK SCOPE (CALIBRATION)

bull

bull ESTABLISH MULTIPLE BIAS FACTORS

Use Nine Base Case Platforms

middot - Capacity Analysis for Revised recipe

Foundation Blas Factor from Foundation JIP

bull DEVELOP WEIGHTING APPROACH

bull UPDATE MULTIPLE BIAS FACTORS

bullEffect of Platforms Not In the Nine Base Cases

bull EXAMPLE APPLICATION OF BIAS FACTORS I

bullbull

PHASE I PHASE II

CAPACITYbull ANALYSISRu bull

---------------------1----middot

Jc1 - iJbullraquoJ dbull

FttlS frt[f~l ratdhJJ

-

LIKELIHOOD FUNCTION

DEVELOPMENT

lk Jb Jfall1N) bull Jb)

111 Ill Isurvtvoq bull 1 - bl

-middot -------------------shy ~---------------------middot-middot

r111ir8 Cb1 111 Cbl BAYESIAN

UPDATING

~cove

R1R2 bull middot---------------------~----

Jblbull P(g cOJ

1 bull JbRI (MBSI I bull 1

~middot

llcJllJfalluro) bull P(gcOJ Ill 111 b21fallun) bull P ((g1 CO) U 1112 c OD

middot--------------------shy ----------------------shy bull

r111111r8 Cbl 111 Ill i-1nro1

deg ~1b2)laquottfrr (b1b21-lnfo) re Ill bull 1-11middot b2) bull lodegamp

middot

i1 covamp i2 cov~i

-

shy

--

Likelihood Function vs b All Success Case STISI K STl30Q STl34W WD90A MC311 MC397

I 0000

0 9000

OIOOO

0 7000

~ 0amp000

ta osoOO

~ 0 4000

0 3000

0 2000

0 1000

bull 0 0000

bull

I

middot

middot bull I

Ibull

-sibull ~ -middot-middot-middot-middot -middot-

----7----+-----~ I

Ibull bull bull I

rbull

--- Jk ISIK[blsuccess)

- bull - bull - lk 130Q[blsuccess) bull

lk 134W[blsuccess)

Jk-WD90A(blsuccc~I

- - - lk MC397[blsucccss)

--ltgt-- lk middotMC31 l[blsucccss)

--- Jk [bl 6 out of6succcsHascs)

05 I 5 2 250

b

Likelihood Functionsmiddot Success Cases - Phbullsc L

shy

bullbull bull bull bull bull bull -------------------~

bull bull bullbull bullbull bullbull bullbull bullbull bullbull bull

bull bullbull

bullbullbullbull

bullbull

bullbullbullbullbullbullbull

bullbullbullbullbullbull

bullbullbullbullbullbullbull

z 0

t z ) LL

c 0 0c i ~ i

J

~ sect ~--------------~ e l5

I

I

SOFTWARE

bull PF PROGRAM

bull CALREL

bull PF +CALREL

bull RELACS

bull

gt i

l

SENSITIVITY STUDIES

POSTERIOR DISTRIBUTION OF B

bull EFFECT OF CHANGE IN DISTRIBUTION OF PRIOR OF B

bull EFFECT OF REMOVAL OF ONE PLATFORM FROM EACH CATEGORY

bull EFFECT OF REMOVAL OF THREE DAMAGE CASES

bull EFFECT OF ADDITIONAL 13 SIMILAR PLATFORMS

-I

Prior and Posterior Distributions of B Variation In COV of PriorofB fromOl to04 -- -

4500

-4000

-

- -

3500 --middot - bull bull - bull bull Prior-r ICOV =02

3000

- - - - Prior-r ICOV =03 -

bull -~ e 2soo -

bull bull bull bull bull bull bull Pnor-r ICOV =o 4t frJ ~ 2000 Posterior rlCOV =0 2

bull bull Postenor rlCOV =0 3 bull bull 1500 bull middot bull - shy -middot- Posterior rlCOV =04

middot~ I 000 - bull

-middotmiddot ~ ~middotmiddot M -~

- middot 0 500

wbullbullbull _ I middot~ middotmiddotmiddotmiddotmiddot middotmiddotmiddotmiddotmiddot - -~ ~ __ middot_ _ bull -shy0 000

0 000 0200 0 400 0 600 0 800 IOOO 1200 1400 1600 1800 2000 -

- - b

-

=shy

Posterior Distribution of Bias Factor (B) bull Effect of Variation In COV of Prior of B

bull

bullPrior and Posterior Distributions of B Evaluating Effect of Removal of One Platform from Each Category from the Base Case of 13 Platforms

400

~ ~ 350

j IA i

3 00 bull I bull I ---Prior-r I bull l

250 1 I 1

I bull bull ---Posterior r lall 13 casesshy- ~E 2 00 bull bull bull bull bull bull bull Posterior r without STISIK

- ~ 1 bull - bull I - bull bull - bull bull Posterior r Iwithout T2I I 50 I

J-tt~i - bull - bull - Posterior r Iwithout T2S

tOO [-

I I ~ ~ I 0 50

0 00

___ r J i

l~T

~--

0 00 0 20 0 40 0 60 080 I 00 120 140 160 110 2 00

b

zshy

Posterior Distribution of Bias Factor (8) bull Effect of Removal of middotPlatform from Any Category

bullPrior and Posterior Distributions or B Evaluating Effect or Removal or 3 Damage Cases from the Base Case or 13 Platrorms

4

35

3

25

e L 2 ~ bulle shy

I 15

bull I

05

0

0 02 04 06 I 2 I 4 16 IS 2

I I

I

I

I Ibull

J -tmiddot L--

bullI

~

I

fl

II ~

I I v I

middot bull bull

bull L

~ ll- I-~lt

OS I

b

I

bull

1middot bullbullbullbull

~ __

bull

- - - - Prior-r

- - - - Postenor flall failure cases

Postenor flall success cases

--- Posterior flall 13 cases

---Posterior fI success failure cases only

IPosterior or B Mean =t11 COV =0121

-

_I

Posterir Distribution or Bias Factor (B) bull Effect or Removal or 3-Damage Cases

-

bullPrior and Posterior Distributions or B Evaluating Effect or Additional 13 Similar Observations as

Used in Andrew JIP The POsterior or B from Andrew-JIP Considered as New Prior or B

- 500

450

400

350

bull bull

bull bull bull bull J bull

bull bull f ~ I

I ~ Ibull bull

bullbull bull

l bull I -bullbull bull bull

I

I bull bull

bull I I~~bull bull bull

bull ~

I bull bull

~ ~_ ~~~

- - New Prior r bull

Posterior 114 new failure cases ~ 300- bull bull bull bull bull bull bull Posterior f 13 new damage casesl 150 ~

- bull bull - bull bull Posterior 116 new success cases i 2 00

Postenor 1113 new success failure -150 damage cases

I 00

0 50 -

0 00 0 00 0 20 0 40 0 60 0 80 I 00 120 I 40 160 110 2 00

Posterior ol B Mean= 120 COV= 0071 b

shy

shy

_rshy

Posterior Distribution of Bias Factor (B) bull Effect of Additional 13-Similar Observations

1

1 I

WEIGHTING APPROACH bull

bull ASSESSMENT OF PLATFORMS DATA

Select Platforms Most Mappable to the Analyzed Platforms

Unlikely SurvivalDamage Cases Identification

Based on Cut-Off Criteria Agemiddot Hs etc

bull INFORMATION REQUIRED FOR COMPARISON

bull CATEGORIZATION OF ADDITIONALmiddot PLATFORMS

bull UPDATING OF CALIBRATION RESULTS

--

OVERVIEW OF PLATFORMS SURVIVEDDAMAGED -~

Operator -

- Platforms

Survived CS)

Platforms

Damaged CD)

Platforms Used In

Phase IJIP -

JI~ ~Bill~IRBDll -

Amoco -

Chevron -

Exxon

Mob II

Phllllps

Shell

Unocal -

30

143

72 22

7

39 25shy -

1

12

bull

1 bull

1

1(S)+1 (D)

3 (S) +3(D)

2 (S)

bull shy

bull

bull

bull

Total Partlclnants Platforms 338 - 15 10C6S+4D) -

Other Operators

310

13

3 (D)

Total Platforms 648 28 13C6S+7D) -

Summary of Andrew JIP bull Phase I Platforms

Plairshy w v- N-ber Lea-le 1 shy car Hind_ o-w N-bull Deplh fl fl IJpe la bull n Yeadon Jadtet lbelledtf liefalit -- _ -OD

IL H - -shy _ICllshy IL IL - - rQClto ~ _

K STUIK 137 1963 8 365 608S4 Cellardeck

-middot STl-shy 170 1964 4 -~ K 39 62269 Cellardeck - --ST134W 137 1911 4 43 608S4 Smvlnlshy- _ - -

WD90A 184 1964 8 K 267Sbull34 83 SOS31 SUboftllar deck SmvMI

MC311 343 1978 8 x S7 62 - MC397 468 1991 4 x shy S33bull6SO 65911 -shySmeSmvlvol -- gt -

r-- T23ST52 63 1969 4 ton- shyKshy 52594 S03 -T25SSl39 62 1969 4 Kshy SOS94 S063 - middot ton-shy__ -

bullbull a ton--shyST161A 118 1964 8 K - 347S 51495 Cellarclecti

r~

ST1771 142 1965 8 K 39 60167 CeDordeck failme

STISIH 137 1964 8 K 395 608S4 CeDardeck Failme

K ST130A 140 1958 a SS 6095 I lblv Pai1uno - T21ST72 61 1969 4 Kshy S4S94 4974 I lhlv Pai1uno -

- -

JE

INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON

bull WATER DEPTH

bull bull YEAR OF INSTALLATIONDESIGN

bull PLATFORM ORIENTATION

bull PLATFORM LATITUDE AND LONGITUDE

bull KEY SKETCHES OF PLATFORM STRUCTURAL FRAMING

Number of Legs Plies Leg Batter Number of Conductors bull Deck Elevations (Including sub-cellar deck If any) Jacket Framing Elevations Brace type In Vertical Frames Sizes of Primary Members (leg pile vertical frames braces primary horizontals) Joint Details (can gap etc) LegPiie Annulus Grouting Status

bull middotPOST ANDREW INSPECTION REPORT Significant Wave Height Hs (ft)

bull INFORMATION ON DAMAGES PRIOR TO ANDREW

bull RESULTS FROM IN-HOUSE STUDIES IF AVAILABLE

bull ~RIGINAL DESIGN CRITERIA INFORMATION IF AVAILABLE

bull SOIL SHEAR STRENGTH PROFILE IF AVAILABLE

Platforms Identified for Evaluating Therr Likelihood of Falling Under Unexpected Survival Category

OPlRATOR

middotADA

BLOClt

llmucnJllE

NAME

YEAll

INSTAUJD

WA1ER

DBPTll

DISTANCE

JlllOM SHOllE

(MILU)

Bl

(M)

AMOCO 111113

STl11

WDI075

A

bull D

St

17

120

1n

44

20

10 I

CllllVllQfl STlllO c 61 110 21 11 STIUO D Q 1110 21 11

STl134 p Q 137 29 11

ST0151 G 61 137 11

ST0151 I 3 121 11

ST0151 J Q 140 11

STl151 It 137 11

ST0151 PJl001 Q UT 11

ST0151 PROD-2 Q UT 11

ST017 A 3 140 35 11

WDIUT c 65 214 34 10

llXXON STI055 II 51 6T 14 I

STl15 A t3 34

WDI073

WD0073

Amiddotshy Q

1111 1111

22

22

I

I

WDI073 c 1n 22 I

- WDOCl93-shy II

65 1110 2T A 3 194 31 MOBU 111121

SS IOl7

A B

501

50

50

31

T

I

T 6

MtlllPllY SS114 B T 50 T T

PlllllIPS SS0149 A 5 55 33 T

SAMllDAN STln A 107 T T

SBllLL WDl103 A 65 223 2T WDl103 B 65 221 2T WD0104 c 65 221 2T WD0104 D 6T 20 2T

UNOCAL SS020I

SS 020I SSl20I

SS 0209

SS0209

II p

PURll

A

B

3

n 61

103

97

97

95

100

34

34

34

34 34

bullbull bullbullbull

- SS0209 D u 95 34 I

- SS 1209 SSl253

SS0253

G A

c

6T

2

100

165

1T5

34

54

54

bullbull I

I I

I I

SOUTH MARCH

~LAND

N

SOUTH PaTOI ENE ISWltD AREA

MMS Defined Limits ror Offshore Structures Requiring Inspections Following Andrew

t---Jbulln3 T25bull I bullTtl

bull1$1Htl1Abull bull1SIK1779bull

~

middot-DaTA AREA~T

CRANO rsLE

AREA bull1300

bullll4W

=shy

bullPath of Hurricane Andrew with Platforms Used In Calibration

Likelihood Function vs b All Success Case STISI K STIJOQ ST134W WD90A MCJll MCJ97

I 0000

0 9000

0 8000

07000

-Bo6000

tiiii 0 5000 -

04000

0 3000

0 2000

01000

0 0000

0 05 25

ri1rt middot-middot-middot-middot-middot-

I 5 2

b

---11 middot151K(blsu~ccsI

- bull - bull - fk 130Q(bfU~lCSS)

lie bull134W(blsuccc]

lk-WD90A[blsultCS]

- - - lk middotMC397(bfUtlCS]

--o-- lk-MCJll(bbuctcsJ

--- ll (bl 6 out of 6 Mlttcs case]

_

( Likelihood Functions middotSuccess Cases - p - 1 c I

CATEGORIZATION OF ADDITIONAL PLATFORMS

I I I

bull CATEGORIES

Unexpected Survivals

Expected Survivals

bull APPROACH

Identify Corresponding Platform(s) Evaluated In the JIP

Determine Max Wave Height at Platform Location

Evaluate Whether Wave Hit the Deck

middot - Evaluate Likelihood of R and S to Differ from the Base Cases

Review Joint Information

Assess Special Features

bull IDENTIFY NUMBER OF PLATFORMS IN EACH CATEGORY

I

UPDATING OF CALIBRATION RESULTS

bull INCLUDE EFFECT OF ADDITIONAL PLATFORMS ON bullBbull

Consider Increased 1Uncertalntles In R and S

Develop bullNeW Ukellhood Functions for Different Categories

middotLet bullxbull Number of Ukely Unexpected Survival Cases and y Number of Ukely Expected Survival Cases

Determine Combined Likelihood Function lk (b Iadditional platforms) middot

bull REVISED LIKELIHOOD FUNCTION

lk(b I all cases) = lk(b I 9 platforms) middot x lk(b Ix platforms) x lk(b I y platforms)

bull ESTABLISH REVISED BIAS FACTOR B

f8 (b) u f8 (b) lk (b I all cases)

II

t t ~

shy

APPLICATIONS OF BIAS FACTORS

ii APPLICABLE TO PARTICULAR LOAD AND RESISTANCE RECIPE

bull USE BY THE API COMMITTEES AND MMS

Criteria Revisions

Recommended Design Guidelines Revisions bull

bullbull REASSESSMENT STUDIES AND DECISION MAKING

middot

bull RISK ASSESSMENT STUDIES BY THE OPERATORS

Individual Platforms

Fleet of Platforms

middot IMR Program Implementation

bullbull NOT TO BE USED IN DESIGN

I I fl J

EXAMPLE USE OF BIAS FACTORS

bull PHASE I EXAMPLE APPLICATION

bull EVALUATION OF ANNUAL PROBABILITY OF FAILURE

Establish Annual Maximum Seastate Data

Determine Coefficients to Define Base Shear

Determine Ultimate Capacity of Platform

Determine Probablllty of Failure Versus bullbbull

-bull DETERMINE UPDATED PROBABILITY OF FAILURE

Pf - P [S gt R]

Pf I P [~ gt Rb I b] f 8 (b) db

Pr - J P [S gt Rb I b] f bull 8 (b) db

bull TOTAL PROBABILITIES OF FAILURE

Total Pf - 1 bull Il [1 middotPf] - l Pf bull

Total Pf - 1 bull Il [1middot Pf] - l Pfmiddot

Total Pr = 1 - n [1middot PfbullJ = l Pfshy

I I I

EXAMPLE APPLICATION bull PHASE I

PLATFORM ST 151 K

bull GULF OF MEXICO

bull WATER DEPTH = 137 FT

bull ANNUAL MAXIMUM HS MEDIAN 175 ft COV 031

bull ASSUMED STORM DURATION 3 hoursmiddot

bull ASSUMED NUMBER OF WAVES 800

bull MEDIAN HS FOR STORM HOURS 1 2 AND 3

bull MEDIAN CURRENT VALUE bull

bull DETERMINE ULTIMATE CAPACITY FOR THREE DIRECTIONS

bull DETERMINE ANNUAL FAILURE PROBABILITIES

Broadsidemiddot Diagonal End-on

0037 0025 0007

0071 0044 0018

0022 0016 0004

DETERMINE TOTAL ANNUAL PROBABILITY OF FAILURE

Pf - 007

Pf - 013

Pr - 004

bull

EXAMPLE USE OF BIAS FACTORS - PHASE II middot

bull DETERMINE PREDICTIVE PROBABILITY OF FAILURE

Case - 1

Pf _ S S Pf (b b2) fa1a2 (b b2) db db2

g bull -1 (b1 R I S) - 1

92 - (b2 R2 I S) - 1

Case -2

Pf s ~ [bJ f11 (b) db

Pf [bJ p [g lt 0)

g - bull 1 (b1 R1 I S) - 1

1

bull1) l

I

BENEFITS OF ANDREW JIP bull PHASE II

lt

bull IMPROVED LOAD AND RESISTANCE RECIPE

bull IMPROVED CALIBRATION PROCEDURE

bull MULTIPLE BIAS FACTORS

bull PROCEDURE TO INCLUDE EFFECT OF OTHER PLATFORMS

bull PROCEDURE TO USE MULTIPLE BIAS FACTORS

  • Structure Bookmarks
    • ANDREW PHASE II JOINT1 INDUSTRY PROJECT
    • PROGRESS MEETING NUMBER 1
    • ANDREW PHASE II JOINT INDUSTRY PROJECT PROGRESS MEETING NUMBER 1 NOVEMBER 16 AGENDA
    • ANDREW PHASE II UST OF PARTICIPANTS
    • ANDREW PHASE I OBJECTIVES AND RESULTS
    • ANDREW PHASE II PROJECT OBJECTIVES
    • SUBSEQUENT PROJECTS
    • ANDREW PHASE II PROJECT TEAM
    • ANDREW PHASE II UPDATE MODELING middotAND ANALYSIS RECIPE
    • -ANDREW PHASE II CONDUCTOR RESISTANCE
    • ANDREW PHASE Il WA VE FORCE PROFILE
    • ANDREW PHASE II WA VE FORCE PROFILEmiddot
    • ANDREW PHASE II BRACE MODELING
    • ANDREW PHASE II FOUNDATIONS
    • ANDREW PHASE II middot middot JOINT MODELING
    • ANDREW PHASE Il JOINT MODELING OPTIONS DISCUSSION
    • ANDREW PHASE II JOINT MODEJJNG OPTIONS ILLUSTRATIONS
    • ANDREW PHASE II JOINT MODELING PHASE II METHOD
    • ANDREW PHASE II ANDREW STORM CONDmONS UPDATED
    • Platform Selection bullbullbull1111111111111
    • BasisCriteria for Selection bullbullbull1111111111111
    • CALIBRATION
    • WHERE TO INTRODUCE BIAS FACTOR (B
    • PHASE I -KEY CONCLUSIONS
    • bull CALIBRATION ISSUES PHASE II
    • PHASE II WORK SCOPE (CALIBRATION)
    • SOFTWARE
    • SENSITIVITY STUDIES POSTERIOR DISTRIBUTION OF B
    • WEIGHTING APPROACH bull
    • INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON
    • CATEGORIZATION OF ADDITIONAL PLATFORMS
    • UPDATING OF CALIBRATION RESULTS
    • APPLICATIONS OF BIAS FACTORS
    • EXAMPLE USE OF BIAS FACTORS
    • EXAMPLE APPLICATION bull PHASE I PLATFORM ST 151 K
    • EXAMPLE USE OF BIAS FACTORS -PHASE II middot
    • BENEFITS OF ANDREW JIP bull PHASE II
Page 27:  · . ' ANDREW PHASE II PROJECT OBJECTIVES . ' • ·UPDATE MODELING AND ANALYSIS RECIPE· JOINTS ' ' ·BRACES WAVE PROFILE FOUNDATIONS . ' ' CONDUCTORS· • ANALVZE PLATFORMS USING

t

Comparison of Pushover Curves for ST 151K Variable Load Pattern and Monotonic Load Pattern (Andrew)

Soll Improved 100

--- --------- --- ----6000

5000

- 4000 8 - I

1 3000 Ul

81 m 2000

_J -shy

1000

---Variable Load

- - Andrew Load

0 f-~~~~~r--~~~~-t-~~~~~-t-~~~~~+-~~~~----t~~~~~-1

0 1 - 2 3 4 5 6

Deck Displacement (ft)

-middotshy

1 t I I ~

ANDREW PHASE II

BRACE MODELING

bull MARSHALL STRUTS USED FORPHASE I

bull PHASE II WILL ASSESS DIFFERENCE IN USING BUCKLING BEAMS AND MARSHALL STRUTS

middot bull COMPARE FOR 2D FRAME

bull COMPARE FOR PLATFORM 177 WHERE LOWER

BAY EXHIBITS LARGE BRACE MOMENTS

Project andrew Model beamframe Version 1 Mon Nov 14 145447 1994

~

middot Frame Test buckling beam-column model

-

Project andrew Model strutframe Version 1 Mon Nov 14 145754 1994

bull

Frame Test strut model

middot-middot _____________--_

_middotmiddotr_~ --ll--~-middot__

Buckling Beam bull

Strut (K=O 65)

Lateral Displacement

ANDREW PHASE II

FOUNDATIONS

bull ~LICIT NONLINEAR API FOUNDATION USED INiPHASE I

bull FOUNDATIONS

JIP SELECTED MINI-VANE DATA FOR

STRENGTH PROFILE

bull COMMENTS ON BENCHMARK COMPARISONS

- SIGNIFICANT VARIABILITY DUE TO FOUNDATION MODELING METHODS

II Ii

z

ctV 4x SPlO caissonfy-42s soil-api(lf-097step-114)

Inelastic Events Legend Elastic Strut Buckling

Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield

--------shyBeam Clmn Fully Plastic Fracture

~--~~~~~~~~~--~----~__i~~---~~~~__--~

Pro)ect apipile Model South Pelto 10 Version Wed May 18072235 1994 - -

1bullt

+

t

f i bullbullbull

bullbull bullbull

I I

5

~x SPlO Caissonfy=42s soil=api(lf=lOOstep=137)

Inelastic Events Legend

Elastic Strut Bucklilg

Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield

earn Clmn Full~ Plas1c Fracture

ANDREW PHASE II middot middot

JOINT MODELING

middotbull API RIGID-PLASTIC USED IN PHASE I

a ELASTIC-BRITTLE USED IN PHASE IB

amiddot MSL WILL PROVIDE NONLINEAR FORCE-DEFORMATION

AND MOMENT-DEFORMATION RELATIONS FOR PHASE II

a MODELING OPTIONS FOR PHASE II -

l

- UNCO~LEDmiddotONE DIMENSIONAL MEMBER END JOINT MODELING

J Ii

ANDREW PHASE Il

JOINT MODELING OPTIONS DISCUSSION

bull JD COUPLED INTERACTION SURFACE

- IDEAL

- DATA NOT AVAILABLE TO CALIBRATE

bull 2D COUPLED INTERACTION SURFACE

- SIGNIFICANT MODELING IMPROVEMENT

- DATAAVAILABLE

- IMPLEMENTATION PRACTICAL BUT NEEDS DEVELOPMENT AND CALIBRATION

bull 2D UNCOUPLED INTERACTION SURFACE

- SIGNIFICANT MODELING IMPROVEMENT

- DATA AVAILABLE

- IMPLEMENTATION PRACTICAL AND AVAILABLE

ANDREW PHASE II

JOINT MODEJJNG OPTIONS ILLUSTRATIONS

P

e

ANDREW PHASE II

JOINT MODELING PHASE II METHOD

e

I I I

ANDREW PHASE II

ANDREW STORM CONDmONS UPDATED

bull -PHASE I USED OCEANWEATHER HINDCAST-(Nov 1992)

bull middotmiddotPHASE II WILL USE THE REVISED HINDCAST (Nov 1994)

bull lt c lt I lt

2-2

l---4-~~~+-~~-+-~~~1--~~-+-~~--l--~T--+-~~--i24

f-tJVtMfiz~ 1Cfl (r STvb(

Plollol bullbull 5middotNOYmiddot9214bull1J Irbullbull Illbull [ANDREVl~NDREVCUSTbullMAXll 5-NOY-1992 10bull54 - 9208

HEIGHT lml

I~ MAXI

I L 1~middotamp SZ~- ~A I 13c

I J) ( r I X~0iWJ~0gtJ gt L gt gt I 1c I 120

-- N

~ deg

I I I I I I e C I ~=-41 122 -94 -92 -90 -88 -86 -84 -82 -80

-

Figure 7 contours of maximum hindcast significant wave height

_

~

~tl ~

NOltNttIJ 17lt ltNS S iJJ)y

Maximum Significant Wave Height - Hurricane Andrew 1992

30 ~

26 ~ I ~-~~~- == -shy -~ 35-_

24deg1--shy --------~- middot-middot-------middotmiddotmiddot----middot------middot-middot~middot- - - ~-----1---middot

22middot ~~~~~~~-~ -94 -92 -90 -88 -84deg-86 -82 -somiddot

c

-

Platform Selection bullbullbull1111111111111

0 BasisCriteria for Selection middot

0 Review of Phase I Platforms middot 0 Alternative Platforms

0 PMB Recommendations

0 Participants Comments

gt

L

l I

BasisCriteria for Selection bullbullbull1111111111111

0 Benefit for Calibration - Classification (UnexpectedExpected)

- Failure Mode (StructuresFoundations)

- Damaged StructuresKnown response

- Better Representation of Full ~opulation

0 Availability of Data - Existing Models

- Inspection Data

- Damage Data

PREDICTED BEHAVIOR

-

OBSERVED

BEHAVIOR

-SURVIVED DAMAGED FAILED

SURVIVED

ST134W WD90A

-MC311 -

MC397 -

--

-

-

ST151K ST130Q

-

-

-

DAMAGED - -

-

-

--

-

-

ST52 SS139 ST177B -

ST161A

-_

-

FAILED

-

-

-

ST151H ST130A ST72

Operator - Platfonns - Platfonns Survived Damagedamp

- Falled -

bull

-

-

Jle eBDIGIPllDbl -Amocoshy 30 1 Chevron -

143 12 Exxon 72 Mobll 22 1 Phllllps 7 bull

-Shell 39 Unocal 25 1

Total Partlclnts Platfonns 338 15

Other Operators bull- 310 13

-

Total Platfonns - -

648 28

fbml

Platforms Considered 6 7

tie of Total Platfonns 1 25

OVERVIEW OF PLATFORMS SURVIVEDDAMAGEDFAILED

-

shy

v

Platforms Evaluated in Phase I

-

-Platfonn Water Year Number Analysis Avallabillty of Availability Platform

- Name Depth rl of Model DetaDed rl ClassUlcatlon

Installation legs Inspection Soll Report Based on Information Phase I Calibration - -ft

- -- Suaill flalfnrm ~BSi -STISIK 137 1963 8 CAP Yes Yes Un eel Sarvlval

-STIJOO 170 1964 4 CAPmiddotshy Yes ST134data u ST134W 137 1981 4 CAP Yes Yesmiddot E

-

edSarvlval -

middotsurvival

WD90A 184 - 1964 8 CAP Yes E -

- Sarvlvill

MC3ll 343 1978 8 KARMA

MC397 468 1991 4 KARMA -

bull ~reglmmiddot- -

Yes Sure Survival

bull Yes - Sare Sanlval-

-T23STS2 63 1969 4 CAP

T2SSS139 62 1969 4 CAP -

Yes

Yes

Yes Ex

Yes Une

bull SurvivalLikely to DamalI

middot bull SanlvalLlkely to DamalI

ST161A 118 1964 8 USFOS

Eail11a pound1al(bullllm Clss ST177B 142 --1965 8 CAP

STISIH 137 1964 - 8 CAP C

STIJOA 140 19S8 8 CAP - -

Yes

Yes

ollamed

Yes

- Yes E

-

at about 1 mUe

Yes

ST134data

bull SafvlvalLlkelv to Dam- bull- -

E middotrahre

Emeried Failure -

Likely FaUare

T21ST72 61 1969 4 CAP Collamed No Likely FaUure

Platforms Identified for Evaluating Their Likelihood of Falling Under UnexpeCted Survival Category

CllEllATOa

AllB

llLOCll

Sl1lvcnJllE

NAME

YEAR

INSTAUJCD

WATER

IJBPJB

middotmiddotDISTANCE

ftOMSllOllE

(MIUIS)

Ill

(M)

AMOCO BllW

STllI

WDI075

A

bull D

4

n uo 17J

44

32

bull 20

10

bull CllEVllOH STtllO c I 1bull 21 11

STl130 D 0 161 21 11 STllJ4 I 0 137 D - 11

STIUI G I 137 32 11

STIUI I 0 121 32 11

STIUI

J 0 140 32 11

STIUI It 4 137 32 11

STllSI PROl)I 0 137 32 11

STllSI PROD-2 0 137 32 11

STll7 A 0

140 35 11

WDl117 c 65 214 34 10

DXON STshy E 51 14 bull STl165 A t3 bull 34 WD0073

WD0073

WD0073

A _ c

4

62

4

IA IA

17J

22

22

22

bullbullbull WD003-shy E 65 161 Z7

A 0 IN 31 MOBD Bl12I

ssn A bull

50

50

31

7

bull 7

MIJllPllY SS114 B 7 50 7 7

PlllUIPS SSl14t A 55 33 7

SAMEDAN ST17J A 4 117 7 bull 7

SJIELL WD0103 A 65 223 27 WDllOJ bull 65 221 27 WDll04 c 65 221 27 WDOI04 D 67 2 27

UNOCAL

SSl20I

SSl20I

bull SSl20I

SS0209

SS0209

SS0209

SS0209

SS 0253

SS 0253

E

I

PtJllE

A

bull D

G A

c

0

4

4

7J

61

0

0

69

103

97

97

9S

100

95

100

165

175

34

34

34

34

34

34

34

54

54

bullbullbullbullbullbullbullbullbull

Platforms Recommended for Phase II

llalr y_ w- Aot Av- _ I 1 Doplla Model nm bull bull _ bull I I H11 s18-1 Irht lbwl~ bullbullbull(-illtdlel OM lllMtD

II ---

-STISIIC 137 1963 I CAP Yeo Yeo I Yeo

STtshy -170 1964 4 CAP Yeo ST134dlU I Yeo

STl34W 137 Yeo Yeo -1981 4 CAP D Yeo

WD90A 114 1964 I CAP Yeo - D

MC311 30 1971 I KARMA Yeo -sre-

MC397 middot 461 1991 4 ICARMA Yeo Sireshy - - r- -

__ T13ST5l 63 1969 4 CAP Yeo Yeobull middot- ~n- D Yeo _ Tl5SSl39 62 1969 4 CAP Yeo Yeo I Yeo

_ STIT711 142 1965 I CAP Yeo I Yeo middot-1shySTl61A Ill 1964 I USFOS Yeo Yeo - D bulloshy

- r-

STISIH 137 1964 I CAP y D y -_ STl31A 140 1951 I CAP Yeo STl34dlU I Yeo

T21Sll 61 1969 4 CAP No rar-middot I Yeo

~-rmiddot-middot SS209A 95 1967 16 CAP Yeo D -SSU4R

11 UI A 41 -in D SS 2741AIM CAP bull

_ _ -shy - -shy- -shy

CALIBRATION

bull PHASEIAPPROACH

bull CALIBRATION ISSUES

bull PHASE II WORK SCOPE

bull SOFlWARE AVAILABILITY

bull EFFECT OF ADDITIONAL PLATFORMS

bull WEIGHTING APPROACH

bull EXAMPLE APPLICATION middot

(l=Ofce Mode] Wind

~

fcurrent l -middot

~SW ~leldl4

ultimate resistance force I first

design load Illgt

i _

deflection

-

bull

4

Historical Hurrlcane Dariiage and Survival of Platforms

Components Required for Calibration (Petrauskas 1992)

PRIOR Distribution

PLATFORM EXPERIENCE

BAYESIAN Survivals and Failures

UPDATING

Waves Winds and Currents bull

POSTERIOR Distribution

- A_middotJA ~ Survival Data Failure Data

Result Result

Load Resistance

middotfl

-

Bayesian Updaing middotGeneral Description (Petrauskas 1992)

L Imiddot

WHERE TO INTRODUCE BIAS FACTOR (B

On Resistance (Rmue - B (RPREDICreD I

middotOn Loads (Smue middot - B (SPRE01cTEo

On Safety Margin (RSmue - B (RSPREDICTED

CALIBRATION METHODOLOGY

gt

r

FIELD OBSERVATIONS

Observed Behavior

survivals Damages

Failures

r

bull

CAPACITY ANALYSIS

Predicted Load and Resistance Predicted Failure Modes

Survival Damage Failure cases

RELIABILITY ANALYSIS

- Probability of Failure ~

Ukellhood Failure

BAYESIAN UPDATING

Prior Distnbution of B Combined Likelihood Function

Posterior Dlstnbution of B

BIAS FACTOR

Mean Value cov bull

Represents Modeling Uncertainties

Prior Distribution of B Likelihood Functionmiddot Observing Success case STISIK and Posterior Distribution of B

2 00

uo

160

140

120

e 100 L $ uoIshy

0 60 ~ ~ 040

020

0 00

0 00 0 50

_ I 00

b

I 50 2 00

-shy lk [bl STISIK)

--shy Pnor-f(b)

--shy Pos1enor r(bl ISIK)

Prior or B Mean bull 10 COV bull 030

Posterlor of Bbull Meanbull 136COV 0 16

2 50

Likelihood Function and Posterior Distribution Example Platform STISIK

- -

~

- Prior and Posterior Distributions or B Failure cases (ST177B STlSlH ST130A T21) Damage cases (T23 T25 ST161A) Success cases (STlSlK ST130Q ST134W WD90A MC311 MC397)

400

350

300

2 so 0-

I

bull bull

bull bull gt

- bullbull ~ bull bull I bull

_ bull bull bullbull I bull bull bullII V -

bull bull

I bull ~ bull

bullbull bullI bull bullI bull bull bull

middot -~

bull__ ~ middotI middotmiddot

L 2 00

e ~ I SO

-I 00

- 0 so

000 gt

0 00 020 0 40 060 0 80 I 00 120 140 160 110 200

b -

-

shy

Prior-r

- - - - Posterior fLlll failure cases

bull bull bull bull bull bull bull Posterior flall damage cases

- bull bull - bull middot Postenor flall success cases

Posterior flall success fulure damage cases

-

Posterior of B Mean= 119 COV 0101

Posterior Distribution or Blas Factor (8) bull Different Categories and All Cases Combined

shy

l

PHASE I - KEY CONCLUSIONS

bull GLOBAL CORRECTION FACTOR

bull 15 TO 20 CONSERVATISM MEAN)

bull UNEXPECTED SURVIVALS HAVE GREATEST IMPACT

bull DAMAGED CASES MORE INFORMATIVE

bull CALIBRATION ISSUES PHASE II

bull

EFFECT OF IMPROVED CAPACITY ANALYSIS RECIPE

bull APPLICABILITY TO MULTIPLE FAILURE MODES

bull NO FOUNDATION DAMAGE CASE INCLUDED

bull IMPROVED REPRESENTATION OF FULL ANDREW DISTRIBUTION

-bull NUMBER OF PLATFORMS CONSIDERED

I 1 I

I t ) bull ~

PHASE II WORK SCOPE (CALIBRATION)

bull

bull ESTABLISH MULTIPLE BIAS FACTORS

Use Nine Base Case Platforms

middot - Capacity Analysis for Revised recipe

Foundation Blas Factor from Foundation JIP

bull DEVELOP WEIGHTING APPROACH

bull UPDATE MULTIPLE BIAS FACTORS

bullEffect of Platforms Not In the Nine Base Cases

bull EXAMPLE APPLICATION OF BIAS FACTORS I

bullbull

PHASE I PHASE II

CAPACITYbull ANALYSISRu bull

---------------------1----middot

Jc1 - iJbullraquoJ dbull

FttlS frt[f~l ratdhJJ

-

LIKELIHOOD FUNCTION

DEVELOPMENT

lk Jb Jfall1N) bull Jb)

111 Ill Isurvtvoq bull 1 - bl

-middot -------------------shy ~---------------------middot-middot

r111ir8 Cb1 111 Cbl BAYESIAN

UPDATING

~cove

R1R2 bull middot---------------------~----

Jblbull P(g cOJ

1 bull JbRI (MBSI I bull 1

~middot

llcJllJfalluro) bull P(gcOJ Ill 111 b21fallun) bull P ((g1 CO) U 1112 c OD

middot--------------------shy ----------------------shy bull

r111111r8 Cbl 111 Ill i-1nro1

deg ~1b2)laquottfrr (b1b21-lnfo) re Ill bull 1-11middot b2) bull lodegamp

middot

i1 covamp i2 cov~i

-

shy

--

Likelihood Function vs b All Success Case STISI K STl30Q STl34W WD90A MC311 MC397

I 0000

0 9000

OIOOO

0 7000

~ 0amp000

ta osoOO

~ 0 4000

0 3000

0 2000

0 1000

bull 0 0000

bull

I

middot

middot bull I

Ibull

-sibull ~ -middot-middot-middot-middot -middot-

----7----+-----~ I

Ibull bull bull I

rbull

--- Jk ISIK[blsuccess)

- bull - bull - lk 130Q[blsuccess) bull

lk 134W[blsuccess)

Jk-WD90A(blsuccc~I

- - - lk MC397[blsucccss)

--ltgt-- lk middotMC31 l[blsucccss)

--- Jk [bl 6 out of6succcsHascs)

05 I 5 2 250

b

Likelihood Functionsmiddot Success Cases - Phbullsc L

shy

bullbull bull bull bull bull bull -------------------~

bull bull bullbull bullbull bullbull bullbull bullbull bullbull bull

bull bullbull

bullbullbullbull

bullbull

bullbullbullbullbullbullbull

bullbullbullbullbullbull

bullbullbullbullbullbullbull

z 0

t z ) LL

c 0 0c i ~ i

J

~ sect ~--------------~ e l5

I

I

SOFTWARE

bull PF PROGRAM

bull CALREL

bull PF +CALREL

bull RELACS

bull

gt i

l

SENSITIVITY STUDIES

POSTERIOR DISTRIBUTION OF B

bull EFFECT OF CHANGE IN DISTRIBUTION OF PRIOR OF B

bull EFFECT OF REMOVAL OF ONE PLATFORM FROM EACH CATEGORY

bull EFFECT OF REMOVAL OF THREE DAMAGE CASES

bull EFFECT OF ADDITIONAL 13 SIMILAR PLATFORMS

-I

Prior and Posterior Distributions of B Variation In COV of PriorofB fromOl to04 -- -

4500

-4000

-

- -

3500 --middot - bull bull - bull bull Prior-r ICOV =02

3000

- - - - Prior-r ICOV =03 -

bull -~ e 2soo -

bull bull bull bull bull bull bull Pnor-r ICOV =o 4t frJ ~ 2000 Posterior rlCOV =0 2

bull bull Postenor rlCOV =0 3 bull bull 1500 bull middot bull - shy -middot- Posterior rlCOV =04

middot~ I 000 - bull

-middotmiddot ~ ~middotmiddot M -~

- middot 0 500

wbullbullbull _ I middot~ middotmiddotmiddotmiddotmiddot middotmiddotmiddotmiddotmiddot - -~ ~ __ middot_ _ bull -shy0 000

0 000 0200 0 400 0 600 0 800 IOOO 1200 1400 1600 1800 2000 -

- - b

-

=shy

Posterior Distribution of Bias Factor (B) bull Effect of Variation In COV of Prior of B

bull

bullPrior and Posterior Distributions of B Evaluating Effect of Removal of One Platform from Each Category from the Base Case of 13 Platforms

400

~ ~ 350

j IA i

3 00 bull I bull I ---Prior-r I bull l

250 1 I 1

I bull bull ---Posterior r lall 13 casesshy- ~E 2 00 bull bull bull bull bull bull bull Posterior r without STISIK

- ~ 1 bull - bull I - bull bull - bull bull Posterior r Iwithout T2I I 50 I

J-tt~i - bull - bull - Posterior r Iwithout T2S

tOO [-

I I ~ ~ I 0 50

0 00

___ r J i

l~T

~--

0 00 0 20 0 40 0 60 080 I 00 120 140 160 110 2 00

b

zshy

Posterior Distribution of Bias Factor (8) bull Effect of Removal of middotPlatform from Any Category

bullPrior and Posterior Distributions or B Evaluating Effect or Removal or 3 Damage Cases from the Base Case or 13 Platrorms

4

35

3

25

e L 2 ~ bulle shy

I 15

bull I

05

0

0 02 04 06 I 2 I 4 16 IS 2

I I

I

I

I Ibull

J -tmiddot L--

bullI

~

I

fl

II ~

I I v I

middot bull bull

bull L

~ ll- I-~lt

OS I

b

I

bull

1middot bullbullbullbull

~ __

bull

- - - - Prior-r

- - - - Postenor flall failure cases

Postenor flall success cases

--- Posterior flall 13 cases

---Posterior fI success failure cases only

IPosterior or B Mean =t11 COV =0121

-

_I

Posterir Distribution or Bias Factor (B) bull Effect or Removal or 3-Damage Cases

-

bullPrior and Posterior Distributions or B Evaluating Effect or Additional 13 Similar Observations as

Used in Andrew JIP The POsterior or B from Andrew-JIP Considered as New Prior or B

- 500

450

400

350

bull bull

bull bull bull bull J bull

bull bull f ~ I

I ~ Ibull bull

bullbull bull

l bull I -bullbull bull bull

I

I bull bull

bull I I~~bull bull bull

bull ~

I bull bull

~ ~_ ~~~

- - New Prior r bull

Posterior 114 new failure cases ~ 300- bull bull bull bull bull bull bull Posterior f 13 new damage casesl 150 ~

- bull bull - bull bull Posterior 116 new success cases i 2 00

Postenor 1113 new success failure -150 damage cases

I 00

0 50 -

0 00 0 00 0 20 0 40 0 60 0 80 I 00 120 I 40 160 110 2 00

Posterior ol B Mean= 120 COV= 0071 b

shy

shy

_rshy

Posterior Distribution of Bias Factor (B) bull Effect of Additional 13-Similar Observations

1

1 I

WEIGHTING APPROACH bull

bull ASSESSMENT OF PLATFORMS DATA

Select Platforms Most Mappable to the Analyzed Platforms

Unlikely SurvivalDamage Cases Identification

Based on Cut-Off Criteria Agemiddot Hs etc

bull INFORMATION REQUIRED FOR COMPARISON

bull CATEGORIZATION OF ADDITIONALmiddot PLATFORMS

bull UPDATING OF CALIBRATION RESULTS

--

OVERVIEW OF PLATFORMS SURVIVEDDAMAGED -~

Operator -

- Platforms

Survived CS)

Platforms

Damaged CD)

Platforms Used In

Phase IJIP -

JI~ ~Bill~IRBDll -

Amoco -

Chevron -

Exxon

Mob II

Phllllps

Shell

Unocal -

30

143

72 22

7

39 25shy -

1

12

bull

1 bull

1

1(S)+1 (D)

3 (S) +3(D)

2 (S)

bull shy

bull

bull

bull

Total Partlclnants Platforms 338 - 15 10C6S+4D) -

Other Operators

310

13

3 (D)

Total Platforms 648 28 13C6S+7D) -

Summary of Andrew JIP bull Phase I Platforms

Plairshy w v- N-ber Lea-le 1 shy car Hind_ o-w N-bull Deplh fl fl IJpe la bull n Yeadon Jadtet lbelledtf liefalit -- _ -OD

IL H - -shy _ICllshy IL IL - - rQClto ~ _

K STUIK 137 1963 8 365 608S4 Cellardeck

-middot STl-shy 170 1964 4 -~ K 39 62269 Cellardeck - --ST134W 137 1911 4 43 608S4 Smvlnlshy- _ - -

WD90A 184 1964 8 K 267Sbull34 83 SOS31 SUboftllar deck SmvMI

MC311 343 1978 8 x S7 62 - MC397 468 1991 4 x shy S33bull6SO 65911 -shySmeSmvlvol -- gt -

r-- T23ST52 63 1969 4 ton- shyKshy 52594 S03 -T25SSl39 62 1969 4 Kshy SOS94 S063 - middot ton-shy__ -

bullbull a ton--shyST161A 118 1964 8 K - 347S 51495 Cellarclecti

r~

ST1771 142 1965 8 K 39 60167 CeDordeck failme

STISIH 137 1964 8 K 395 608S4 CeDardeck Failme

K ST130A 140 1958 a SS 6095 I lblv Pai1uno - T21ST72 61 1969 4 Kshy S4S94 4974 I lhlv Pai1uno -

- -

JE

INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON

bull WATER DEPTH

bull bull YEAR OF INSTALLATIONDESIGN

bull PLATFORM ORIENTATION

bull PLATFORM LATITUDE AND LONGITUDE

bull KEY SKETCHES OF PLATFORM STRUCTURAL FRAMING

Number of Legs Plies Leg Batter Number of Conductors bull Deck Elevations (Including sub-cellar deck If any) Jacket Framing Elevations Brace type In Vertical Frames Sizes of Primary Members (leg pile vertical frames braces primary horizontals) Joint Details (can gap etc) LegPiie Annulus Grouting Status

bull middotPOST ANDREW INSPECTION REPORT Significant Wave Height Hs (ft)

bull INFORMATION ON DAMAGES PRIOR TO ANDREW

bull RESULTS FROM IN-HOUSE STUDIES IF AVAILABLE

bull ~RIGINAL DESIGN CRITERIA INFORMATION IF AVAILABLE

bull SOIL SHEAR STRENGTH PROFILE IF AVAILABLE

Platforms Identified for Evaluating Therr Likelihood of Falling Under Unexpected Survival Category

OPlRATOR

middotADA

BLOClt

llmucnJllE

NAME

YEAll

INSTAUJD

WA1ER

DBPTll

DISTANCE

JlllOM SHOllE

(MILU)

Bl

(M)

AMOCO 111113

STl11

WDI075

A

bull D

St

17

120

1n

44

20

10 I

CllllVllQfl STlllO c 61 110 21 11 STIUO D Q 1110 21 11

STl134 p Q 137 29 11

ST0151 G 61 137 11

ST0151 I 3 121 11

ST0151 J Q 140 11

STl151 It 137 11

ST0151 PJl001 Q UT 11

ST0151 PROD-2 Q UT 11

ST017 A 3 140 35 11

WDIUT c 65 214 34 10

llXXON STI055 II 51 6T 14 I

STl15 A t3 34

WDI073

WD0073

Amiddotshy Q

1111 1111

22

22

I

I

WDI073 c 1n 22 I

- WDOCl93-shy II

65 1110 2T A 3 194 31 MOBU 111121

SS IOl7

A B

501

50

50

31

T

I

T 6

MtlllPllY SS114 B T 50 T T

PlllllIPS SS0149 A 5 55 33 T

SAMllDAN STln A 107 T T

SBllLL WDl103 A 65 223 2T WDl103 B 65 221 2T WD0104 c 65 221 2T WD0104 D 6T 20 2T

UNOCAL SS020I

SS 020I SSl20I

SS 0209

SS0209

II p

PURll

A

B

3

n 61

103

97

97

95

100

34

34

34

34 34

bullbull bullbullbull

- SS0209 D u 95 34 I

- SS 1209 SSl253

SS0253

G A

c

6T

2

100

165

1T5

34

54

54

bullbull I

I I

I I

SOUTH MARCH

~LAND

N

SOUTH PaTOI ENE ISWltD AREA

MMS Defined Limits ror Offshore Structures Requiring Inspections Following Andrew

t---Jbulln3 T25bull I bullTtl

bull1$1Htl1Abull bull1SIK1779bull

~

middot-DaTA AREA~T

CRANO rsLE

AREA bull1300

bullll4W

=shy

bullPath of Hurricane Andrew with Platforms Used In Calibration

Likelihood Function vs b All Success Case STISI K STIJOQ ST134W WD90A MCJll MCJ97

I 0000

0 9000

0 8000

07000

-Bo6000

tiiii 0 5000 -

04000

0 3000

0 2000

01000

0 0000

0 05 25

ri1rt middot-middot-middot-middot-middot-

I 5 2

b

---11 middot151K(blsu~ccsI

- bull - bull - fk 130Q(bfU~lCSS)

lie bull134W(blsuccc]

lk-WD90A[blsultCS]

- - - lk middotMC397(bfUtlCS]

--o-- lk-MCJll(bbuctcsJ

--- ll (bl 6 out of 6 Mlttcs case]

_

( Likelihood Functions middotSuccess Cases - p - 1 c I

CATEGORIZATION OF ADDITIONAL PLATFORMS

I I I

bull CATEGORIES

Unexpected Survivals

Expected Survivals

bull APPROACH

Identify Corresponding Platform(s) Evaluated In the JIP

Determine Max Wave Height at Platform Location

Evaluate Whether Wave Hit the Deck

middot - Evaluate Likelihood of R and S to Differ from the Base Cases

Review Joint Information

Assess Special Features

bull IDENTIFY NUMBER OF PLATFORMS IN EACH CATEGORY

I

UPDATING OF CALIBRATION RESULTS

bull INCLUDE EFFECT OF ADDITIONAL PLATFORMS ON bullBbull

Consider Increased 1Uncertalntles In R and S

Develop bullNeW Ukellhood Functions for Different Categories

middotLet bullxbull Number of Ukely Unexpected Survival Cases and y Number of Ukely Expected Survival Cases

Determine Combined Likelihood Function lk (b Iadditional platforms) middot

bull REVISED LIKELIHOOD FUNCTION

lk(b I all cases) = lk(b I 9 platforms) middot x lk(b Ix platforms) x lk(b I y platforms)

bull ESTABLISH REVISED BIAS FACTOR B

f8 (b) u f8 (b) lk (b I all cases)

II

t t ~

shy

APPLICATIONS OF BIAS FACTORS

ii APPLICABLE TO PARTICULAR LOAD AND RESISTANCE RECIPE

bull USE BY THE API COMMITTEES AND MMS

Criteria Revisions

Recommended Design Guidelines Revisions bull

bullbull REASSESSMENT STUDIES AND DECISION MAKING

middot

bull RISK ASSESSMENT STUDIES BY THE OPERATORS

Individual Platforms

Fleet of Platforms

middot IMR Program Implementation

bullbull NOT TO BE USED IN DESIGN

I I fl J

EXAMPLE USE OF BIAS FACTORS

bull PHASE I EXAMPLE APPLICATION

bull EVALUATION OF ANNUAL PROBABILITY OF FAILURE

Establish Annual Maximum Seastate Data

Determine Coefficients to Define Base Shear

Determine Ultimate Capacity of Platform

Determine Probablllty of Failure Versus bullbbull

-bull DETERMINE UPDATED PROBABILITY OF FAILURE

Pf - P [S gt R]

Pf I P [~ gt Rb I b] f 8 (b) db

Pr - J P [S gt Rb I b] f bull 8 (b) db

bull TOTAL PROBABILITIES OF FAILURE

Total Pf - 1 bull Il [1 middotPf] - l Pf bull

Total Pf - 1 bull Il [1middot Pf] - l Pfmiddot

Total Pr = 1 - n [1middot PfbullJ = l Pfshy

I I I

EXAMPLE APPLICATION bull PHASE I

PLATFORM ST 151 K

bull GULF OF MEXICO

bull WATER DEPTH = 137 FT

bull ANNUAL MAXIMUM HS MEDIAN 175 ft COV 031

bull ASSUMED STORM DURATION 3 hoursmiddot

bull ASSUMED NUMBER OF WAVES 800

bull MEDIAN HS FOR STORM HOURS 1 2 AND 3

bull MEDIAN CURRENT VALUE bull

bull DETERMINE ULTIMATE CAPACITY FOR THREE DIRECTIONS

bull DETERMINE ANNUAL FAILURE PROBABILITIES

Broadsidemiddot Diagonal End-on

0037 0025 0007

0071 0044 0018

0022 0016 0004

DETERMINE TOTAL ANNUAL PROBABILITY OF FAILURE

Pf - 007

Pf - 013

Pr - 004

bull

EXAMPLE USE OF BIAS FACTORS - PHASE II middot

bull DETERMINE PREDICTIVE PROBABILITY OF FAILURE

Case - 1

Pf _ S S Pf (b b2) fa1a2 (b b2) db db2

g bull -1 (b1 R I S) - 1

92 - (b2 R2 I S) - 1

Case -2

Pf s ~ [bJ f11 (b) db

Pf [bJ p [g lt 0)

g - bull 1 (b1 R1 I S) - 1

1

bull1) l

I

BENEFITS OF ANDREW JIP bull PHASE II

lt

bull IMPROVED LOAD AND RESISTANCE RECIPE

bull IMPROVED CALIBRATION PROCEDURE

bull MULTIPLE BIAS FACTORS

bull PROCEDURE TO INCLUDE EFFECT OF OTHER PLATFORMS

bull PROCEDURE TO USE MULTIPLE BIAS FACTORS

  • Structure Bookmarks
    • ANDREW PHASE II JOINT1 INDUSTRY PROJECT
    • PROGRESS MEETING NUMBER 1
    • ANDREW PHASE II JOINT INDUSTRY PROJECT PROGRESS MEETING NUMBER 1 NOVEMBER 16 AGENDA
    • ANDREW PHASE II UST OF PARTICIPANTS
    • ANDREW PHASE I OBJECTIVES AND RESULTS
    • ANDREW PHASE II PROJECT OBJECTIVES
    • SUBSEQUENT PROJECTS
    • ANDREW PHASE II PROJECT TEAM
    • ANDREW PHASE II UPDATE MODELING middotAND ANALYSIS RECIPE
    • -ANDREW PHASE II CONDUCTOR RESISTANCE
    • ANDREW PHASE Il WA VE FORCE PROFILE
    • ANDREW PHASE II WA VE FORCE PROFILEmiddot
    • ANDREW PHASE II BRACE MODELING
    • ANDREW PHASE II FOUNDATIONS
    • ANDREW PHASE II middot middot JOINT MODELING
    • ANDREW PHASE Il JOINT MODELING OPTIONS DISCUSSION
    • ANDREW PHASE II JOINT MODEJJNG OPTIONS ILLUSTRATIONS
    • ANDREW PHASE II JOINT MODELING PHASE II METHOD
    • ANDREW PHASE II ANDREW STORM CONDmONS UPDATED
    • Platform Selection bullbullbull1111111111111
    • BasisCriteria for Selection bullbullbull1111111111111
    • CALIBRATION
    • WHERE TO INTRODUCE BIAS FACTOR (B
    • PHASE I -KEY CONCLUSIONS
    • bull CALIBRATION ISSUES PHASE II
    • PHASE II WORK SCOPE (CALIBRATION)
    • SOFTWARE
    • SENSITIVITY STUDIES POSTERIOR DISTRIBUTION OF B
    • WEIGHTING APPROACH bull
    • INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON
    • CATEGORIZATION OF ADDITIONAL PLATFORMS
    • UPDATING OF CALIBRATION RESULTS
    • APPLICATIONS OF BIAS FACTORS
    • EXAMPLE USE OF BIAS FACTORS
    • EXAMPLE APPLICATION bull PHASE I PLATFORM ST 151 K
    • EXAMPLE USE OF BIAS FACTORS -PHASE II middot
    • BENEFITS OF ANDREW JIP bull PHASE II
Page 28:  · . ' ANDREW PHASE II PROJECT OBJECTIVES . ' • ·UPDATE MODELING AND ANALYSIS RECIPE· JOINTS ' ' ·BRACES WAVE PROFILE FOUNDATIONS . ' ' CONDUCTORS· • ANALVZE PLATFORMS USING

1 t I I ~

ANDREW PHASE II

BRACE MODELING

bull MARSHALL STRUTS USED FORPHASE I

bull PHASE II WILL ASSESS DIFFERENCE IN USING BUCKLING BEAMS AND MARSHALL STRUTS

middot bull COMPARE FOR 2D FRAME

bull COMPARE FOR PLATFORM 177 WHERE LOWER

BAY EXHIBITS LARGE BRACE MOMENTS

Project andrew Model beamframe Version 1 Mon Nov 14 145447 1994

~

middot Frame Test buckling beam-column model

-

Project andrew Model strutframe Version 1 Mon Nov 14 145754 1994

bull

Frame Test strut model

middot-middot _____________--_

_middotmiddotr_~ --ll--~-middot__

Buckling Beam bull

Strut (K=O 65)

Lateral Displacement

ANDREW PHASE II

FOUNDATIONS

bull ~LICIT NONLINEAR API FOUNDATION USED INiPHASE I

bull FOUNDATIONS

JIP SELECTED MINI-VANE DATA FOR

STRENGTH PROFILE

bull COMMENTS ON BENCHMARK COMPARISONS

- SIGNIFICANT VARIABILITY DUE TO FOUNDATION MODELING METHODS

II Ii

z

ctV 4x SPlO caissonfy-42s soil-api(lf-097step-114)

Inelastic Events Legend Elastic Strut Buckling

Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield

--------shyBeam Clmn Fully Plastic Fracture

~--~~~~~~~~~--~----~__i~~---~~~~__--~

Pro)ect apipile Model South Pelto 10 Version Wed May 18072235 1994 - -

1bullt

+

t

f i bullbullbull

bullbull bullbull

I I

5

~x SPlO Caissonfy=42s soil=api(lf=lOOstep=137)

Inelastic Events Legend

Elastic Strut Bucklilg

Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield

earn Clmn Full~ Plas1c Fracture

ANDREW PHASE II middot middot

JOINT MODELING

middotbull API RIGID-PLASTIC USED IN PHASE I

a ELASTIC-BRITTLE USED IN PHASE IB

amiddot MSL WILL PROVIDE NONLINEAR FORCE-DEFORMATION

AND MOMENT-DEFORMATION RELATIONS FOR PHASE II

a MODELING OPTIONS FOR PHASE II -

l

- UNCO~LEDmiddotONE DIMENSIONAL MEMBER END JOINT MODELING

J Ii

ANDREW PHASE Il

JOINT MODELING OPTIONS DISCUSSION

bull JD COUPLED INTERACTION SURFACE

- IDEAL

- DATA NOT AVAILABLE TO CALIBRATE

bull 2D COUPLED INTERACTION SURFACE

- SIGNIFICANT MODELING IMPROVEMENT

- DATAAVAILABLE

- IMPLEMENTATION PRACTICAL BUT NEEDS DEVELOPMENT AND CALIBRATION

bull 2D UNCOUPLED INTERACTION SURFACE

- SIGNIFICANT MODELING IMPROVEMENT

- DATA AVAILABLE

- IMPLEMENTATION PRACTICAL AND AVAILABLE

ANDREW PHASE II

JOINT MODEJJNG OPTIONS ILLUSTRATIONS

P

e

ANDREW PHASE II

JOINT MODELING PHASE II METHOD

e

I I I

ANDREW PHASE II

ANDREW STORM CONDmONS UPDATED

bull -PHASE I USED OCEANWEATHER HINDCAST-(Nov 1992)

bull middotmiddotPHASE II WILL USE THE REVISED HINDCAST (Nov 1994)

bull lt c lt I lt

2-2

l---4-~~~+-~~-+-~~~1--~~-+-~~--l--~T--+-~~--i24

f-tJVtMfiz~ 1Cfl (r STvb(

Plollol bullbull 5middotNOYmiddot9214bull1J Irbullbull Illbull [ANDREVl~NDREVCUSTbullMAXll 5-NOY-1992 10bull54 - 9208

HEIGHT lml

I~ MAXI

I L 1~middotamp SZ~- ~A I 13c

I J) ( r I X~0iWJ~0gtJ gt L gt gt I 1c I 120

-- N

~ deg

I I I I I I e C I ~=-41 122 -94 -92 -90 -88 -86 -84 -82 -80

-

Figure 7 contours of maximum hindcast significant wave height

_

~

~tl ~

NOltNttIJ 17lt ltNS S iJJ)y

Maximum Significant Wave Height - Hurricane Andrew 1992

30 ~

26 ~ I ~-~~~- == -shy -~ 35-_

24deg1--shy --------~- middot-middot-------middotmiddotmiddot----middot------middot-middot~middot- - - ~-----1---middot

22middot ~~~~~~~-~ -94 -92 -90 -88 -84deg-86 -82 -somiddot

c

-

Platform Selection bullbullbull1111111111111

0 BasisCriteria for Selection middot

0 Review of Phase I Platforms middot 0 Alternative Platforms

0 PMB Recommendations

0 Participants Comments

gt

L

l I

BasisCriteria for Selection bullbullbull1111111111111

0 Benefit for Calibration - Classification (UnexpectedExpected)

- Failure Mode (StructuresFoundations)

- Damaged StructuresKnown response

- Better Representation of Full ~opulation

0 Availability of Data - Existing Models

- Inspection Data

- Damage Data

PREDICTED BEHAVIOR

-

OBSERVED

BEHAVIOR

-SURVIVED DAMAGED FAILED

SURVIVED

ST134W WD90A

-MC311 -

MC397 -

--

-

-

ST151K ST130Q

-

-

-

DAMAGED - -

-

-

--

-

-

ST52 SS139 ST177B -

ST161A

-_

-

FAILED

-

-

-

ST151H ST130A ST72

Operator - Platfonns - Platfonns Survived Damagedamp

- Falled -

bull

-

-

Jle eBDIGIPllDbl -Amocoshy 30 1 Chevron -

143 12 Exxon 72 Mobll 22 1 Phllllps 7 bull

-Shell 39 Unocal 25 1

Total Partlclnts Platfonns 338 15

Other Operators bull- 310 13

-

Total Platfonns - -

648 28

fbml

Platforms Considered 6 7

tie of Total Platfonns 1 25

OVERVIEW OF PLATFORMS SURVIVEDDAMAGEDFAILED

-

shy

v

Platforms Evaluated in Phase I

-

-Platfonn Water Year Number Analysis Avallabillty of Availability Platform

- Name Depth rl of Model DetaDed rl ClassUlcatlon

Installation legs Inspection Soll Report Based on Information Phase I Calibration - -ft

- -- Suaill flalfnrm ~BSi -STISIK 137 1963 8 CAP Yes Yes Un eel Sarvlval

-STIJOO 170 1964 4 CAPmiddotshy Yes ST134data u ST134W 137 1981 4 CAP Yes Yesmiddot E

-

edSarvlval -

middotsurvival

WD90A 184 - 1964 8 CAP Yes E -

- Sarvlvill

MC3ll 343 1978 8 KARMA

MC397 468 1991 4 KARMA -

bull ~reglmmiddot- -

Yes Sure Survival

bull Yes - Sare Sanlval-

-T23STS2 63 1969 4 CAP

T2SSS139 62 1969 4 CAP -

Yes

Yes

Yes Ex

Yes Une

bull SurvivalLikely to DamalI

middot bull SanlvalLlkely to DamalI

ST161A 118 1964 8 USFOS

Eail11a pound1al(bullllm Clss ST177B 142 --1965 8 CAP

STISIH 137 1964 - 8 CAP C

STIJOA 140 19S8 8 CAP - -

Yes

Yes

ollamed

Yes

- Yes E

-

at about 1 mUe

Yes

ST134data

bull SafvlvalLlkelv to Dam- bull- -

E middotrahre

Emeried Failure -

Likely FaUare

T21ST72 61 1969 4 CAP Collamed No Likely FaUure

Platforms Identified for Evaluating Their Likelihood of Falling Under UnexpeCted Survival Category

CllEllATOa

AllB

llLOCll

Sl1lvcnJllE

NAME

YEAR

INSTAUJCD

WATER

IJBPJB

middotmiddotDISTANCE

ftOMSllOllE

(MIUIS)

Ill

(M)

AMOCO BllW

STllI

WDI075

A

bull D

4

n uo 17J

44

32

bull 20

10

bull CllEVllOH STtllO c I 1bull 21 11

STl130 D 0 161 21 11 STllJ4 I 0 137 D - 11

STIUI G I 137 32 11

STIUI I 0 121 32 11

STIUI

J 0 140 32 11

STIUI It 4 137 32 11

STllSI PROl)I 0 137 32 11

STllSI PROD-2 0 137 32 11

STll7 A 0

140 35 11

WDl117 c 65 214 34 10

DXON STshy E 51 14 bull STl165 A t3 bull 34 WD0073

WD0073

WD0073

A _ c

4

62

4

IA IA

17J

22

22

22

bullbullbull WD003-shy E 65 161 Z7

A 0 IN 31 MOBD Bl12I

ssn A bull

50

50

31

7

bull 7

MIJllPllY SS114 B 7 50 7 7

PlllUIPS SSl14t A 55 33 7

SAMEDAN ST17J A 4 117 7 bull 7

SJIELL WD0103 A 65 223 27 WDllOJ bull 65 221 27 WDll04 c 65 221 27 WDOI04 D 67 2 27

UNOCAL

SSl20I

SSl20I

bull SSl20I

SS0209

SS0209

SS0209

SS0209

SS 0253

SS 0253

E

I

PtJllE

A

bull D

G A

c

0

4

4

7J

61

0

0

69

103

97

97

9S

100

95

100

165

175

34

34

34

34

34

34

34

54

54

bullbullbullbullbullbullbullbullbull

Platforms Recommended for Phase II

llalr y_ w- Aot Av- _ I 1 Doplla Model nm bull bull _ bull I I H11 s18-1 Irht lbwl~ bullbullbull(-illtdlel OM lllMtD

II ---

-STISIIC 137 1963 I CAP Yeo Yeo I Yeo

STtshy -170 1964 4 CAP Yeo ST134dlU I Yeo

STl34W 137 Yeo Yeo -1981 4 CAP D Yeo

WD90A 114 1964 I CAP Yeo - D

MC311 30 1971 I KARMA Yeo -sre-

MC397 middot 461 1991 4 ICARMA Yeo Sireshy - - r- -

__ T13ST5l 63 1969 4 CAP Yeo Yeobull middot- ~n- D Yeo _ Tl5SSl39 62 1969 4 CAP Yeo Yeo I Yeo

_ STIT711 142 1965 I CAP Yeo I Yeo middot-1shySTl61A Ill 1964 I USFOS Yeo Yeo - D bulloshy

- r-

STISIH 137 1964 I CAP y D y -_ STl31A 140 1951 I CAP Yeo STl34dlU I Yeo

T21Sll 61 1969 4 CAP No rar-middot I Yeo

~-rmiddot-middot SS209A 95 1967 16 CAP Yeo D -SSU4R

11 UI A 41 -in D SS 2741AIM CAP bull

_ _ -shy - -shy- -shy

CALIBRATION

bull PHASEIAPPROACH

bull CALIBRATION ISSUES

bull PHASE II WORK SCOPE

bull SOFlWARE AVAILABILITY

bull EFFECT OF ADDITIONAL PLATFORMS

bull WEIGHTING APPROACH

bull EXAMPLE APPLICATION middot

(l=Ofce Mode] Wind

~

fcurrent l -middot

~SW ~leldl4

ultimate resistance force I first

design load Illgt

i _

deflection

-

bull

4

Historical Hurrlcane Dariiage and Survival of Platforms

Components Required for Calibration (Petrauskas 1992)

PRIOR Distribution

PLATFORM EXPERIENCE

BAYESIAN Survivals and Failures

UPDATING

Waves Winds and Currents bull

POSTERIOR Distribution

- A_middotJA ~ Survival Data Failure Data

Result Result

Load Resistance

middotfl

-

Bayesian Updaing middotGeneral Description (Petrauskas 1992)

L Imiddot

WHERE TO INTRODUCE BIAS FACTOR (B

On Resistance (Rmue - B (RPREDICreD I

middotOn Loads (Smue middot - B (SPRE01cTEo

On Safety Margin (RSmue - B (RSPREDICTED

CALIBRATION METHODOLOGY

gt

r

FIELD OBSERVATIONS

Observed Behavior

survivals Damages

Failures

r

bull

CAPACITY ANALYSIS

Predicted Load and Resistance Predicted Failure Modes

Survival Damage Failure cases

RELIABILITY ANALYSIS

- Probability of Failure ~

Ukellhood Failure

BAYESIAN UPDATING

Prior Distnbution of B Combined Likelihood Function

Posterior Dlstnbution of B

BIAS FACTOR

Mean Value cov bull

Represents Modeling Uncertainties

Prior Distribution of B Likelihood Functionmiddot Observing Success case STISIK and Posterior Distribution of B

2 00

uo

160

140

120

e 100 L $ uoIshy

0 60 ~ ~ 040

020

0 00

0 00 0 50

_ I 00

b

I 50 2 00

-shy lk [bl STISIK)

--shy Pnor-f(b)

--shy Pos1enor r(bl ISIK)

Prior or B Mean bull 10 COV bull 030

Posterlor of Bbull Meanbull 136COV 0 16

2 50

Likelihood Function and Posterior Distribution Example Platform STISIK

- -

~

- Prior and Posterior Distributions or B Failure cases (ST177B STlSlH ST130A T21) Damage cases (T23 T25 ST161A) Success cases (STlSlK ST130Q ST134W WD90A MC311 MC397)

400

350

300

2 so 0-

I

bull bull

bull bull gt

- bullbull ~ bull bull I bull

_ bull bull bullbull I bull bull bullII V -

bull bull

I bull ~ bull

bullbull bullI bull bullI bull bull bull

middot -~

bull__ ~ middotI middotmiddot

L 2 00

e ~ I SO

-I 00

- 0 so

000 gt

0 00 020 0 40 060 0 80 I 00 120 140 160 110 200

b -

-

shy

Prior-r

- - - - Posterior fLlll failure cases

bull bull bull bull bull bull bull Posterior flall damage cases

- bull bull - bull middot Postenor flall success cases

Posterior flall success fulure damage cases

-

Posterior of B Mean= 119 COV 0101

Posterior Distribution or Blas Factor (8) bull Different Categories and All Cases Combined

shy

l

PHASE I - KEY CONCLUSIONS

bull GLOBAL CORRECTION FACTOR

bull 15 TO 20 CONSERVATISM MEAN)

bull UNEXPECTED SURVIVALS HAVE GREATEST IMPACT

bull DAMAGED CASES MORE INFORMATIVE

bull CALIBRATION ISSUES PHASE II

bull

EFFECT OF IMPROVED CAPACITY ANALYSIS RECIPE

bull APPLICABILITY TO MULTIPLE FAILURE MODES

bull NO FOUNDATION DAMAGE CASE INCLUDED

bull IMPROVED REPRESENTATION OF FULL ANDREW DISTRIBUTION

-bull NUMBER OF PLATFORMS CONSIDERED

I 1 I

I t ) bull ~

PHASE II WORK SCOPE (CALIBRATION)

bull

bull ESTABLISH MULTIPLE BIAS FACTORS

Use Nine Base Case Platforms

middot - Capacity Analysis for Revised recipe

Foundation Blas Factor from Foundation JIP

bull DEVELOP WEIGHTING APPROACH

bull UPDATE MULTIPLE BIAS FACTORS

bullEffect of Platforms Not In the Nine Base Cases

bull EXAMPLE APPLICATION OF BIAS FACTORS I

bullbull

PHASE I PHASE II

CAPACITYbull ANALYSISRu bull

---------------------1----middot

Jc1 - iJbullraquoJ dbull

FttlS frt[f~l ratdhJJ

-

LIKELIHOOD FUNCTION

DEVELOPMENT

lk Jb Jfall1N) bull Jb)

111 Ill Isurvtvoq bull 1 - bl

-middot -------------------shy ~---------------------middot-middot

r111ir8 Cb1 111 Cbl BAYESIAN

UPDATING

~cove

R1R2 bull middot---------------------~----

Jblbull P(g cOJ

1 bull JbRI (MBSI I bull 1

~middot

llcJllJfalluro) bull P(gcOJ Ill 111 b21fallun) bull P ((g1 CO) U 1112 c OD

middot--------------------shy ----------------------shy bull

r111111r8 Cbl 111 Ill i-1nro1

deg ~1b2)laquottfrr (b1b21-lnfo) re Ill bull 1-11middot b2) bull lodegamp

middot

i1 covamp i2 cov~i

-

shy

--

Likelihood Function vs b All Success Case STISI K STl30Q STl34W WD90A MC311 MC397

I 0000

0 9000

OIOOO

0 7000

~ 0amp000

ta osoOO

~ 0 4000

0 3000

0 2000

0 1000

bull 0 0000

bull

I

middot

middot bull I

Ibull

-sibull ~ -middot-middot-middot-middot -middot-

----7----+-----~ I

Ibull bull bull I

rbull

--- Jk ISIK[blsuccess)

- bull - bull - lk 130Q[blsuccess) bull

lk 134W[blsuccess)

Jk-WD90A(blsuccc~I

- - - lk MC397[blsucccss)

--ltgt-- lk middotMC31 l[blsucccss)

--- Jk [bl 6 out of6succcsHascs)

05 I 5 2 250

b

Likelihood Functionsmiddot Success Cases - Phbullsc L

shy

bullbull bull bull bull bull bull -------------------~

bull bull bullbull bullbull bullbull bullbull bullbull bullbull bull

bull bullbull

bullbullbullbull

bullbull

bullbullbullbullbullbullbull

bullbullbullbullbullbull

bullbullbullbullbullbullbull

z 0

t z ) LL

c 0 0c i ~ i

J

~ sect ~--------------~ e l5

I

I

SOFTWARE

bull PF PROGRAM

bull CALREL

bull PF +CALREL

bull RELACS

bull

gt i

l

SENSITIVITY STUDIES

POSTERIOR DISTRIBUTION OF B

bull EFFECT OF CHANGE IN DISTRIBUTION OF PRIOR OF B

bull EFFECT OF REMOVAL OF ONE PLATFORM FROM EACH CATEGORY

bull EFFECT OF REMOVAL OF THREE DAMAGE CASES

bull EFFECT OF ADDITIONAL 13 SIMILAR PLATFORMS

-I

Prior and Posterior Distributions of B Variation In COV of PriorofB fromOl to04 -- -

4500

-4000

-

- -

3500 --middot - bull bull - bull bull Prior-r ICOV =02

3000

- - - - Prior-r ICOV =03 -

bull -~ e 2soo -

bull bull bull bull bull bull bull Pnor-r ICOV =o 4t frJ ~ 2000 Posterior rlCOV =0 2

bull bull Postenor rlCOV =0 3 bull bull 1500 bull middot bull - shy -middot- Posterior rlCOV =04

middot~ I 000 - bull

-middotmiddot ~ ~middotmiddot M -~

- middot 0 500

wbullbullbull _ I middot~ middotmiddotmiddotmiddotmiddot middotmiddotmiddotmiddotmiddot - -~ ~ __ middot_ _ bull -shy0 000

0 000 0200 0 400 0 600 0 800 IOOO 1200 1400 1600 1800 2000 -

- - b

-

=shy

Posterior Distribution of Bias Factor (B) bull Effect of Variation In COV of Prior of B

bull

bullPrior and Posterior Distributions of B Evaluating Effect of Removal of One Platform from Each Category from the Base Case of 13 Platforms

400

~ ~ 350

j IA i

3 00 bull I bull I ---Prior-r I bull l

250 1 I 1

I bull bull ---Posterior r lall 13 casesshy- ~E 2 00 bull bull bull bull bull bull bull Posterior r without STISIK

- ~ 1 bull - bull I - bull bull - bull bull Posterior r Iwithout T2I I 50 I

J-tt~i - bull - bull - Posterior r Iwithout T2S

tOO [-

I I ~ ~ I 0 50

0 00

___ r J i

l~T

~--

0 00 0 20 0 40 0 60 080 I 00 120 140 160 110 2 00

b

zshy

Posterior Distribution of Bias Factor (8) bull Effect of Removal of middotPlatform from Any Category

bullPrior and Posterior Distributions or B Evaluating Effect or Removal or 3 Damage Cases from the Base Case or 13 Platrorms

4

35

3

25

e L 2 ~ bulle shy

I 15

bull I

05

0

0 02 04 06 I 2 I 4 16 IS 2

I I

I

I

I Ibull

J -tmiddot L--

bullI

~

I

fl

II ~

I I v I

middot bull bull

bull L

~ ll- I-~lt

OS I

b

I

bull

1middot bullbullbullbull

~ __

bull

- - - - Prior-r

- - - - Postenor flall failure cases

Postenor flall success cases

--- Posterior flall 13 cases

---Posterior fI success failure cases only

IPosterior or B Mean =t11 COV =0121

-

_I

Posterir Distribution or Bias Factor (B) bull Effect or Removal or 3-Damage Cases

-

bullPrior and Posterior Distributions or B Evaluating Effect or Additional 13 Similar Observations as

Used in Andrew JIP The POsterior or B from Andrew-JIP Considered as New Prior or B

- 500

450

400

350

bull bull

bull bull bull bull J bull

bull bull f ~ I

I ~ Ibull bull

bullbull bull

l bull I -bullbull bull bull

I

I bull bull

bull I I~~bull bull bull

bull ~

I bull bull

~ ~_ ~~~

- - New Prior r bull

Posterior 114 new failure cases ~ 300- bull bull bull bull bull bull bull Posterior f 13 new damage casesl 150 ~

- bull bull - bull bull Posterior 116 new success cases i 2 00

Postenor 1113 new success failure -150 damage cases

I 00

0 50 -

0 00 0 00 0 20 0 40 0 60 0 80 I 00 120 I 40 160 110 2 00

Posterior ol B Mean= 120 COV= 0071 b

shy

shy

_rshy

Posterior Distribution of Bias Factor (B) bull Effect of Additional 13-Similar Observations

1

1 I

WEIGHTING APPROACH bull

bull ASSESSMENT OF PLATFORMS DATA

Select Platforms Most Mappable to the Analyzed Platforms

Unlikely SurvivalDamage Cases Identification

Based on Cut-Off Criteria Agemiddot Hs etc

bull INFORMATION REQUIRED FOR COMPARISON

bull CATEGORIZATION OF ADDITIONALmiddot PLATFORMS

bull UPDATING OF CALIBRATION RESULTS

--

OVERVIEW OF PLATFORMS SURVIVEDDAMAGED -~

Operator -

- Platforms

Survived CS)

Platforms

Damaged CD)

Platforms Used In

Phase IJIP -

JI~ ~Bill~IRBDll -

Amoco -

Chevron -

Exxon

Mob II

Phllllps

Shell

Unocal -

30

143

72 22

7

39 25shy -

1

12

bull

1 bull

1

1(S)+1 (D)

3 (S) +3(D)

2 (S)

bull shy

bull

bull

bull

Total Partlclnants Platforms 338 - 15 10C6S+4D) -

Other Operators

310

13

3 (D)

Total Platforms 648 28 13C6S+7D) -

Summary of Andrew JIP bull Phase I Platforms

Plairshy w v- N-ber Lea-le 1 shy car Hind_ o-w N-bull Deplh fl fl IJpe la bull n Yeadon Jadtet lbelledtf liefalit -- _ -OD

IL H - -shy _ICllshy IL IL - - rQClto ~ _

K STUIK 137 1963 8 365 608S4 Cellardeck

-middot STl-shy 170 1964 4 -~ K 39 62269 Cellardeck - --ST134W 137 1911 4 43 608S4 Smvlnlshy- _ - -

WD90A 184 1964 8 K 267Sbull34 83 SOS31 SUboftllar deck SmvMI

MC311 343 1978 8 x S7 62 - MC397 468 1991 4 x shy S33bull6SO 65911 -shySmeSmvlvol -- gt -

r-- T23ST52 63 1969 4 ton- shyKshy 52594 S03 -T25SSl39 62 1969 4 Kshy SOS94 S063 - middot ton-shy__ -

bullbull a ton--shyST161A 118 1964 8 K - 347S 51495 Cellarclecti

r~

ST1771 142 1965 8 K 39 60167 CeDordeck failme

STISIH 137 1964 8 K 395 608S4 CeDardeck Failme

K ST130A 140 1958 a SS 6095 I lblv Pai1uno - T21ST72 61 1969 4 Kshy S4S94 4974 I lhlv Pai1uno -

- -

JE

INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON

bull WATER DEPTH

bull bull YEAR OF INSTALLATIONDESIGN

bull PLATFORM ORIENTATION

bull PLATFORM LATITUDE AND LONGITUDE

bull KEY SKETCHES OF PLATFORM STRUCTURAL FRAMING

Number of Legs Plies Leg Batter Number of Conductors bull Deck Elevations (Including sub-cellar deck If any) Jacket Framing Elevations Brace type In Vertical Frames Sizes of Primary Members (leg pile vertical frames braces primary horizontals) Joint Details (can gap etc) LegPiie Annulus Grouting Status

bull middotPOST ANDREW INSPECTION REPORT Significant Wave Height Hs (ft)

bull INFORMATION ON DAMAGES PRIOR TO ANDREW

bull RESULTS FROM IN-HOUSE STUDIES IF AVAILABLE

bull ~RIGINAL DESIGN CRITERIA INFORMATION IF AVAILABLE

bull SOIL SHEAR STRENGTH PROFILE IF AVAILABLE

Platforms Identified for Evaluating Therr Likelihood of Falling Under Unexpected Survival Category

OPlRATOR

middotADA

BLOClt

llmucnJllE

NAME

YEAll

INSTAUJD

WA1ER

DBPTll

DISTANCE

JlllOM SHOllE

(MILU)

Bl

(M)

AMOCO 111113

STl11

WDI075

A

bull D

St

17

120

1n

44

20

10 I

CllllVllQfl STlllO c 61 110 21 11 STIUO D Q 1110 21 11

STl134 p Q 137 29 11

ST0151 G 61 137 11

ST0151 I 3 121 11

ST0151 J Q 140 11

STl151 It 137 11

ST0151 PJl001 Q UT 11

ST0151 PROD-2 Q UT 11

ST017 A 3 140 35 11

WDIUT c 65 214 34 10

llXXON STI055 II 51 6T 14 I

STl15 A t3 34

WDI073

WD0073

Amiddotshy Q

1111 1111

22

22

I

I

WDI073 c 1n 22 I

- WDOCl93-shy II

65 1110 2T A 3 194 31 MOBU 111121

SS IOl7

A B

501

50

50

31

T

I

T 6

MtlllPllY SS114 B T 50 T T

PlllllIPS SS0149 A 5 55 33 T

SAMllDAN STln A 107 T T

SBllLL WDl103 A 65 223 2T WDl103 B 65 221 2T WD0104 c 65 221 2T WD0104 D 6T 20 2T

UNOCAL SS020I

SS 020I SSl20I

SS 0209

SS0209

II p

PURll

A

B

3

n 61

103

97

97

95

100

34

34

34

34 34

bullbull bullbullbull

- SS0209 D u 95 34 I

- SS 1209 SSl253

SS0253

G A

c

6T

2

100

165

1T5

34

54

54

bullbull I

I I

I I

SOUTH MARCH

~LAND

N

SOUTH PaTOI ENE ISWltD AREA

MMS Defined Limits ror Offshore Structures Requiring Inspections Following Andrew

t---Jbulln3 T25bull I bullTtl

bull1$1Htl1Abull bull1SIK1779bull

~

middot-DaTA AREA~T

CRANO rsLE

AREA bull1300

bullll4W

=shy

bullPath of Hurricane Andrew with Platforms Used In Calibration

Likelihood Function vs b All Success Case STISI K STIJOQ ST134W WD90A MCJll MCJ97

I 0000

0 9000

0 8000

07000

-Bo6000

tiiii 0 5000 -

04000

0 3000

0 2000

01000

0 0000

0 05 25

ri1rt middot-middot-middot-middot-middot-

I 5 2

b

---11 middot151K(blsu~ccsI

- bull - bull - fk 130Q(bfU~lCSS)

lie bull134W(blsuccc]

lk-WD90A[blsultCS]

- - - lk middotMC397(bfUtlCS]

--o-- lk-MCJll(bbuctcsJ

--- ll (bl 6 out of 6 Mlttcs case]

_

( Likelihood Functions middotSuccess Cases - p - 1 c I

CATEGORIZATION OF ADDITIONAL PLATFORMS

I I I

bull CATEGORIES

Unexpected Survivals

Expected Survivals

bull APPROACH

Identify Corresponding Platform(s) Evaluated In the JIP

Determine Max Wave Height at Platform Location

Evaluate Whether Wave Hit the Deck

middot - Evaluate Likelihood of R and S to Differ from the Base Cases

Review Joint Information

Assess Special Features

bull IDENTIFY NUMBER OF PLATFORMS IN EACH CATEGORY

I

UPDATING OF CALIBRATION RESULTS

bull INCLUDE EFFECT OF ADDITIONAL PLATFORMS ON bullBbull

Consider Increased 1Uncertalntles In R and S

Develop bullNeW Ukellhood Functions for Different Categories

middotLet bullxbull Number of Ukely Unexpected Survival Cases and y Number of Ukely Expected Survival Cases

Determine Combined Likelihood Function lk (b Iadditional platforms) middot

bull REVISED LIKELIHOOD FUNCTION

lk(b I all cases) = lk(b I 9 platforms) middot x lk(b Ix platforms) x lk(b I y platforms)

bull ESTABLISH REVISED BIAS FACTOR B

f8 (b) u f8 (b) lk (b I all cases)

II

t t ~

shy

APPLICATIONS OF BIAS FACTORS

ii APPLICABLE TO PARTICULAR LOAD AND RESISTANCE RECIPE

bull USE BY THE API COMMITTEES AND MMS

Criteria Revisions

Recommended Design Guidelines Revisions bull

bullbull REASSESSMENT STUDIES AND DECISION MAKING

middot

bull RISK ASSESSMENT STUDIES BY THE OPERATORS

Individual Platforms

Fleet of Platforms

middot IMR Program Implementation

bullbull NOT TO BE USED IN DESIGN

I I fl J

EXAMPLE USE OF BIAS FACTORS

bull PHASE I EXAMPLE APPLICATION

bull EVALUATION OF ANNUAL PROBABILITY OF FAILURE

Establish Annual Maximum Seastate Data

Determine Coefficients to Define Base Shear

Determine Ultimate Capacity of Platform

Determine Probablllty of Failure Versus bullbbull

-bull DETERMINE UPDATED PROBABILITY OF FAILURE

Pf - P [S gt R]

Pf I P [~ gt Rb I b] f 8 (b) db

Pr - J P [S gt Rb I b] f bull 8 (b) db

bull TOTAL PROBABILITIES OF FAILURE

Total Pf - 1 bull Il [1 middotPf] - l Pf bull

Total Pf - 1 bull Il [1middot Pf] - l Pfmiddot

Total Pr = 1 - n [1middot PfbullJ = l Pfshy

I I I

EXAMPLE APPLICATION bull PHASE I

PLATFORM ST 151 K

bull GULF OF MEXICO

bull WATER DEPTH = 137 FT

bull ANNUAL MAXIMUM HS MEDIAN 175 ft COV 031

bull ASSUMED STORM DURATION 3 hoursmiddot

bull ASSUMED NUMBER OF WAVES 800

bull MEDIAN HS FOR STORM HOURS 1 2 AND 3

bull MEDIAN CURRENT VALUE bull

bull DETERMINE ULTIMATE CAPACITY FOR THREE DIRECTIONS

bull DETERMINE ANNUAL FAILURE PROBABILITIES

Broadsidemiddot Diagonal End-on

0037 0025 0007

0071 0044 0018

0022 0016 0004

DETERMINE TOTAL ANNUAL PROBABILITY OF FAILURE

Pf - 007

Pf - 013

Pr - 004

bull

EXAMPLE USE OF BIAS FACTORS - PHASE II middot

bull DETERMINE PREDICTIVE PROBABILITY OF FAILURE

Case - 1

Pf _ S S Pf (b b2) fa1a2 (b b2) db db2

g bull -1 (b1 R I S) - 1

92 - (b2 R2 I S) - 1

Case -2

Pf s ~ [bJ f11 (b) db

Pf [bJ p [g lt 0)

g - bull 1 (b1 R1 I S) - 1

1

bull1) l

I

BENEFITS OF ANDREW JIP bull PHASE II

lt

bull IMPROVED LOAD AND RESISTANCE RECIPE

bull IMPROVED CALIBRATION PROCEDURE

bull MULTIPLE BIAS FACTORS

bull PROCEDURE TO INCLUDE EFFECT OF OTHER PLATFORMS

bull PROCEDURE TO USE MULTIPLE BIAS FACTORS

  • Structure Bookmarks
    • ANDREW PHASE II JOINT1 INDUSTRY PROJECT
    • PROGRESS MEETING NUMBER 1
    • ANDREW PHASE II JOINT INDUSTRY PROJECT PROGRESS MEETING NUMBER 1 NOVEMBER 16 AGENDA
    • ANDREW PHASE II UST OF PARTICIPANTS
    • ANDREW PHASE I OBJECTIVES AND RESULTS
    • ANDREW PHASE II PROJECT OBJECTIVES
    • SUBSEQUENT PROJECTS
    • ANDREW PHASE II PROJECT TEAM
    • ANDREW PHASE II UPDATE MODELING middotAND ANALYSIS RECIPE
    • -ANDREW PHASE II CONDUCTOR RESISTANCE
    • ANDREW PHASE Il WA VE FORCE PROFILE
    • ANDREW PHASE II WA VE FORCE PROFILEmiddot
    • ANDREW PHASE II BRACE MODELING
    • ANDREW PHASE II FOUNDATIONS
    • ANDREW PHASE II middot middot JOINT MODELING
    • ANDREW PHASE Il JOINT MODELING OPTIONS DISCUSSION
    • ANDREW PHASE II JOINT MODEJJNG OPTIONS ILLUSTRATIONS
    • ANDREW PHASE II JOINT MODELING PHASE II METHOD
    • ANDREW PHASE II ANDREW STORM CONDmONS UPDATED
    • Platform Selection bullbullbull1111111111111
    • BasisCriteria for Selection bullbullbull1111111111111
    • CALIBRATION
    • WHERE TO INTRODUCE BIAS FACTOR (B
    • PHASE I -KEY CONCLUSIONS
    • bull CALIBRATION ISSUES PHASE II
    • PHASE II WORK SCOPE (CALIBRATION)
    • SOFTWARE
    • SENSITIVITY STUDIES POSTERIOR DISTRIBUTION OF B
    • WEIGHTING APPROACH bull
    • INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON
    • CATEGORIZATION OF ADDITIONAL PLATFORMS
    • UPDATING OF CALIBRATION RESULTS
    • APPLICATIONS OF BIAS FACTORS
    • EXAMPLE USE OF BIAS FACTORS
    • EXAMPLE APPLICATION bull PHASE I PLATFORM ST 151 K
    • EXAMPLE USE OF BIAS FACTORS -PHASE II middot
    • BENEFITS OF ANDREW JIP bull PHASE II
Page 29:  · . ' ANDREW PHASE II PROJECT OBJECTIVES . ' • ·UPDATE MODELING AND ANALYSIS RECIPE· JOINTS ' ' ·BRACES WAVE PROFILE FOUNDATIONS . ' ' CONDUCTORS· • ANALVZE PLATFORMS USING

Project andrew Model beamframe Version 1 Mon Nov 14 145447 1994

~

middot Frame Test buckling beam-column model

-

Project andrew Model strutframe Version 1 Mon Nov 14 145754 1994

bull

Frame Test strut model

middot-middot _____________--_

_middotmiddotr_~ --ll--~-middot__

Buckling Beam bull

Strut (K=O 65)

Lateral Displacement

ANDREW PHASE II

FOUNDATIONS

bull ~LICIT NONLINEAR API FOUNDATION USED INiPHASE I

bull FOUNDATIONS

JIP SELECTED MINI-VANE DATA FOR

STRENGTH PROFILE

bull COMMENTS ON BENCHMARK COMPARISONS

- SIGNIFICANT VARIABILITY DUE TO FOUNDATION MODELING METHODS

II Ii

z

ctV 4x SPlO caissonfy-42s soil-api(lf-097step-114)

Inelastic Events Legend Elastic Strut Buckling

Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield

--------shyBeam Clmn Fully Plastic Fracture

~--~~~~~~~~~--~----~__i~~---~~~~__--~

Pro)ect apipile Model South Pelto 10 Version Wed May 18072235 1994 - -

1bullt

+

t

f i bullbullbull

bullbull bullbull

I I

5

~x SPlO Caissonfy=42s soil=api(lf=lOOstep=137)

Inelastic Events Legend

Elastic Strut Bucklilg

Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield

earn Clmn Full~ Plas1c Fracture

ANDREW PHASE II middot middot

JOINT MODELING

middotbull API RIGID-PLASTIC USED IN PHASE I

a ELASTIC-BRITTLE USED IN PHASE IB

amiddot MSL WILL PROVIDE NONLINEAR FORCE-DEFORMATION

AND MOMENT-DEFORMATION RELATIONS FOR PHASE II

a MODELING OPTIONS FOR PHASE II -

l

- UNCO~LEDmiddotONE DIMENSIONAL MEMBER END JOINT MODELING

J Ii

ANDREW PHASE Il

JOINT MODELING OPTIONS DISCUSSION

bull JD COUPLED INTERACTION SURFACE

- IDEAL

- DATA NOT AVAILABLE TO CALIBRATE

bull 2D COUPLED INTERACTION SURFACE

- SIGNIFICANT MODELING IMPROVEMENT

- DATAAVAILABLE

- IMPLEMENTATION PRACTICAL BUT NEEDS DEVELOPMENT AND CALIBRATION

bull 2D UNCOUPLED INTERACTION SURFACE

- SIGNIFICANT MODELING IMPROVEMENT

- DATA AVAILABLE

- IMPLEMENTATION PRACTICAL AND AVAILABLE

ANDREW PHASE II

JOINT MODEJJNG OPTIONS ILLUSTRATIONS

P

e

ANDREW PHASE II

JOINT MODELING PHASE II METHOD

e

I I I

ANDREW PHASE II

ANDREW STORM CONDmONS UPDATED

bull -PHASE I USED OCEANWEATHER HINDCAST-(Nov 1992)

bull middotmiddotPHASE II WILL USE THE REVISED HINDCAST (Nov 1994)

bull lt c lt I lt

2-2

l---4-~~~+-~~-+-~~~1--~~-+-~~--l--~T--+-~~--i24

f-tJVtMfiz~ 1Cfl (r STvb(

Plollol bullbull 5middotNOYmiddot9214bull1J Irbullbull Illbull [ANDREVl~NDREVCUSTbullMAXll 5-NOY-1992 10bull54 - 9208

HEIGHT lml

I~ MAXI

I L 1~middotamp SZ~- ~A I 13c

I J) ( r I X~0iWJ~0gtJ gt L gt gt I 1c I 120

-- N

~ deg

I I I I I I e C I ~=-41 122 -94 -92 -90 -88 -86 -84 -82 -80

-

Figure 7 contours of maximum hindcast significant wave height

_

~

~tl ~

NOltNttIJ 17lt ltNS S iJJ)y

Maximum Significant Wave Height - Hurricane Andrew 1992

30 ~

26 ~ I ~-~~~- == -shy -~ 35-_

24deg1--shy --------~- middot-middot-------middotmiddotmiddot----middot------middot-middot~middot- - - ~-----1---middot

22middot ~~~~~~~-~ -94 -92 -90 -88 -84deg-86 -82 -somiddot

c

-

Platform Selection bullbullbull1111111111111

0 BasisCriteria for Selection middot

0 Review of Phase I Platforms middot 0 Alternative Platforms

0 PMB Recommendations

0 Participants Comments

gt

L

l I

BasisCriteria for Selection bullbullbull1111111111111

0 Benefit for Calibration - Classification (UnexpectedExpected)

- Failure Mode (StructuresFoundations)

- Damaged StructuresKnown response

- Better Representation of Full ~opulation

0 Availability of Data - Existing Models

- Inspection Data

- Damage Data

PREDICTED BEHAVIOR

-

OBSERVED

BEHAVIOR

-SURVIVED DAMAGED FAILED

SURVIVED

ST134W WD90A

-MC311 -

MC397 -

--

-

-

ST151K ST130Q

-

-

-

DAMAGED - -

-

-

--

-

-

ST52 SS139 ST177B -

ST161A

-_

-

FAILED

-

-

-

ST151H ST130A ST72

Operator - Platfonns - Platfonns Survived Damagedamp

- Falled -

bull

-

-

Jle eBDIGIPllDbl -Amocoshy 30 1 Chevron -

143 12 Exxon 72 Mobll 22 1 Phllllps 7 bull

-Shell 39 Unocal 25 1

Total Partlclnts Platfonns 338 15

Other Operators bull- 310 13

-

Total Platfonns - -

648 28

fbml

Platforms Considered 6 7

tie of Total Platfonns 1 25

OVERVIEW OF PLATFORMS SURVIVEDDAMAGEDFAILED

-

shy

v

Platforms Evaluated in Phase I

-

-Platfonn Water Year Number Analysis Avallabillty of Availability Platform

- Name Depth rl of Model DetaDed rl ClassUlcatlon

Installation legs Inspection Soll Report Based on Information Phase I Calibration - -ft

- -- Suaill flalfnrm ~BSi -STISIK 137 1963 8 CAP Yes Yes Un eel Sarvlval

-STIJOO 170 1964 4 CAPmiddotshy Yes ST134data u ST134W 137 1981 4 CAP Yes Yesmiddot E

-

edSarvlval -

middotsurvival

WD90A 184 - 1964 8 CAP Yes E -

- Sarvlvill

MC3ll 343 1978 8 KARMA

MC397 468 1991 4 KARMA -

bull ~reglmmiddot- -

Yes Sure Survival

bull Yes - Sare Sanlval-

-T23STS2 63 1969 4 CAP

T2SSS139 62 1969 4 CAP -

Yes

Yes

Yes Ex

Yes Une

bull SurvivalLikely to DamalI

middot bull SanlvalLlkely to DamalI

ST161A 118 1964 8 USFOS

Eail11a pound1al(bullllm Clss ST177B 142 --1965 8 CAP

STISIH 137 1964 - 8 CAP C

STIJOA 140 19S8 8 CAP - -

Yes

Yes

ollamed

Yes

- Yes E

-

at about 1 mUe

Yes

ST134data

bull SafvlvalLlkelv to Dam- bull- -

E middotrahre

Emeried Failure -

Likely FaUare

T21ST72 61 1969 4 CAP Collamed No Likely FaUure

Platforms Identified for Evaluating Their Likelihood of Falling Under UnexpeCted Survival Category

CllEllATOa

AllB

llLOCll

Sl1lvcnJllE

NAME

YEAR

INSTAUJCD

WATER

IJBPJB

middotmiddotDISTANCE

ftOMSllOllE

(MIUIS)

Ill

(M)

AMOCO BllW

STllI

WDI075

A

bull D

4

n uo 17J

44

32

bull 20

10

bull CllEVllOH STtllO c I 1bull 21 11

STl130 D 0 161 21 11 STllJ4 I 0 137 D - 11

STIUI G I 137 32 11

STIUI I 0 121 32 11

STIUI

J 0 140 32 11

STIUI It 4 137 32 11

STllSI PROl)I 0 137 32 11

STllSI PROD-2 0 137 32 11

STll7 A 0

140 35 11

WDl117 c 65 214 34 10

DXON STshy E 51 14 bull STl165 A t3 bull 34 WD0073

WD0073

WD0073

A _ c

4

62

4

IA IA

17J

22

22

22

bullbullbull WD003-shy E 65 161 Z7

A 0 IN 31 MOBD Bl12I

ssn A bull

50

50

31

7

bull 7

MIJllPllY SS114 B 7 50 7 7

PlllUIPS SSl14t A 55 33 7

SAMEDAN ST17J A 4 117 7 bull 7

SJIELL WD0103 A 65 223 27 WDllOJ bull 65 221 27 WDll04 c 65 221 27 WDOI04 D 67 2 27

UNOCAL

SSl20I

SSl20I

bull SSl20I

SS0209

SS0209

SS0209

SS0209

SS 0253

SS 0253

E

I

PtJllE

A

bull D

G A

c

0

4

4

7J

61

0

0

69

103

97

97

9S

100

95

100

165

175

34

34

34

34

34

34

34

54

54

bullbullbullbullbullbullbullbullbull

Platforms Recommended for Phase II

llalr y_ w- Aot Av- _ I 1 Doplla Model nm bull bull _ bull I I H11 s18-1 Irht lbwl~ bullbullbull(-illtdlel OM lllMtD

II ---

-STISIIC 137 1963 I CAP Yeo Yeo I Yeo

STtshy -170 1964 4 CAP Yeo ST134dlU I Yeo

STl34W 137 Yeo Yeo -1981 4 CAP D Yeo

WD90A 114 1964 I CAP Yeo - D

MC311 30 1971 I KARMA Yeo -sre-

MC397 middot 461 1991 4 ICARMA Yeo Sireshy - - r- -

__ T13ST5l 63 1969 4 CAP Yeo Yeobull middot- ~n- D Yeo _ Tl5SSl39 62 1969 4 CAP Yeo Yeo I Yeo

_ STIT711 142 1965 I CAP Yeo I Yeo middot-1shySTl61A Ill 1964 I USFOS Yeo Yeo - D bulloshy

- r-

STISIH 137 1964 I CAP y D y -_ STl31A 140 1951 I CAP Yeo STl34dlU I Yeo

T21Sll 61 1969 4 CAP No rar-middot I Yeo

~-rmiddot-middot SS209A 95 1967 16 CAP Yeo D -SSU4R

11 UI A 41 -in D SS 2741AIM CAP bull

_ _ -shy - -shy- -shy

CALIBRATION

bull PHASEIAPPROACH

bull CALIBRATION ISSUES

bull PHASE II WORK SCOPE

bull SOFlWARE AVAILABILITY

bull EFFECT OF ADDITIONAL PLATFORMS

bull WEIGHTING APPROACH

bull EXAMPLE APPLICATION middot

(l=Ofce Mode] Wind

~

fcurrent l -middot

~SW ~leldl4

ultimate resistance force I first

design load Illgt

i _

deflection

-

bull

4

Historical Hurrlcane Dariiage and Survival of Platforms

Components Required for Calibration (Petrauskas 1992)

PRIOR Distribution

PLATFORM EXPERIENCE

BAYESIAN Survivals and Failures

UPDATING

Waves Winds and Currents bull

POSTERIOR Distribution

- A_middotJA ~ Survival Data Failure Data

Result Result

Load Resistance

middotfl

-

Bayesian Updaing middotGeneral Description (Petrauskas 1992)

L Imiddot

WHERE TO INTRODUCE BIAS FACTOR (B

On Resistance (Rmue - B (RPREDICreD I

middotOn Loads (Smue middot - B (SPRE01cTEo

On Safety Margin (RSmue - B (RSPREDICTED

CALIBRATION METHODOLOGY

gt

r

FIELD OBSERVATIONS

Observed Behavior

survivals Damages

Failures

r

bull

CAPACITY ANALYSIS

Predicted Load and Resistance Predicted Failure Modes

Survival Damage Failure cases

RELIABILITY ANALYSIS

- Probability of Failure ~

Ukellhood Failure

BAYESIAN UPDATING

Prior Distnbution of B Combined Likelihood Function

Posterior Dlstnbution of B

BIAS FACTOR

Mean Value cov bull

Represents Modeling Uncertainties

Prior Distribution of B Likelihood Functionmiddot Observing Success case STISIK and Posterior Distribution of B

2 00

uo

160

140

120

e 100 L $ uoIshy

0 60 ~ ~ 040

020

0 00

0 00 0 50

_ I 00

b

I 50 2 00

-shy lk [bl STISIK)

--shy Pnor-f(b)

--shy Pos1enor r(bl ISIK)

Prior or B Mean bull 10 COV bull 030

Posterlor of Bbull Meanbull 136COV 0 16

2 50

Likelihood Function and Posterior Distribution Example Platform STISIK

- -

~

- Prior and Posterior Distributions or B Failure cases (ST177B STlSlH ST130A T21) Damage cases (T23 T25 ST161A) Success cases (STlSlK ST130Q ST134W WD90A MC311 MC397)

400

350

300

2 so 0-

I

bull bull

bull bull gt

- bullbull ~ bull bull I bull

_ bull bull bullbull I bull bull bullII V -

bull bull

I bull ~ bull

bullbull bullI bull bullI bull bull bull

middot -~

bull__ ~ middotI middotmiddot

L 2 00

e ~ I SO

-I 00

- 0 so

000 gt

0 00 020 0 40 060 0 80 I 00 120 140 160 110 200

b -

-

shy

Prior-r

- - - - Posterior fLlll failure cases

bull bull bull bull bull bull bull Posterior flall damage cases

- bull bull - bull middot Postenor flall success cases

Posterior flall success fulure damage cases

-

Posterior of B Mean= 119 COV 0101

Posterior Distribution or Blas Factor (8) bull Different Categories and All Cases Combined

shy

l

PHASE I - KEY CONCLUSIONS

bull GLOBAL CORRECTION FACTOR

bull 15 TO 20 CONSERVATISM MEAN)

bull UNEXPECTED SURVIVALS HAVE GREATEST IMPACT

bull DAMAGED CASES MORE INFORMATIVE

bull CALIBRATION ISSUES PHASE II

bull

EFFECT OF IMPROVED CAPACITY ANALYSIS RECIPE

bull APPLICABILITY TO MULTIPLE FAILURE MODES

bull NO FOUNDATION DAMAGE CASE INCLUDED

bull IMPROVED REPRESENTATION OF FULL ANDREW DISTRIBUTION

-bull NUMBER OF PLATFORMS CONSIDERED

I 1 I

I t ) bull ~

PHASE II WORK SCOPE (CALIBRATION)

bull

bull ESTABLISH MULTIPLE BIAS FACTORS

Use Nine Base Case Platforms

middot - Capacity Analysis for Revised recipe

Foundation Blas Factor from Foundation JIP

bull DEVELOP WEIGHTING APPROACH

bull UPDATE MULTIPLE BIAS FACTORS

bullEffect of Platforms Not In the Nine Base Cases

bull EXAMPLE APPLICATION OF BIAS FACTORS I

bullbull

PHASE I PHASE II

CAPACITYbull ANALYSISRu bull

---------------------1----middot

Jc1 - iJbullraquoJ dbull

FttlS frt[f~l ratdhJJ

-

LIKELIHOOD FUNCTION

DEVELOPMENT

lk Jb Jfall1N) bull Jb)

111 Ill Isurvtvoq bull 1 - bl

-middot -------------------shy ~---------------------middot-middot

r111ir8 Cb1 111 Cbl BAYESIAN

UPDATING

~cove

R1R2 bull middot---------------------~----

Jblbull P(g cOJ

1 bull JbRI (MBSI I bull 1

~middot

llcJllJfalluro) bull P(gcOJ Ill 111 b21fallun) bull P ((g1 CO) U 1112 c OD

middot--------------------shy ----------------------shy bull

r111111r8 Cbl 111 Ill i-1nro1

deg ~1b2)laquottfrr (b1b21-lnfo) re Ill bull 1-11middot b2) bull lodegamp

middot

i1 covamp i2 cov~i

-

shy

--

Likelihood Function vs b All Success Case STISI K STl30Q STl34W WD90A MC311 MC397

I 0000

0 9000

OIOOO

0 7000

~ 0amp000

ta osoOO

~ 0 4000

0 3000

0 2000

0 1000

bull 0 0000

bull

I

middot

middot bull I

Ibull

-sibull ~ -middot-middot-middot-middot -middot-

----7----+-----~ I

Ibull bull bull I

rbull

--- Jk ISIK[blsuccess)

- bull - bull - lk 130Q[blsuccess) bull

lk 134W[blsuccess)

Jk-WD90A(blsuccc~I

- - - lk MC397[blsucccss)

--ltgt-- lk middotMC31 l[blsucccss)

--- Jk [bl 6 out of6succcsHascs)

05 I 5 2 250

b

Likelihood Functionsmiddot Success Cases - Phbullsc L

shy

bullbull bull bull bull bull bull -------------------~

bull bull bullbull bullbull bullbull bullbull bullbull bullbull bull

bull bullbull

bullbullbullbull

bullbull

bullbullbullbullbullbullbull

bullbullbullbullbullbull

bullbullbullbullbullbullbull

z 0

t z ) LL

c 0 0c i ~ i

J

~ sect ~--------------~ e l5

I

I

SOFTWARE

bull PF PROGRAM

bull CALREL

bull PF +CALREL

bull RELACS

bull

gt i

l

SENSITIVITY STUDIES

POSTERIOR DISTRIBUTION OF B

bull EFFECT OF CHANGE IN DISTRIBUTION OF PRIOR OF B

bull EFFECT OF REMOVAL OF ONE PLATFORM FROM EACH CATEGORY

bull EFFECT OF REMOVAL OF THREE DAMAGE CASES

bull EFFECT OF ADDITIONAL 13 SIMILAR PLATFORMS

-I

Prior and Posterior Distributions of B Variation In COV of PriorofB fromOl to04 -- -

4500

-4000

-

- -

3500 --middot - bull bull - bull bull Prior-r ICOV =02

3000

- - - - Prior-r ICOV =03 -

bull -~ e 2soo -

bull bull bull bull bull bull bull Pnor-r ICOV =o 4t frJ ~ 2000 Posterior rlCOV =0 2

bull bull Postenor rlCOV =0 3 bull bull 1500 bull middot bull - shy -middot- Posterior rlCOV =04

middot~ I 000 - bull

-middotmiddot ~ ~middotmiddot M -~

- middot 0 500

wbullbullbull _ I middot~ middotmiddotmiddotmiddotmiddot middotmiddotmiddotmiddotmiddot - -~ ~ __ middot_ _ bull -shy0 000

0 000 0200 0 400 0 600 0 800 IOOO 1200 1400 1600 1800 2000 -

- - b

-

=shy

Posterior Distribution of Bias Factor (B) bull Effect of Variation In COV of Prior of B

bull

bullPrior and Posterior Distributions of B Evaluating Effect of Removal of One Platform from Each Category from the Base Case of 13 Platforms

400

~ ~ 350

j IA i

3 00 bull I bull I ---Prior-r I bull l

250 1 I 1

I bull bull ---Posterior r lall 13 casesshy- ~E 2 00 bull bull bull bull bull bull bull Posterior r without STISIK

- ~ 1 bull - bull I - bull bull - bull bull Posterior r Iwithout T2I I 50 I

J-tt~i - bull - bull - Posterior r Iwithout T2S

tOO [-

I I ~ ~ I 0 50

0 00

___ r J i

l~T

~--

0 00 0 20 0 40 0 60 080 I 00 120 140 160 110 2 00

b

zshy

Posterior Distribution of Bias Factor (8) bull Effect of Removal of middotPlatform from Any Category

bullPrior and Posterior Distributions or B Evaluating Effect or Removal or 3 Damage Cases from the Base Case or 13 Platrorms

4

35

3

25

e L 2 ~ bulle shy

I 15

bull I

05

0

0 02 04 06 I 2 I 4 16 IS 2

I I

I

I

I Ibull

J -tmiddot L--

bullI

~

I

fl

II ~

I I v I

middot bull bull

bull L

~ ll- I-~lt

OS I

b

I

bull

1middot bullbullbullbull

~ __

bull

- - - - Prior-r

- - - - Postenor flall failure cases

Postenor flall success cases

--- Posterior flall 13 cases

---Posterior fI success failure cases only

IPosterior or B Mean =t11 COV =0121

-

_I

Posterir Distribution or Bias Factor (B) bull Effect or Removal or 3-Damage Cases

-

bullPrior and Posterior Distributions or B Evaluating Effect or Additional 13 Similar Observations as

Used in Andrew JIP The POsterior or B from Andrew-JIP Considered as New Prior or B

- 500

450

400

350

bull bull

bull bull bull bull J bull

bull bull f ~ I

I ~ Ibull bull

bullbull bull

l bull I -bullbull bull bull

I

I bull bull

bull I I~~bull bull bull

bull ~

I bull bull

~ ~_ ~~~

- - New Prior r bull

Posterior 114 new failure cases ~ 300- bull bull bull bull bull bull bull Posterior f 13 new damage casesl 150 ~

- bull bull - bull bull Posterior 116 new success cases i 2 00

Postenor 1113 new success failure -150 damage cases

I 00

0 50 -

0 00 0 00 0 20 0 40 0 60 0 80 I 00 120 I 40 160 110 2 00

Posterior ol B Mean= 120 COV= 0071 b

shy

shy

_rshy

Posterior Distribution of Bias Factor (B) bull Effect of Additional 13-Similar Observations

1

1 I

WEIGHTING APPROACH bull

bull ASSESSMENT OF PLATFORMS DATA

Select Platforms Most Mappable to the Analyzed Platforms

Unlikely SurvivalDamage Cases Identification

Based on Cut-Off Criteria Agemiddot Hs etc

bull INFORMATION REQUIRED FOR COMPARISON

bull CATEGORIZATION OF ADDITIONALmiddot PLATFORMS

bull UPDATING OF CALIBRATION RESULTS

--

OVERVIEW OF PLATFORMS SURVIVEDDAMAGED -~

Operator -

- Platforms

Survived CS)

Platforms

Damaged CD)

Platforms Used In

Phase IJIP -

JI~ ~Bill~IRBDll -

Amoco -

Chevron -

Exxon

Mob II

Phllllps

Shell

Unocal -

30

143

72 22

7

39 25shy -

1

12

bull

1 bull

1

1(S)+1 (D)

3 (S) +3(D)

2 (S)

bull shy

bull

bull

bull

Total Partlclnants Platforms 338 - 15 10C6S+4D) -

Other Operators

310

13

3 (D)

Total Platforms 648 28 13C6S+7D) -

Summary of Andrew JIP bull Phase I Platforms

Plairshy w v- N-ber Lea-le 1 shy car Hind_ o-w N-bull Deplh fl fl IJpe la bull n Yeadon Jadtet lbelledtf liefalit -- _ -OD

IL H - -shy _ICllshy IL IL - - rQClto ~ _

K STUIK 137 1963 8 365 608S4 Cellardeck

-middot STl-shy 170 1964 4 -~ K 39 62269 Cellardeck - --ST134W 137 1911 4 43 608S4 Smvlnlshy- _ - -

WD90A 184 1964 8 K 267Sbull34 83 SOS31 SUboftllar deck SmvMI

MC311 343 1978 8 x S7 62 - MC397 468 1991 4 x shy S33bull6SO 65911 -shySmeSmvlvol -- gt -

r-- T23ST52 63 1969 4 ton- shyKshy 52594 S03 -T25SSl39 62 1969 4 Kshy SOS94 S063 - middot ton-shy__ -

bullbull a ton--shyST161A 118 1964 8 K - 347S 51495 Cellarclecti

r~

ST1771 142 1965 8 K 39 60167 CeDordeck failme

STISIH 137 1964 8 K 395 608S4 CeDardeck Failme

K ST130A 140 1958 a SS 6095 I lblv Pai1uno - T21ST72 61 1969 4 Kshy S4S94 4974 I lhlv Pai1uno -

- -

JE

INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON

bull WATER DEPTH

bull bull YEAR OF INSTALLATIONDESIGN

bull PLATFORM ORIENTATION

bull PLATFORM LATITUDE AND LONGITUDE

bull KEY SKETCHES OF PLATFORM STRUCTURAL FRAMING

Number of Legs Plies Leg Batter Number of Conductors bull Deck Elevations (Including sub-cellar deck If any) Jacket Framing Elevations Brace type In Vertical Frames Sizes of Primary Members (leg pile vertical frames braces primary horizontals) Joint Details (can gap etc) LegPiie Annulus Grouting Status

bull middotPOST ANDREW INSPECTION REPORT Significant Wave Height Hs (ft)

bull INFORMATION ON DAMAGES PRIOR TO ANDREW

bull RESULTS FROM IN-HOUSE STUDIES IF AVAILABLE

bull ~RIGINAL DESIGN CRITERIA INFORMATION IF AVAILABLE

bull SOIL SHEAR STRENGTH PROFILE IF AVAILABLE

Platforms Identified for Evaluating Therr Likelihood of Falling Under Unexpected Survival Category

OPlRATOR

middotADA

BLOClt

llmucnJllE

NAME

YEAll

INSTAUJD

WA1ER

DBPTll

DISTANCE

JlllOM SHOllE

(MILU)

Bl

(M)

AMOCO 111113

STl11

WDI075

A

bull D

St

17

120

1n

44

20

10 I

CllllVllQfl STlllO c 61 110 21 11 STIUO D Q 1110 21 11

STl134 p Q 137 29 11

ST0151 G 61 137 11

ST0151 I 3 121 11

ST0151 J Q 140 11

STl151 It 137 11

ST0151 PJl001 Q UT 11

ST0151 PROD-2 Q UT 11

ST017 A 3 140 35 11

WDIUT c 65 214 34 10

llXXON STI055 II 51 6T 14 I

STl15 A t3 34

WDI073

WD0073

Amiddotshy Q

1111 1111

22

22

I

I

WDI073 c 1n 22 I

- WDOCl93-shy II

65 1110 2T A 3 194 31 MOBU 111121

SS IOl7

A B

501

50

50

31

T

I

T 6

MtlllPllY SS114 B T 50 T T

PlllllIPS SS0149 A 5 55 33 T

SAMllDAN STln A 107 T T

SBllLL WDl103 A 65 223 2T WDl103 B 65 221 2T WD0104 c 65 221 2T WD0104 D 6T 20 2T

UNOCAL SS020I

SS 020I SSl20I

SS 0209

SS0209

II p

PURll

A

B

3

n 61

103

97

97

95

100

34

34

34

34 34

bullbull bullbullbull

- SS0209 D u 95 34 I

- SS 1209 SSl253

SS0253

G A

c

6T

2

100

165

1T5

34

54

54

bullbull I

I I

I I

SOUTH MARCH

~LAND

N

SOUTH PaTOI ENE ISWltD AREA

MMS Defined Limits ror Offshore Structures Requiring Inspections Following Andrew

t---Jbulln3 T25bull I bullTtl

bull1$1Htl1Abull bull1SIK1779bull

~

middot-DaTA AREA~T

CRANO rsLE

AREA bull1300

bullll4W

=shy

bullPath of Hurricane Andrew with Platforms Used In Calibration

Likelihood Function vs b All Success Case STISI K STIJOQ ST134W WD90A MCJll MCJ97

I 0000

0 9000

0 8000

07000

-Bo6000

tiiii 0 5000 -

04000

0 3000

0 2000

01000

0 0000

0 05 25

ri1rt middot-middot-middot-middot-middot-

I 5 2

b

---11 middot151K(blsu~ccsI

- bull - bull - fk 130Q(bfU~lCSS)

lie bull134W(blsuccc]

lk-WD90A[blsultCS]

- - - lk middotMC397(bfUtlCS]

--o-- lk-MCJll(bbuctcsJ

--- ll (bl 6 out of 6 Mlttcs case]

_

( Likelihood Functions middotSuccess Cases - p - 1 c I

CATEGORIZATION OF ADDITIONAL PLATFORMS

I I I

bull CATEGORIES

Unexpected Survivals

Expected Survivals

bull APPROACH

Identify Corresponding Platform(s) Evaluated In the JIP

Determine Max Wave Height at Platform Location

Evaluate Whether Wave Hit the Deck

middot - Evaluate Likelihood of R and S to Differ from the Base Cases

Review Joint Information

Assess Special Features

bull IDENTIFY NUMBER OF PLATFORMS IN EACH CATEGORY

I

UPDATING OF CALIBRATION RESULTS

bull INCLUDE EFFECT OF ADDITIONAL PLATFORMS ON bullBbull

Consider Increased 1Uncertalntles In R and S

Develop bullNeW Ukellhood Functions for Different Categories

middotLet bullxbull Number of Ukely Unexpected Survival Cases and y Number of Ukely Expected Survival Cases

Determine Combined Likelihood Function lk (b Iadditional platforms) middot

bull REVISED LIKELIHOOD FUNCTION

lk(b I all cases) = lk(b I 9 platforms) middot x lk(b Ix platforms) x lk(b I y platforms)

bull ESTABLISH REVISED BIAS FACTOR B

f8 (b) u f8 (b) lk (b I all cases)

II

t t ~

shy

APPLICATIONS OF BIAS FACTORS

ii APPLICABLE TO PARTICULAR LOAD AND RESISTANCE RECIPE

bull USE BY THE API COMMITTEES AND MMS

Criteria Revisions

Recommended Design Guidelines Revisions bull

bullbull REASSESSMENT STUDIES AND DECISION MAKING

middot

bull RISK ASSESSMENT STUDIES BY THE OPERATORS

Individual Platforms

Fleet of Platforms

middot IMR Program Implementation

bullbull NOT TO BE USED IN DESIGN

I I fl J

EXAMPLE USE OF BIAS FACTORS

bull PHASE I EXAMPLE APPLICATION

bull EVALUATION OF ANNUAL PROBABILITY OF FAILURE

Establish Annual Maximum Seastate Data

Determine Coefficients to Define Base Shear

Determine Ultimate Capacity of Platform

Determine Probablllty of Failure Versus bullbbull

-bull DETERMINE UPDATED PROBABILITY OF FAILURE

Pf - P [S gt R]

Pf I P [~ gt Rb I b] f 8 (b) db

Pr - J P [S gt Rb I b] f bull 8 (b) db

bull TOTAL PROBABILITIES OF FAILURE

Total Pf - 1 bull Il [1 middotPf] - l Pf bull

Total Pf - 1 bull Il [1middot Pf] - l Pfmiddot

Total Pr = 1 - n [1middot PfbullJ = l Pfshy

I I I

EXAMPLE APPLICATION bull PHASE I

PLATFORM ST 151 K

bull GULF OF MEXICO

bull WATER DEPTH = 137 FT

bull ANNUAL MAXIMUM HS MEDIAN 175 ft COV 031

bull ASSUMED STORM DURATION 3 hoursmiddot

bull ASSUMED NUMBER OF WAVES 800

bull MEDIAN HS FOR STORM HOURS 1 2 AND 3

bull MEDIAN CURRENT VALUE bull

bull DETERMINE ULTIMATE CAPACITY FOR THREE DIRECTIONS

bull DETERMINE ANNUAL FAILURE PROBABILITIES

Broadsidemiddot Diagonal End-on

0037 0025 0007

0071 0044 0018

0022 0016 0004

DETERMINE TOTAL ANNUAL PROBABILITY OF FAILURE

Pf - 007

Pf - 013

Pr - 004

bull

EXAMPLE USE OF BIAS FACTORS - PHASE II middot

bull DETERMINE PREDICTIVE PROBABILITY OF FAILURE

Case - 1

Pf _ S S Pf (b b2) fa1a2 (b b2) db db2

g bull -1 (b1 R I S) - 1

92 - (b2 R2 I S) - 1

Case -2

Pf s ~ [bJ f11 (b) db

Pf [bJ p [g lt 0)

g - bull 1 (b1 R1 I S) - 1

1

bull1) l

I

BENEFITS OF ANDREW JIP bull PHASE II

lt

bull IMPROVED LOAD AND RESISTANCE RECIPE

bull IMPROVED CALIBRATION PROCEDURE

bull MULTIPLE BIAS FACTORS

bull PROCEDURE TO INCLUDE EFFECT OF OTHER PLATFORMS

bull PROCEDURE TO USE MULTIPLE BIAS FACTORS

  • Structure Bookmarks
    • ANDREW PHASE II JOINT1 INDUSTRY PROJECT
    • PROGRESS MEETING NUMBER 1
    • ANDREW PHASE II JOINT INDUSTRY PROJECT PROGRESS MEETING NUMBER 1 NOVEMBER 16 AGENDA
    • ANDREW PHASE II UST OF PARTICIPANTS
    • ANDREW PHASE I OBJECTIVES AND RESULTS
    • ANDREW PHASE II PROJECT OBJECTIVES
    • SUBSEQUENT PROJECTS
    • ANDREW PHASE II PROJECT TEAM
    • ANDREW PHASE II UPDATE MODELING middotAND ANALYSIS RECIPE
    • -ANDREW PHASE II CONDUCTOR RESISTANCE
    • ANDREW PHASE Il WA VE FORCE PROFILE
    • ANDREW PHASE II WA VE FORCE PROFILEmiddot
    • ANDREW PHASE II BRACE MODELING
    • ANDREW PHASE II FOUNDATIONS
    • ANDREW PHASE II middot middot JOINT MODELING
    • ANDREW PHASE Il JOINT MODELING OPTIONS DISCUSSION
    • ANDREW PHASE II JOINT MODEJJNG OPTIONS ILLUSTRATIONS
    • ANDREW PHASE II JOINT MODELING PHASE II METHOD
    • ANDREW PHASE II ANDREW STORM CONDmONS UPDATED
    • Platform Selection bullbullbull1111111111111
    • BasisCriteria for Selection bullbullbull1111111111111
    • CALIBRATION
    • WHERE TO INTRODUCE BIAS FACTOR (B
    • PHASE I -KEY CONCLUSIONS
    • bull CALIBRATION ISSUES PHASE II
    • PHASE II WORK SCOPE (CALIBRATION)
    • SOFTWARE
    • SENSITIVITY STUDIES POSTERIOR DISTRIBUTION OF B
    • WEIGHTING APPROACH bull
    • INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON
    • CATEGORIZATION OF ADDITIONAL PLATFORMS
    • UPDATING OF CALIBRATION RESULTS
    • APPLICATIONS OF BIAS FACTORS
    • EXAMPLE USE OF BIAS FACTORS
    • EXAMPLE APPLICATION bull PHASE I PLATFORM ST 151 K
    • EXAMPLE USE OF BIAS FACTORS -PHASE II middot
    • BENEFITS OF ANDREW JIP bull PHASE II
Page 30:  · . ' ANDREW PHASE II PROJECT OBJECTIVES . ' • ·UPDATE MODELING AND ANALYSIS RECIPE· JOINTS ' ' ·BRACES WAVE PROFILE FOUNDATIONS . ' ' CONDUCTORS· • ANALVZE PLATFORMS USING

-

Project andrew Model strutframe Version 1 Mon Nov 14 145754 1994

bull

Frame Test strut model

middot-middot _____________--_

_middotmiddotr_~ --ll--~-middot__

Buckling Beam bull

Strut (K=O 65)

Lateral Displacement

ANDREW PHASE II

FOUNDATIONS

bull ~LICIT NONLINEAR API FOUNDATION USED INiPHASE I

bull FOUNDATIONS

JIP SELECTED MINI-VANE DATA FOR

STRENGTH PROFILE

bull COMMENTS ON BENCHMARK COMPARISONS

- SIGNIFICANT VARIABILITY DUE TO FOUNDATION MODELING METHODS

II Ii

z

ctV 4x SPlO caissonfy-42s soil-api(lf-097step-114)

Inelastic Events Legend Elastic Strut Buckling

Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield

--------shyBeam Clmn Fully Plastic Fracture

~--~~~~~~~~~--~----~__i~~---~~~~__--~

Pro)ect apipile Model South Pelto 10 Version Wed May 18072235 1994 - -

1bullt

+

t

f i bullbullbull

bullbull bullbull

I I

5

~x SPlO Caissonfy=42s soil=api(lf=lOOstep=137)

Inelastic Events Legend

Elastic Strut Bucklilg

Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield

earn Clmn Full~ Plas1c Fracture

ANDREW PHASE II middot middot

JOINT MODELING

middotbull API RIGID-PLASTIC USED IN PHASE I

a ELASTIC-BRITTLE USED IN PHASE IB

amiddot MSL WILL PROVIDE NONLINEAR FORCE-DEFORMATION

AND MOMENT-DEFORMATION RELATIONS FOR PHASE II

a MODELING OPTIONS FOR PHASE II -

l

- UNCO~LEDmiddotONE DIMENSIONAL MEMBER END JOINT MODELING

J Ii

ANDREW PHASE Il

JOINT MODELING OPTIONS DISCUSSION

bull JD COUPLED INTERACTION SURFACE

- IDEAL

- DATA NOT AVAILABLE TO CALIBRATE

bull 2D COUPLED INTERACTION SURFACE

- SIGNIFICANT MODELING IMPROVEMENT

- DATAAVAILABLE

- IMPLEMENTATION PRACTICAL BUT NEEDS DEVELOPMENT AND CALIBRATION

bull 2D UNCOUPLED INTERACTION SURFACE

- SIGNIFICANT MODELING IMPROVEMENT

- DATA AVAILABLE

- IMPLEMENTATION PRACTICAL AND AVAILABLE

ANDREW PHASE II

JOINT MODEJJNG OPTIONS ILLUSTRATIONS

P

e

ANDREW PHASE II

JOINT MODELING PHASE II METHOD

e

I I I

ANDREW PHASE II

ANDREW STORM CONDmONS UPDATED

bull -PHASE I USED OCEANWEATHER HINDCAST-(Nov 1992)

bull middotmiddotPHASE II WILL USE THE REVISED HINDCAST (Nov 1994)

bull lt c lt I lt

2-2

l---4-~~~+-~~-+-~~~1--~~-+-~~--l--~T--+-~~--i24

f-tJVtMfiz~ 1Cfl (r STvb(

Plollol bullbull 5middotNOYmiddot9214bull1J Irbullbull Illbull [ANDREVl~NDREVCUSTbullMAXll 5-NOY-1992 10bull54 - 9208

HEIGHT lml

I~ MAXI

I L 1~middotamp SZ~- ~A I 13c

I J) ( r I X~0iWJ~0gtJ gt L gt gt I 1c I 120

-- N

~ deg

I I I I I I e C I ~=-41 122 -94 -92 -90 -88 -86 -84 -82 -80

-

Figure 7 contours of maximum hindcast significant wave height

_

~

~tl ~

NOltNttIJ 17lt ltNS S iJJ)y

Maximum Significant Wave Height - Hurricane Andrew 1992

30 ~

26 ~ I ~-~~~- == -shy -~ 35-_

24deg1--shy --------~- middot-middot-------middotmiddotmiddot----middot------middot-middot~middot- - - ~-----1---middot

22middot ~~~~~~~-~ -94 -92 -90 -88 -84deg-86 -82 -somiddot

c

-

Platform Selection bullbullbull1111111111111

0 BasisCriteria for Selection middot

0 Review of Phase I Platforms middot 0 Alternative Platforms

0 PMB Recommendations

0 Participants Comments

gt

L

l I

BasisCriteria for Selection bullbullbull1111111111111

0 Benefit for Calibration - Classification (UnexpectedExpected)

- Failure Mode (StructuresFoundations)

- Damaged StructuresKnown response

- Better Representation of Full ~opulation

0 Availability of Data - Existing Models

- Inspection Data

- Damage Data

PREDICTED BEHAVIOR

-

OBSERVED

BEHAVIOR

-SURVIVED DAMAGED FAILED

SURVIVED

ST134W WD90A

-MC311 -

MC397 -

--

-

-

ST151K ST130Q

-

-

-

DAMAGED - -

-

-

--

-

-

ST52 SS139 ST177B -

ST161A

-_

-

FAILED

-

-

-

ST151H ST130A ST72

Operator - Platfonns - Platfonns Survived Damagedamp

- Falled -

bull

-

-

Jle eBDIGIPllDbl -Amocoshy 30 1 Chevron -

143 12 Exxon 72 Mobll 22 1 Phllllps 7 bull

-Shell 39 Unocal 25 1

Total Partlclnts Platfonns 338 15

Other Operators bull- 310 13

-

Total Platfonns - -

648 28

fbml

Platforms Considered 6 7

tie of Total Platfonns 1 25

OVERVIEW OF PLATFORMS SURVIVEDDAMAGEDFAILED

-

shy

v

Platforms Evaluated in Phase I

-

-Platfonn Water Year Number Analysis Avallabillty of Availability Platform

- Name Depth rl of Model DetaDed rl ClassUlcatlon

Installation legs Inspection Soll Report Based on Information Phase I Calibration - -ft

- -- Suaill flalfnrm ~BSi -STISIK 137 1963 8 CAP Yes Yes Un eel Sarvlval

-STIJOO 170 1964 4 CAPmiddotshy Yes ST134data u ST134W 137 1981 4 CAP Yes Yesmiddot E

-

edSarvlval -

middotsurvival

WD90A 184 - 1964 8 CAP Yes E -

- Sarvlvill

MC3ll 343 1978 8 KARMA

MC397 468 1991 4 KARMA -

bull ~reglmmiddot- -

Yes Sure Survival

bull Yes - Sare Sanlval-

-T23STS2 63 1969 4 CAP

T2SSS139 62 1969 4 CAP -

Yes

Yes

Yes Ex

Yes Une

bull SurvivalLikely to DamalI

middot bull SanlvalLlkely to DamalI

ST161A 118 1964 8 USFOS

Eail11a pound1al(bullllm Clss ST177B 142 --1965 8 CAP

STISIH 137 1964 - 8 CAP C

STIJOA 140 19S8 8 CAP - -

Yes

Yes

ollamed

Yes

- Yes E

-

at about 1 mUe

Yes

ST134data

bull SafvlvalLlkelv to Dam- bull- -

E middotrahre

Emeried Failure -

Likely FaUare

T21ST72 61 1969 4 CAP Collamed No Likely FaUure

Platforms Identified for Evaluating Their Likelihood of Falling Under UnexpeCted Survival Category

CllEllATOa

AllB

llLOCll

Sl1lvcnJllE

NAME

YEAR

INSTAUJCD

WATER

IJBPJB

middotmiddotDISTANCE

ftOMSllOllE

(MIUIS)

Ill

(M)

AMOCO BllW

STllI

WDI075

A

bull D

4

n uo 17J

44

32

bull 20

10

bull CllEVllOH STtllO c I 1bull 21 11

STl130 D 0 161 21 11 STllJ4 I 0 137 D - 11

STIUI G I 137 32 11

STIUI I 0 121 32 11

STIUI

J 0 140 32 11

STIUI It 4 137 32 11

STllSI PROl)I 0 137 32 11

STllSI PROD-2 0 137 32 11

STll7 A 0

140 35 11

WDl117 c 65 214 34 10

DXON STshy E 51 14 bull STl165 A t3 bull 34 WD0073

WD0073

WD0073

A _ c

4

62

4

IA IA

17J

22

22

22

bullbullbull WD003-shy E 65 161 Z7

A 0 IN 31 MOBD Bl12I

ssn A bull

50

50

31

7

bull 7

MIJllPllY SS114 B 7 50 7 7

PlllUIPS SSl14t A 55 33 7

SAMEDAN ST17J A 4 117 7 bull 7

SJIELL WD0103 A 65 223 27 WDllOJ bull 65 221 27 WDll04 c 65 221 27 WDOI04 D 67 2 27

UNOCAL

SSl20I

SSl20I

bull SSl20I

SS0209

SS0209

SS0209

SS0209

SS 0253

SS 0253

E

I

PtJllE

A

bull D

G A

c

0

4

4

7J

61

0

0

69

103

97

97

9S

100

95

100

165

175

34

34

34

34

34

34

34

54

54

bullbullbullbullbullbullbullbullbull

Platforms Recommended for Phase II

llalr y_ w- Aot Av- _ I 1 Doplla Model nm bull bull _ bull I I H11 s18-1 Irht lbwl~ bullbullbull(-illtdlel OM lllMtD

II ---

-STISIIC 137 1963 I CAP Yeo Yeo I Yeo

STtshy -170 1964 4 CAP Yeo ST134dlU I Yeo

STl34W 137 Yeo Yeo -1981 4 CAP D Yeo

WD90A 114 1964 I CAP Yeo - D

MC311 30 1971 I KARMA Yeo -sre-

MC397 middot 461 1991 4 ICARMA Yeo Sireshy - - r- -

__ T13ST5l 63 1969 4 CAP Yeo Yeobull middot- ~n- D Yeo _ Tl5SSl39 62 1969 4 CAP Yeo Yeo I Yeo

_ STIT711 142 1965 I CAP Yeo I Yeo middot-1shySTl61A Ill 1964 I USFOS Yeo Yeo - D bulloshy

- r-

STISIH 137 1964 I CAP y D y -_ STl31A 140 1951 I CAP Yeo STl34dlU I Yeo

T21Sll 61 1969 4 CAP No rar-middot I Yeo

~-rmiddot-middot SS209A 95 1967 16 CAP Yeo D -SSU4R

11 UI A 41 -in D SS 2741AIM CAP bull

_ _ -shy - -shy- -shy

CALIBRATION

bull PHASEIAPPROACH

bull CALIBRATION ISSUES

bull PHASE II WORK SCOPE

bull SOFlWARE AVAILABILITY

bull EFFECT OF ADDITIONAL PLATFORMS

bull WEIGHTING APPROACH

bull EXAMPLE APPLICATION middot

(l=Ofce Mode] Wind

~

fcurrent l -middot

~SW ~leldl4

ultimate resistance force I first

design load Illgt

i _

deflection

-

bull

4

Historical Hurrlcane Dariiage and Survival of Platforms

Components Required for Calibration (Petrauskas 1992)

PRIOR Distribution

PLATFORM EXPERIENCE

BAYESIAN Survivals and Failures

UPDATING

Waves Winds and Currents bull

POSTERIOR Distribution

- A_middotJA ~ Survival Data Failure Data

Result Result

Load Resistance

middotfl

-

Bayesian Updaing middotGeneral Description (Petrauskas 1992)

L Imiddot

WHERE TO INTRODUCE BIAS FACTOR (B

On Resistance (Rmue - B (RPREDICreD I

middotOn Loads (Smue middot - B (SPRE01cTEo

On Safety Margin (RSmue - B (RSPREDICTED

CALIBRATION METHODOLOGY

gt

r

FIELD OBSERVATIONS

Observed Behavior

survivals Damages

Failures

r

bull

CAPACITY ANALYSIS

Predicted Load and Resistance Predicted Failure Modes

Survival Damage Failure cases

RELIABILITY ANALYSIS

- Probability of Failure ~

Ukellhood Failure

BAYESIAN UPDATING

Prior Distnbution of B Combined Likelihood Function

Posterior Dlstnbution of B

BIAS FACTOR

Mean Value cov bull

Represents Modeling Uncertainties

Prior Distribution of B Likelihood Functionmiddot Observing Success case STISIK and Posterior Distribution of B

2 00

uo

160

140

120

e 100 L $ uoIshy

0 60 ~ ~ 040

020

0 00

0 00 0 50

_ I 00

b

I 50 2 00

-shy lk [bl STISIK)

--shy Pnor-f(b)

--shy Pos1enor r(bl ISIK)

Prior or B Mean bull 10 COV bull 030

Posterlor of Bbull Meanbull 136COV 0 16

2 50

Likelihood Function and Posterior Distribution Example Platform STISIK

- -

~

- Prior and Posterior Distributions or B Failure cases (ST177B STlSlH ST130A T21) Damage cases (T23 T25 ST161A) Success cases (STlSlK ST130Q ST134W WD90A MC311 MC397)

400

350

300

2 so 0-

I

bull bull

bull bull gt

- bullbull ~ bull bull I bull

_ bull bull bullbull I bull bull bullII V -

bull bull

I bull ~ bull

bullbull bullI bull bullI bull bull bull

middot -~

bull__ ~ middotI middotmiddot

L 2 00

e ~ I SO

-I 00

- 0 so

000 gt

0 00 020 0 40 060 0 80 I 00 120 140 160 110 200

b -

-

shy

Prior-r

- - - - Posterior fLlll failure cases

bull bull bull bull bull bull bull Posterior flall damage cases

- bull bull - bull middot Postenor flall success cases

Posterior flall success fulure damage cases

-

Posterior of B Mean= 119 COV 0101

Posterior Distribution or Blas Factor (8) bull Different Categories and All Cases Combined

shy

l

PHASE I - KEY CONCLUSIONS

bull GLOBAL CORRECTION FACTOR

bull 15 TO 20 CONSERVATISM MEAN)

bull UNEXPECTED SURVIVALS HAVE GREATEST IMPACT

bull DAMAGED CASES MORE INFORMATIVE

bull CALIBRATION ISSUES PHASE II

bull

EFFECT OF IMPROVED CAPACITY ANALYSIS RECIPE

bull APPLICABILITY TO MULTIPLE FAILURE MODES

bull NO FOUNDATION DAMAGE CASE INCLUDED

bull IMPROVED REPRESENTATION OF FULL ANDREW DISTRIBUTION

-bull NUMBER OF PLATFORMS CONSIDERED

I 1 I

I t ) bull ~

PHASE II WORK SCOPE (CALIBRATION)

bull

bull ESTABLISH MULTIPLE BIAS FACTORS

Use Nine Base Case Platforms

middot - Capacity Analysis for Revised recipe

Foundation Blas Factor from Foundation JIP

bull DEVELOP WEIGHTING APPROACH

bull UPDATE MULTIPLE BIAS FACTORS

bullEffect of Platforms Not In the Nine Base Cases

bull EXAMPLE APPLICATION OF BIAS FACTORS I

bullbull

PHASE I PHASE II

CAPACITYbull ANALYSISRu bull

---------------------1----middot

Jc1 - iJbullraquoJ dbull

FttlS frt[f~l ratdhJJ

-

LIKELIHOOD FUNCTION

DEVELOPMENT

lk Jb Jfall1N) bull Jb)

111 Ill Isurvtvoq bull 1 - bl

-middot -------------------shy ~---------------------middot-middot

r111ir8 Cb1 111 Cbl BAYESIAN

UPDATING

~cove

R1R2 bull middot---------------------~----

Jblbull P(g cOJ

1 bull JbRI (MBSI I bull 1

~middot

llcJllJfalluro) bull P(gcOJ Ill 111 b21fallun) bull P ((g1 CO) U 1112 c OD

middot--------------------shy ----------------------shy bull

r111111r8 Cbl 111 Ill i-1nro1

deg ~1b2)laquottfrr (b1b21-lnfo) re Ill bull 1-11middot b2) bull lodegamp

middot

i1 covamp i2 cov~i

-

shy

--

Likelihood Function vs b All Success Case STISI K STl30Q STl34W WD90A MC311 MC397

I 0000

0 9000

OIOOO

0 7000

~ 0amp000

ta osoOO

~ 0 4000

0 3000

0 2000

0 1000

bull 0 0000

bull

I

middot

middot bull I

Ibull

-sibull ~ -middot-middot-middot-middot -middot-

----7----+-----~ I

Ibull bull bull I

rbull

--- Jk ISIK[blsuccess)

- bull - bull - lk 130Q[blsuccess) bull

lk 134W[blsuccess)

Jk-WD90A(blsuccc~I

- - - lk MC397[blsucccss)

--ltgt-- lk middotMC31 l[blsucccss)

--- Jk [bl 6 out of6succcsHascs)

05 I 5 2 250

b

Likelihood Functionsmiddot Success Cases - Phbullsc L

shy

bullbull bull bull bull bull bull -------------------~

bull bull bullbull bullbull bullbull bullbull bullbull bullbull bull

bull bullbull

bullbullbullbull

bullbull

bullbullbullbullbullbullbull

bullbullbullbullbullbull

bullbullbullbullbullbullbull

z 0

t z ) LL

c 0 0c i ~ i

J

~ sect ~--------------~ e l5

I

I

SOFTWARE

bull PF PROGRAM

bull CALREL

bull PF +CALREL

bull RELACS

bull

gt i

l

SENSITIVITY STUDIES

POSTERIOR DISTRIBUTION OF B

bull EFFECT OF CHANGE IN DISTRIBUTION OF PRIOR OF B

bull EFFECT OF REMOVAL OF ONE PLATFORM FROM EACH CATEGORY

bull EFFECT OF REMOVAL OF THREE DAMAGE CASES

bull EFFECT OF ADDITIONAL 13 SIMILAR PLATFORMS

-I

Prior and Posterior Distributions of B Variation In COV of PriorofB fromOl to04 -- -

4500

-4000

-

- -

3500 --middot - bull bull - bull bull Prior-r ICOV =02

3000

- - - - Prior-r ICOV =03 -

bull -~ e 2soo -

bull bull bull bull bull bull bull Pnor-r ICOV =o 4t frJ ~ 2000 Posterior rlCOV =0 2

bull bull Postenor rlCOV =0 3 bull bull 1500 bull middot bull - shy -middot- Posterior rlCOV =04

middot~ I 000 - bull

-middotmiddot ~ ~middotmiddot M -~

- middot 0 500

wbullbullbull _ I middot~ middotmiddotmiddotmiddotmiddot middotmiddotmiddotmiddotmiddot - -~ ~ __ middot_ _ bull -shy0 000

0 000 0200 0 400 0 600 0 800 IOOO 1200 1400 1600 1800 2000 -

- - b

-

=shy

Posterior Distribution of Bias Factor (B) bull Effect of Variation In COV of Prior of B

bull

bullPrior and Posterior Distributions of B Evaluating Effect of Removal of One Platform from Each Category from the Base Case of 13 Platforms

400

~ ~ 350

j IA i

3 00 bull I bull I ---Prior-r I bull l

250 1 I 1

I bull bull ---Posterior r lall 13 casesshy- ~E 2 00 bull bull bull bull bull bull bull Posterior r without STISIK

- ~ 1 bull - bull I - bull bull - bull bull Posterior r Iwithout T2I I 50 I

J-tt~i - bull - bull - Posterior r Iwithout T2S

tOO [-

I I ~ ~ I 0 50

0 00

___ r J i

l~T

~--

0 00 0 20 0 40 0 60 080 I 00 120 140 160 110 2 00

b

zshy

Posterior Distribution of Bias Factor (8) bull Effect of Removal of middotPlatform from Any Category

bullPrior and Posterior Distributions or B Evaluating Effect or Removal or 3 Damage Cases from the Base Case or 13 Platrorms

4

35

3

25

e L 2 ~ bulle shy

I 15

bull I

05

0

0 02 04 06 I 2 I 4 16 IS 2

I I

I

I

I Ibull

J -tmiddot L--

bullI

~

I

fl

II ~

I I v I

middot bull bull

bull L

~ ll- I-~lt

OS I

b

I

bull

1middot bullbullbullbull

~ __

bull

- - - - Prior-r

- - - - Postenor flall failure cases

Postenor flall success cases

--- Posterior flall 13 cases

---Posterior fI success failure cases only

IPosterior or B Mean =t11 COV =0121

-

_I

Posterir Distribution or Bias Factor (B) bull Effect or Removal or 3-Damage Cases

-

bullPrior and Posterior Distributions or B Evaluating Effect or Additional 13 Similar Observations as

Used in Andrew JIP The POsterior or B from Andrew-JIP Considered as New Prior or B

- 500

450

400

350

bull bull

bull bull bull bull J bull

bull bull f ~ I

I ~ Ibull bull

bullbull bull

l bull I -bullbull bull bull

I

I bull bull

bull I I~~bull bull bull

bull ~

I bull bull

~ ~_ ~~~

- - New Prior r bull

Posterior 114 new failure cases ~ 300- bull bull bull bull bull bull bull Posterior f 13 new damage casesl 150 ~

- bull bull - bull bull Posterior 116 new success cases i 2 00

Postenor 1113 new success failure -150 damage cases

I 00

0 50 -

0 00 0 00 0 20 0 40 0 60 0 80 I 00 120 I 40 160 110 2 00

Posterior ol B Mean= 120 COV= 0071 b

shy

shy

_rshy

Posterior Distribution of Bias Factor (B) bull Effect of Additional 13-Similar Observations

1

1 I

WEIGHTING APPROACH bull

bull ASSESSMENT OF PLATFORMS DATA

Select Platforms Most Mappable to the Analyzed Platforms

Unlikely SurvivalDamage Cases Identification

Based on Cut-Off Criteria Agemiddot Hs etc

bull INFORMATION REQUIRED FOR COMPARISON

bull CATEGORIZATION OF ADDITIONALmiddot PLATFORMS

bull UPDATING OF CALIBRATION RESULTS

--

OVERVIEW OF PLATFORMS SURVIVEDDAMAGED -~

Operator -

- Platforms

Survived CS)

Platforms

Damaged CD)

Platforms Used In

Phase IJIP -

JI~ ~Bill~IRBDll -

Amoco -

Chevron -

Exxon

Mob II

Phllllps

Shell

Unocal -

30

143

72 22

7

39 25shy -

1

12

bull

1 bull

1

1(S)+1 (D)

3 (S) +3(D)

2 (S)

bull shy

bull

bull

bull

Total Partlclnants Platforms 338 - 15 10C6S+4D) -

Other Operators

310

13

3 (D)

Total Platforms 648 28 13C6S+7D) -

Summary of Andrew JIP bull Phase I Platforms

Plairshy w v- N-ber Lea-le 1 shy car Hind_ o-w N-bull Deplh fl fl IJpe la bull n Yeadon Jadtet lbelledtf liefalit -- _ -OD

IL H - -shy _ICllshy IL IL - - rQClto ~ _

K STUIK 137 1963 8 365 608S4 Cellardeck

-middot STl-shy 170 1964 4 -~ K 39 62269 Cellardeck - --ST134W 137 1911 4 43 608S4 Smvlnlshy- _ - -

WD90A 184 1964 8 K 267Sbull34 83 SOS31 SUboftllar deck SmvMI

MC311 343 1978 8 x S7 62 - MC397 468 1991 4 x shy S33bull6SO 65911 -shySmeSmvlvol -- gt -

r-- T23ST52 63 1969 4 ton- shyKshy 52594 S03 -T25SSl39 62 1969 4 Kshy SOS94 S063 - middot ton-shy__ -

bullbull a ton--shyST161A 118 1964 8 K - 347S 51495 Cellarclecti

r~

ST1771 142 1965 8 K 39 60167 CeDordeck failme

STISIH 137 1964 8 K 395 608S4 CeDardeck Failme

K ST130A 140 1958 a SS 6095 I lblv Pai1uno - T21ST72 61 1969 4 Kshy S4S94 4974 I lhlv Pai1uno -

- -

JE

INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON

bull WATER DEPTH

bull bull YEAR OF INSTALLATIONDESIGN

bull PLATFORM ORIENTATION

bull PLATFORM LATITUDE AND LONGITUDE

bull KEY SKETCHES OF PLATFORM STRUCTURAL FRAMING

Number of Legs Plies Leg Batter Number of Conductors bull Deck Elevations (Including sub-cellar deck If any) Jacket Framing Elevations Brace type In Vertical Frames Sizes of Primary Members (leg pile vertical frames braces primary horizontals) Joint Details (can gap etc) LegPiie Annulus Grouting Status

bull middotPOST ANDREW INSPECTION REPORT Significant Wave Height Hs (ft)

bull INFORMATION ON DAMAGES PRIOR TO ANDREW

bull RESULTS FROM IN-HOUSE STUDIES IF AVAILABLE

bull ~RIGINAL DESIGN CRITERIA INFORMATION IF AVAILABLE

bull SOIL SHEAR STRENGTH PROFILE IF AVAILABLE

Platforms Identified for Evaluating Therr Likelihood of Falling Under Unexpected Survival Category

OPlRATOR

middotADA

BLOClt

llmucnJllE

NAME

YEAll

INSTAUJD

WA1ER

DBPTll

DISTANCE

JlllOM SHOllE

(MILU)

Bl

(M)

AMOCO 111113

STl11

WDI075

A

bull D

St

17

120

1n

44

20

10 I

CllllVllQfl STlllO c 61 110 21 11 STIUO D Q 1110 21 11

STl134 p Q 137 29 11

ST0151 G 61 137 11

ST0151 I 3 121 11

ST0151 J Q 140 11

STl151 It 137 11

ST0151 PJl001 Q UT 11

ST0151 PROD-2 Q UT 11

ST017 A 3 140 35 11

WDIUT c 65 214 34 10

llXXON STI055 II 51 6T 14 I

STl15 A t3 34

WDI073

WD0073

Amiddotshy Q

1111 1111

22

22

I

I

WDI073 c 1n 22 I

- WDOCl93-shy II

65 1110 2T A 3 194 31 MOBU 111121

SS IOl7

A B

501

50

50

31

T

I

T 6

MtlllPllY SS114 B T 50 T T

PlllllIPS SS0149 A 5 55 33 T

SAMllDAN STln A 107 T T

SBllLL WDl103 A 65 223 2T WDl103 B 65 221 2T WD0104 c 65 221 2T WD0104 D 6T 20 2T

UNOCAL SS020I

SS 020I SSl20I

SS 0209

SS0209

II p

PURll

A

B

3

n 61

103

97

97

95

100

34

34

34

34 34

bullbull bullbullbull

- SS0209 D u 95 34 I

- SS 1209 SSl253

SS0253

G A

c

6T

2

100

165

1T5

34

54

54

bullbull I

I I

I I

SOUTH MARCH

~LAND

N

SOUTH PaTOI ENE ISWltD AREA

MMS Defined Limits ror Offshore Structures Requiring Inspections Following Andrew

t---Jbulln3 T25bull I bullTtl

bull1$1Htl1Abull bull1SIK1779bull

~

middot-DaTA AREA~T

CRANO rsLE

AREA bull1300

bullll4W

=shy

bullPath of Hurricane Andrew with Platforms Used In Calibration

Likelihood Function vs b All Success Case STISI K STIJOQ ST134W WD90A MCJll MCJ97

I 0000

0 9000

0 8000

07000

-Bo6000

tiiii 0 5000 -

04000

0 3000

0 2000

01000

0 0000

0 05 25

ri1rt middot-middot-middot-middot-middot-

I 5 2

b

---11 middot151K(blsu~ccsI

- bull - bull - fk 130Q(bfU~lCSS)

lie bull134W(blsuccc]

lk-WD90A[blsultCS]

- - - lk middotMC397(bfUtlCS]

--o-- lk-MCJll(bbuctcsJ

--- ll (bl 6 out of 6 Mlttcs case]

_

( Likelihood Functions middotSuccess Cases - p - 1 c I

CATEGORIZATION OF ADDITIONAL PLATFORMS

I I I

bull CATEGORIES

Unexpected Survivals

Expected Survivals

bull APPROACH

Identify Corresponding Platform(s) Evaluated In the JIP

Determine Max Wave Height at Platform Location

Evaluate Whether Wave Hit the Deck

middot - Evaluate Likelihood of R and S to Differ from the Base Cases

Review Joint Information

Assess Special Features

bull IDENTIFY NUMBER OF PLATFORMS IN EACH CATEGORY

I

UPDATING OF CALIBRATION RESULTS

bull INCLUDE EFFECT OF ADDITIONAL PLATFORMS ON bullBbull

Consider Increased 1Uncertalntles In R and S

Develop bullNeW Ukellhood Functions for Different Categories

middotLet bullxbull Number of Ukely Unexpected Survival Cases and y Number of Ukely Expected Survival Cases

Determine Combined Likelihood Function lk (b Iadditional platforms) middot

bull REVISED LIKELIHOOD FUNCTION

lk(b I all cases) = lk(b I 9 platforms) middot x lk(b Ix platforms) x lk(b I y platforms)

bull ESTABLISH REVISED BIAS FACTOR B

f8 (b) u f8 (b) lk (b I all cases)

II

t t ~

shy

APPLICATIONS OF BIAS FACTORS

ii APPLICABLE TO PARTICULAR LOAD AND RESISTANCE RECIPE

bull USE BY THE API COMMITTEES AND MMS

Criteria Revisions

Recommended Design Guidelines Revisions bull

bullbull REASSESSMENT STUDIES AND DECISION MAKING

middot

bull RISK ASSESSMENT STUDIES BY THE OPERATORS

Individual Platforms

Fleet of Platforms

middot IMR Program Implementation

bullbull NOT TO BE USED IN DESIGN

I I fl J

EXAMPLE USE OF BIAS FACTORS

bull PHASE I EXAMPLE APPLICATION

bull EVALUATION OF ANNUAL PROBABILITY OF FAILURE

Establish Annual Maximum Seastate Data

Determine Coefficients to Define Base Shear

Determine Ultimate Capacity of Platform

Determine Probablllty of Failure Versus bullbbull

-bull DETERMINE UPDATED PROBABILITY OF FAILURE

Pf - P [S gt R]

Pf I P [~ gt Rb I b] f 8 (b) db

Pr - J P [S gt Rb I b] f bull 8 (b) db

bull TOTAL PROBABILITIES OF FAILURE

Total Pf - 1 bull Il [1 middotPf] - l Pf bull

Total Pf - 1 bull Il [1middot Pf] - l Pfmiddot

Total Pr = 1 - n [1middot PfbullJ = l Pfshy

I I I

EXAMPLE APPLICATION bull PHASE I

PLATFORM ST 151 K

bull GULF OF MEXICO

bull WATER DEPTH = 137 FT

bull ANNUAL MAXIMUM HS MEDIAN 175 ft COV 031

bull ASSUMED STORM DURATION 3 hoursmiddot

bull ASSUMED NUMBER OF WAVES 800

bull MEDIAN HS FOR STORM HOURS 1 2 AND 3

bull MEDIAN CURRENT VALUE bull

bull DETERMINE ULTIMATE CAPACITY FOR THREE DIRECTIONS

bull DETERMINE ANNUAL FAILURE PROBABILITIES

Broadsidemiddot Diagonal End-on

0037 0025 0007

0071 0044 0018

0022 0016 0004

DETERMINE TOTAL ANNUAL PROBABILITY OF FAILURE

Pf - 007

Pf - 013

Pr - 004

bull

EXAMPLE USE OF BIAS FACTORS - PHASE II middot

bull DETERMINE PREDICTIVE PROBABILITY OF FAILURE

Case - 1

Pf _ S S Pf (b b2) fa1a2 (b b2) db db2

g bull -1 (b1 R I S) - 1

92 - (b2 R2 I S) - 1

Case -2

Pf s ~ [bJ f11 (b) db

Pf [bJ p [g lt 0)

g - bull 1 (b1 R1 I S) - 1

1

bull1) l

I

BENEFITS OF ANDREW JIP bull PHASE II

lt

bull IMPROVED LOAD AND RESISTANCE RECIPE

bull IMPROVED CALIBRATION PROCEDURE

bull MULTIPLE BIAS FACTORS

bull PROCEDURE TO INCLUDE EFFECT OF OTHER PLATFORMS

bull PROCEDURE TO USE MULTIPLE BIAS FACTORS

  • Structure Bookmarks
    • ANDREW PHASE II JOINT1 INDUSTRY PROJECT
    • PROGRESS MEETING NUMBER 1
    • ANDREW PHASE II JOINT INDUSTRY PROJECT PROGRESS MEETING NUMBER 1 NOVEMBER 16 AGENDA
    • ANDREW PHASE II UST OF PARTICIPANTS
    • ANDREW PHASE I OBJECTIVES AND RESULTS
    • ANDREW PHASE II PROJECT OBJECTIVES
    • SUBSEQUENT PROJECTS
    • ANDREW PHASE II PROJECT TEAM
    • ANDREW PHASE II UPDATE MODELING middotAND ANALYSIS RECIPE
    • -ANDREW PHASE II CONDUCTOR RESISTANCE
    • ANDREW PHASE Il WA VE FORCE PROFILE
    • ANDREW PHASE II WA VE FORCE PROFILEmiddot
    • ANDREW PHASE II BRACE MODELING
    • ANDREW PHASE II FOUNDATIONS
    • ANDREW PHASE II middot middot JOINT MODELING
    • ANDREW PHASE Il JOINT MODELING OPTIONS DISCUSSION
    • ANDREW PHASE II JOINT MODEJJNG OPTIONS ILLUSTRATIONS
    • ANDREW PHASE II JOINT MODELING PHASE II METHOD
    • ANDREW PHASE II ANDREW STORM CONDmONS UPDATED
    • Platform Selection bullbullbull1111111111111
    • BasisCriteria for Selection bullbullbull1111111111111
    • CALIBRATION
    • WHERE TO INTRODUCE BIAS FACTOR (B
    • PHASE I -KEY CONCLUSIONS
    • bull CALIBRATION ISSUES PHASE II
    • PHASE II WORK SCOPE (CALIBRATION)
    • SOFTWARE
    • SENSITIVITY STUDIES POSTERIOR DISTRIBUTION OF B
    • WEIGHTING APPROACH bull
    • INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON
    • CATEGORIZATION OF ADDITIONAL PLATFORMS
    • UPDATING OF CALIBRATION RESULTS
    • APPLICATIONS OF BIAS FACTORS
    • EXAMPLE USE OF BIAS FACTORS
    • EXAMPLE APPLICATION bull PHASE I PLATFORM ST 151 K
    • EXAMPLE USE OF BIAS FACTORS -PHASE II middot
    • BENEFITS OF ANDREW JIP bull PHASE II
Page 31:  · . ' ANDREW PHASE II PROJECT OBJECTIVES . ' • ·UPDATE MODELING AND ANALYSIS RECIPE· JOINTS ' ' ·BRACES WAVE PROFILE FOUNDATIONS . ' ' CONDUCTORS· • ANALVZE PLATFORMS USING

middot-middot _____________--_

_middotmiddotr_~ --ll--~-middot__

Buckling Beam bull

Strut (K=O 65)

Lateral Displacement

ANDREW PHASE II

FOUNDATIONS

bull ~LICIT NONLINEAR API FOUNDATION USED INiPHASE I

bull FOUNDATIONS

JIP SELECTED MINI-VANE DATA FOR

STRENGTH PROFILE

bull COMMENTS ON BENCHMARK COMPARISONS

- SIGNIFICANT VARIABILITY DUE TO FOUNDATION MODELING METHODS

II Ii

z

ctV 4x SPlO caissonfy-42s soil-api(lf-097step-114)

Inelastic Events Legend Elastic Strut Buckling

Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield

--------shyBeam Clmn Fully Plastic Fracture

~--~~~~~~~~~--~----~__i~~---~~~~__--~

Pro)ect apipile Model South Pelto 10 Version Wed May 18072235 1994 - -

1bullt

+

t

f i bullbullbull

bullbull bullbull

I I

5

~x SPlO Caissonfy=42s soil=api(lf=lOOstep=137)

Inelastic Events Legend

Elastic Strut Bucklilg

Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield

earn Clmn Full~ Plas1c Fracture

ANDREW PHASE II middot middot

JOINT MODELING

middotbull API RIGID-PLASTIC USED IN PHASE I

a ELASTIC-BRITTLE USED IN PHASE IB

amiddot MSL WILL PROVIDE NONLINEAR FORCE-DEFORMATION

AND MOMENT-DEFORMATION RELATIONS FOR PHASE II

a MODELING OPTIONS FOR PHASE II -

l

- UNCO~LEDmiddotONE DIMENSIONAL MEMBER END JOINT MODELING

J Ii

ANDREW PHASE Il

JOINT MODELING OPTIONS DISCUSSION

bull JD COUPLED INTERACTION SURFACE

- IDEAL

- DATA NOT AVAILABLE TO CALIBRATE

bull 2D COUPLED INTERACTION SURFACE

- SIGNIFICANT MODELING IMPROVEMENT

- DATAAVAILABLE

- IMPLEMENTATION PRACTICAL BUT NEEDS DEVELOPMENT AND CALIBRATION

bull 2D UNCOUPLED INTERACTION SURFACE

- SIGNIFICANT MODELING IMPROVEMENT

- DATA AVAILABLE

- IMPLEMENTATION PRACTICAL AND AVAILABLE

ANDREW PHASE II

JOINT MODEJJNG OPTIONS ILLUSTRATIONS

P

e

ANDREW PHASE II

JOINT MODELING PHASE II METHOD

e

I I I

ANDREW PHASE II

ANDREW STORM CONDmONS UPDATED

bull -PHASE I USED OCEANWEATHER HINDCAST-(Nov 1992)

bull middotmiddotPHASE II WILL USE THE REVISED HINDCAST (Nov 1994)

bull lt c lt I lt

2-2

l---4-~~~+-~~-+-~~~1--~~-+-~~--l--~T--+-~~--i24

f-tJVtMfiz~ 1Cfl (r STvb(

Plollol bullbull 5middotNOYmiddot9214bull1J Irbullbull Illbull [ANDREVl~NDREVCUSTbullMAXll 5-NOY-1992 10bull54 - 9208

HEIGHT lml

I~ MAXI

I L 1~middotamp SZ~- ~A I 13c

I J) ( r I X~0iWJ~0gtJ gt L gt gt I 1c I 120

-- N

~ deg

I I I I I I e C I ~=-41 122 -94 -92 -90 -88 -86 -84 -82 -80

-

Figure 7 contours of maximum hindcast significant wave height

_

~

~tl ~

NOltNttIJ 17lt ltNS S iJJ)y

Maximum Significant Wave Height - Hurricane Andrew 1992

30 ~

26 ~ I ~-~~~- == -shy -~ 35-_

24deg1--shy --------~- middot-middot-------middotmiddotmiddot----middot------middot-middot~middot- - - ~-----1---middot

22middot ~~~~~~~-~ -94 -92 -90 -88 -84deg-86 -82 -somiddot

c

-

Platform Selection bullbullbull1111111111111

0 BasisCriteria for Selection middot

0 Review of Phase I Platforms middot 0 Alternative Platforms

0 PMB Recommendations

0 Participants Comments

gt

L

l I

BasisCriteria for Selection bullbullbull1111111111111

0 Benefit for Calibration - Classification (UnexpectedExpected)

- Failure Mode (StructuresFoundations)

- Damaged StructuresKnown response

- Better Representation of Full ~opulation

0 Availability of Data - Existing Models

- Inspection Data

- Damage Data

PREDICTED BEHAVIOR

-

OBSERVED

BEHAVIOR

-SURVIVED DAMAGED FAILED

SURVIVED

ST134W WD90A

-MC311 -

MC397 -

--

-

-

ST151K ST130Q

-

-

-

DAMAGED - -

-

-

--

-

-

ST52 SS139 ST177B -

ST161A

-_

-

FAILED

-

-

-

ST151H ST130A ST72

Operator - Platfonns - Platfonns Survived Damagedamp

- Falled -

bull

-

-

Jle eBDIGIPllDbl -Amocoshy 30 1 Chevron -

143 12 Exxon 72 Mobll 22 1 Phllllps 7 bull

-Shell 39 Unocal 25 1

Total Partlclnts Platfonns 338 15

Other Operators bull- 310 13

-

Total Platfonns - -

648 28

fbml

Platforms Considered 6 7

tie of Total Platfonns 1 25

OVERVIEW OF PLATFORMS SURVIVEDDAMAGEDFAILED

-

shy

v

Platforms Evaluated in Phase I

-

-Platfonn Water Year Number Analysis Avallabillty of Availability Platform

- Name Depth rl of Model DetaDed rl ClassUlcatlon

Installation legs Inspection Soll Report Based on Information Phase I Calibration - -ft

- -- Suaill flalfnrm ~BSi -STISIK 137 1963 8 CAP Yes Yes Un eel Sarvlval

-STIJOO 170 1964 4 CAPmiddotshy Yes ST134data u ST134W 137 1981 4 CAP Yes Yesmiddot E

-

edSarvlval -

middotsurvival

WD90A 184 - 1964 8 CAP Yes E -

- Sarvlvill

MC3ll 343 1978 8 KARMA

MC397 468 1991 4 KARMA -

bull ~reglmmiddot- -

Yes Sure Survival

bull Yes - Sare Sanlval-

-T23STS2 63 1969 4 CAP

T2SSS139 62 1969 4 CAP -

Yes

Yes

Yes Ex

Yes Une

bull SurvivalLikely to DamalI

middot bull SanlvalLlkely to DamalI

ST161A 118 1964 8 USFOS

Eail11a pound1al(bullllm Clss ST177B 142 --1965 8 CAP

STISIH 137 1964 - 8 CAP C

STIJOA 140 19S8 8 CAP - -

Yes

Yes

ollamed

Yes

- Yes E

-

at about 1 mUe

Yes

ST134data

bull SafvlvalLlkelv to Dam- bull- -

E middotrahre

Emeried Failure -

Likely FaUare

T21ST72 61 1969 4 CAP Collamed No Likely FaUure

Platforms Identified for Evaluating Their Likelihood of Falling Under UnexpeCted Survival Category

CllEllATOa

AllB

llLOCll

Sl1lvcnJllE

NAME

YEAR

INSTAUJCD

WATER

IJBPJB

middotmiddotDISTANCE

ftOMSllOllE

(MIUIS)

Ill

(M)

AMOCO BllW

STllI

WDI075

A

bull D

4

n uo 17J

44

32

bull 20

10

bull CllEVllOH STtllO c I 1bull 21 11

STl130 D 0 161 21 11 STllJ4 I 0 137 D - 11

STIUI G I 137 32 11

STIUI I 0 121 32 11

STIUI

J 0 140 32 11

STIUI It 4 137 32 11

STllSI PROl)I 0 137 32 11

STllSI PROD-2 0 137 32 11

STll7 A 0

140 35 11

WDl117 c 65 214 34 10

DXON STshy E 51 14 bull STl165 A t3 bull 34 WD0073

WD0073

WD0073

A _ c

4

62

4

IA IA

17J

22

22

22

bullbullbull WD003-shy E 65 161 Z7

A 0 IN 31 MOBD Bl12I

ssn A bull

50

50

31

7

bull 7

MIJllPllY SS114 B 7 50 7 7

PlllUIPS SSl14t A 55 33 7

SAMEDAN ST17J A 4 117 7 bull 7

SJIELL WD0103 A 65 223 27 WDllOJ bull 65 221 27 WDll04 c 65 221 27 WDOI04 D 67 2 27

UNOCAL

SSl20I

SSl20I

bull SSl20I

SS0209

SS0209

SS0209

SS0209

SS 0253

SS 0253

E

I

PtJllE

A

bull D

G A

c

0

4

4

7J

61

0

0

69

103

97

97

9S

100

95

100

165

175

34

34

34

34

34

34

34

54

54

bullbullbullbullbullbullbullbullbull

Platforms Recommended for Phase II

llalr y_ w- Aot Av- _ I 1 Doplla Model nm bull bull _ bull I I H11 s18-1 Irht lbwl~ bullbullbull(-illtdlel OM lllMtD

II ---

-STISIIC 137 1963 I CAP Yeo Yeo I Yeo

STtshy -170 1964 4 CAP Yeo ST134dlU I Yeo

STl34W 137 Yeo Yeo -1981 4 CAP D Yeo

WD90A 114 1964 I CAP Yeo - D

MC311 30 1971 I KARMA Yeo -sre-

MC397 middot 461 1991 4 ICARMA Yeo Sireshy - - r- -

__ T13ST5l 63 1969 4 CAP Yeo Yeobull middot- ~n- D Yeo _ Tl5SSl39 62 1969 4 CAP Yeo Yeo I Yeo

_ STIT711 142 1965 I CAP Yeo I Yeo middot-1shySTl61A Ill 1964 I USFOS Yeo Yeo - D bulloshy

- r-

STISIH 137 1964 I CAP y D y -_ STl31A 140 1951 I CAP Yeo STl34dlU I Yeo

T21Sll 61 1969 4 CAP No rar-middot I Yeo

~-rmiddot-middot SS209A 95 1967 16 CAP Yeo D -SSU4R

11 UI A 41 -in D SS 2741AIM CAP bull

_ _ -shy - -shy- -shy

CALIBRATION

bull PHASEIAPPROACH

bull CALIBRATION ISSUES

bull PHASE II WORK SCOPE

bull SOFlWARE AVAILABILITY

bull EFFECT OF ADDITIONAL PLATFORMS

bull WEIGHTING APPROACH

bull EXAMPLE APPLICATION middot

(l=Ofce Mode] Wind

~

fcurrent l -middot

~SW ~leldl4

ultimate resistance force I first

design load Illgt

i _

deflection

-

bull

4

Historical Hurrlcane Dariiage and Survival of Platforms

Components Required for Calibration (Petrauskas 1992)

PRIOR Distribution

PLATFORM EXPERIENCE

BAYESIAN Survivals and Failures

UPDATING

Waves Winds and Currents bull

POSTERIOR Distribution

- A_middotJA ~ Survival Data Failure Data

Result Result

Load Resistance

middotfl

-

Bayesian Updaing middotGeneral Description (Petrauskas 1992)

L Imiddot

WHERE TO INTRODUCE BIAS FACTOR (B

On Resistance (Rmue - B (RPREDICreD I

middotOn Loads (Smue middot - B (SPRE01cTEo

On Safety Margin (RSmue - B (RSPREDICTED

CALIBRATION METHODOLOGY

gt

r

FIELD OBSERVATIONS

Observed Behavior

survivals Damages

Failures

r

bull

CAPACITY ANALYSIS

Predicted Load and Resistance Predicted Failure Modes

Survival Damage Failure cases

RELIABILITY ANALYSIS

- Probability of Failure ~

Ukellhood Failure

BAYESIAN UPDATING

Prior Distnbution of B Combined Likelihood Function

Posterior Dlstnbution of B

BIAS FACTOR

Mean Value cov bull

Represents Modeling Uncertainties

Prior Distribution of B Likelihood Functionmiddot Observing Success case STISIK and Posterior Distribution of B

2 00

uo

160

140

120

e 100 L $ uoIshy

0 60 ~ ~ 040

020

0 00

0 00 0 50

_ I 00

b

I 50 2 00

-shy lk [bl STISIK)

--shy Pnor-f(b)

--shy Pos1enor r(bl ISIK)

Prior or B Mean bull 10 COV bull 030

Posterlor of Bbull Meanbull 136COV 0 16

2 50

Likelihood Function and Posterior Distribution Example Platform STISIK

- -

~

- Prior and Posterior Distributions or B Failure cases (ST177B STlSlH ST130A T21) Damage cases (T23 T25 ST161A) Success cases (STlSlK ST130Q ST134W WD90A MC311 MC397)

400

350

300

2 so 0-

I

bull bull

bull bull gt

- bullbull ~ bull bull I bull

_ bull bull bullbull I bull bull bullII V -

bull bull

I bull ~ bull

bullbull bullI bull bullI bull bull bull

middot -~

bull__ ~ middotI middotmiddot

L 2 00

e ~ I SO

-I 00

- 0 so

000 gt

0 00 020 0 40 060 0 80 I 00 120 140 160 110 200

b -

-

shy

Prior-r

- - - - Posterior fLlll failure cases

bull bull bull bull bull bull bull Posterior flall damage cases

- bull bull - bull middot Postenor flall success cases

Posterior flall success fulure damage cases

-

Posterior of B Mean= 119 COV 0101

Posterior Distribution or Blas Factor (8) bull Different Categories and All Cases Combined

shy

l

PHASE I - KEY CONCLUSIONS

bull GLOBAL CORRECTION FACTOR

bull 15 TO 20 CONSERVATISM MEAN)

bull UNEXPECTED SURVIVALS HAVE GREATEST IMPACT

bull DAMAGED CASES MORE INFORMATIVE

bull CALIBRATION ISSUES PHASE II

bull

EFFECT OF IMPROVED CAPACITY ANALYSIS RECIPE

bull APPLICABILITY TO MULTIPLE FAILURE MODES

bull NO FOUNDATION DAMAGE CASE INCLUDED

bull IMPROVED REPRESENTATION OF FULL ANDREW DISTRIBUTION

-bull NUMBER OF PLATFORMS CONSIDERED

I 1 I

I t ) bull ~

PHASE II WORK SCOPE (CALIBRATION)

bull

bull ESTABLISH MULTIPLE BIAS FACTORS

Use Nine Base Case Platforms

middot - Capacity Analysis for Revised recipe

Foundation Blas Factor from Foundation JIP

bull DEVELOP WEIGHTING APPROACH

bull UPDATE MULTIPLE BIAS FACTORS

bullEffect of Platforms Not In the Nine Base Cases

bull EXAMPLE APPLICATION OF BIAS FACTORS I

bullbull

PHASE I PHASE II

CAPACITYbull ANALYSISRu bull

---------------------1----middot

Jc1 - iJbullraquoJ dbull

FttlS frt[f~l ratdhJJ

-

LIKELIHOOD FUNCTION

DEVELOPMENT

lk Jb Jfall1N) bull Jb)

111 Ill Isurvtvoq bull 1 - bl

-middot -------------------shy ~---------------------middot-middot

r111ir8 Cb1 111 Cbl BAYESIAN

UPDATING

~cove

R1R2 bull middot---------------------~----

Jblbull P(g cOJ

1 bull JbRI (MBSI I bull 1

~middot

llcJllJfalluro) bull P(gcOJ Ill 111 b21fallun) bull P ((g1 CO) U 1112 c OD

middot--------------------shy ----------------------shy bull

r111111r8 Cbl 111 Ill i-1nro1

deg ~1b2)laquottfrr (b1b21-lnfo) re Ill bull 1-11middot b2) bull lodegamp

middot

i1 covamp i2 cov~i

-

shy

--

Likelihood Function vs b All Success Case STISI K STl30Q STl34W WD90A MC311 MC397

I 0000

0 9000

OIOOO

0 7000

~ 0amp000

ta osoOO

~ 0 4000

0 3000

0 2000

0 1000

bull 0 0000

bull

I

middot

middot bull I

Ibull

-sibull ~ -middot-middot-middot-middot -middot-

----7----+-----~ I

Ibull bull bull I

rbull

--- Jk ISIK[blsuccess)

- bull - bull - lk 130Q[blsuccess) bull

lk 134W[blsuccess)

Jk-WD90A(blsuccc~I

- - - lk MC397[blsucccss)

--ltgt-- lk middotMC31 l[blsucccss)

--- Jk [bl 6 out of6succcsHascs)

05 I 5 2 250

b

Likelihood Functionsmiddot Success Cases - Phbullsc L

shy

bullbull bull bull bull bull bull -------------------~

bull bull bullbull bullbull bullbull bullbull bullbull bullbull bull

bull bullbull

bullbullbullbull

bullbull

bullbullbullbullbullbullbull

bullbullbullbullbullbull

bullbullbullbullbullbullbull

z 0

t z ) LL

c 0 0c i ~ i

J

~ sect ~--------------~ e l5

I

I

SOFTWARE

bull PF PROGRAM

bull CALREL

bull PF +CALREL

bull RELACS

bull

gt i

l

SENSITIVITY STUDIES

POSTERIOR DISTRIBUTION OF B

bull EFFECT OF CHANGE IN DISTRIBUTION OF PRIOR OF B

bull EFFECT OF REMOVAL OF ONE PLATFORM FROM EACH CATEGORY

bull EFFECT OF REMOVAL OF THREE DAMAGE CASES

bull EFFECT OF ADDITIONAL 13 SIMILAR PLATFORMS

-I

Prior and Posterior Distributions of B Variation In COV of PriorofB fromOl to04 -- -

4500

-4000

-

- -

3500 --middot - bull bull - bull bull Prior-r ICOV =02

3000

- - - - Prior-r ICOV =03 -

bull -~ e 2soo -

bull bull bull bull bull bull bull Pnor-r ICOV =o 4t frJ ~ 2000 Posterior rlCOV =0 2

bull bull Postenor rlCOV =0 3 bull bull 1500 bull middot bull - shy -middot- Posterior rlCOV =04

middot~ I 000 - bull

-middotmiddot ~ ~middotmiddot M -~

- middot 0 500

wbullbullbull _ I middot~ middotmiddotmiddotmiddotmiddot middotmiddotmiddotmiddotmiddot - -~ ~ __ middot_ _ bull -shy0 000

0 000 0200 0 400 0 600 0 800 IOOO 1200 1400 1600 1800 2000 -

- - b

-

=shy

Posterior Distribution of Bias Factor (B) bull Effect of Variation In COV of Prior of B

bull

bullPrior and Posterior Distributions of B Evaluating Effect of Removal of One Platform from Each Category from the Base Case of 13 Platforms

400

~ ~ 350

j IA i

3 00 bull I bull I ---Prior-r I bull l

250 1 I 1

I bull bull ---Posterior r lall 13 casesshy- ~E 2 00 bull bull bull bull bull bull bull Posterior r without STISIK

- ~ 1 bull - bull I - bull bull - bull bull Posterior r Iwithout T2I I 50 I

J-tt~i - bull - bull - Posterior r Iwithout T2S

tOO [-

I I ~ ~ I 0 50

0 00

___ r J i

l~T

~--

0 00 0 20 0 40 0 60 080 I 00 120 140 160 110 2 00

b

zshy

Posterior Distribution of Bias Factor (8) bull Effect of Removal of middotPlatform from Any Category

bullPrior and Posterior Distributions or B Evaluating Effect or Removal or 3 Damage Cases from the Base Case or 13 Platrorms

4

35

3

25

e L 2 ~ bulle shy

I 15

bull I

05

0

0 02 04 06 I 2 I 4 16 IS 2

I I

I

I

I Ibull

J -tmiddot L--

bullI

~

I

fl

II ~

I I v I

middot bull bull

bull L

~ ll- I-~lt

OS I

b

I

bull

1middot bullbullbullbull

~ __

bull

- - - - Prior-r

- - - - Postenor flall failure cases

Postenor flall success cases

--- Posterior flall 13 cases

---Posterior fI success failure cases only

IPosterior or B Mean =t11 COV =0121

-

_I

Posterir Distribution or Bias Factor (B) bull Effect or Removal or 3-Damage Cases

-

bullPrior and Posterior Distributions or B Evaluating Effect or Additional 13 Similar Observations as

Used in Andrew JIP The POsterior or B from Andrew-JIP Considered as New Prior or B

- 500

450

400

350

bull bull

bull bull bull bull J bull

bull bull f ~ I

I ~ Ibull bull

bullbull bull

l bull I -bullbull bull bull

I

I bull bull

bull I I~~bull bull bull

bull ~

I bull bull

~ ~_ ~~~

- - New Prior r bull

Posterior 114 new failure cases ~ 300- bull bull bull bull bull bull bull Posterior f 13 new damage casesl 150 ~

- bull bull - bull bull Posterior 116 new success cases i 2 00

Postenor 1113 new success failure -150 damage cases

I 00

0 50 -

0 00 0 00 0 20 0 40 0 60 0 80 I 00 120 I 40 160 110 2 00

Posterior ol B Mean= 120 COV= 0071 b

shy

shy

_rshy

Posterior Distribution of Bias Factor (B) bull Effect of Additional 13-Similar Observations

1

1 I

WEIGHTING APPROACH bull

bull ASSESSMENT OF PLATFORMS DATA

Select Platforms Most Mappable to the Analyzed Platforms

Unlikely SurvivalDamage Cases Identification

Based on Cut-Off Criteria Agemiddot Hs etc

bull INFORMATION REQUIRED FOR COMPARISON

bull CATEGORIZATION OF ADDITIONALmiddot PLATFORMS

bull UPDATING OF CALIBRATION RESULTS

--

OVERVIEW OF PLATFORMS SURVIVEDDAMAGED -~

Operator -

- Platforms

Survived CS)

Platforms

Damaged CD)

Platforms Used In

Phase IJIP -

JI~ ~Bill~IRBDll -

Amoco -

Chevron -

Exxon

Mob II

Phllllps

Shell

Unocal -

30

143

72 22

7

39 25shy -

1

12

bull

1 bull

1

1(S)+1 (D)

3 (S) +3(D)

2 (S)

bull shy

bull

bull

bull

Total Partlclnants Platforms 338 - 15 10C6S+4D) -

Other Operators

310

13

3 (D)

Total Platforms 648 28 13C6S+7D) -

Summary of Andrew JIP bull Phase I Platforms

Plairshy w v- N-ber Lea-le 1 shy car Hind_ o-w N-bull Deplh fl fl IJpe la bull n Yeadon Jadtet lbelledtf liefalit -- _ -OD

IL H - -shy _ICllshy IL IL - - rQClto ~ _

K STUIK 137 1963 8 365 608S4 Cellardeck

-middot STl-shy 170 1964 4 -~ K 39 62269 Cellardeck - --ST134W 137 1911 4 43 608S4 Smvlnlshy- _ - -

WD90A 184 1964 8 K 267Sbull34 83 SOS31 SUboftllar deck SmvMI

MC311 343 1978 8 x S7 62 - MC397 468 1991 4 x shy S33bull6SO 65911 -shySmeSmvlvol -- gt -

r-- T23ST52 63 1969 4 ton- shyKshy 52594 S03 -T25SSl39 62 1969 4 Kshy SOS94 S063 - middot ton-shy__ -

bullbull a ton--shyST161A 118 1964 8 K - 347S 51495 Cellarclecti

r~

ST1771 142 1965 8 K 39 60167 CeDordeck failme

STISIH 137 1964 8 K 395 608S4 CeDardeck Failme

K ST130A 140 1958 a SS 6095 I lblv Pai1uno - T21ST72 61 1969 4 Kshy S4S94 4974 I lhlv Pai1uno -

- -

JE

INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON

bull WATER DEPTH

bull bull YEAR OF INSTALLATIONDESIGN

bull PLATFORM ORIENTATION

bull PLATFORM LATITUDE AND LONGITUDE

bull KEY SKETCHES OF PLATFORM STRUCTURAL FRAMING

Number of Legs Plies Leg Batter Number of Conductors bull Deck Elevations (Including sub-cellar deck If any) Jacket Framing Elevations Brace type In Vertical Frames Sizes of Primary Members (leg pile vertical frames braces primary horizontals) Joint Details (can gap etc) LegPiie Annulus Grouting Status

bull middotPOST ANDREW INSPECTION REPORT Significant Wave Height Hs (ft)

bull INFORMATION ON DAMAGES PRIOR TO ANDREW

bull RESULTS FROM IN-HOUSE STUDIES IF AVAILABLE

bull ~RIGINAL DESIGN CRITERIA INFORMATION IF AVAILABLE

bull SOIL SHEAR STRENGTH PROFILE IF AVAILABLE

Platforms Identified for Evaluating Therr Likelihood of Falling Under Unexpected Survival Category

OPlRATOR

middotADA

BLOClt

llmucnJllE

NAME

YEAll

INSTAUJD

WA1ER

DBPTll

DISTANCE

JlllOM SHOllE

(MILU)

Bl

(M)

AMOCO 111113

STl11

WDI075

A

bull D

St

17

120

1n

44

20

10 I

CllllVllQfl STlllO c 61 110 21 11 STIUO D Q 1110 21 11

STl134 p Q 137 29 11

ST0151 G 61 137 11

ST0151 I 3 121 11

ST0151 J Q 140 11

STl151 It 137 11

ST0151 PJl001 Q UT 11

ST0151 PROD-2 Q UT 11

ST017 A 3 140 35 11

WDIUT c 65 214 34 10

llXXON STI055 II 51 6T 14 I

STl15 A t3 34

WDI073

WD0073

Amiddotshy Q

1111 1111

22

22

I

I

WDI073 c 1n 22 I

- WDOCl93-shy II

65 1110 2T A 3 194 31 MOBU 111121

SS IOl7

A B

501

50

50

31

T

I

T 6

MtlllPllY SS114 B T 50 T T

PlllllIPS SS0149 A 5 55 33 T

SAMllDAN STln A 107 T T

SBllLL WDl103 A 65 223 2T WDl103 B 65 221 2T WD0104 c 65 221 2T WD0104 D 6T 20 2T

UNOCAL SS020I

SS 020I SSl20I

SS 0209

SS0209

II p

PURll

A

B

3

n 61

103

97

97

95

100

34

34

34

34 34

bullbull bullbullbull

- SS0209 D u 95 34 I

- SS 1209 SSl253

SS0253

G A

c

6T

2

100

165

1T5

34

54

54

bullbull I

I I

I I

SOUTH MARCH

~LAND

N

SOUTH PaTOI ENE ISWltD AREA

MMS Defined Limits ror Offshore Structures Requiring Inspections Following Andrew

t---Jbulln3 T25bull I bullTtl

bull1$1Htl1Abull bull1SIK1779bull

~

middot-DaTA AREA~T

CRANO rsLE

AREA bull1300

bullll4W

=shy

bullPath of Hurricane Andrew with Platforms Used In Calibration

Likelihood Function vs b All Success Case STISI K STIJOQ ST134W WD90A MCJll MCJ97

I 0000

0 9000

0 8000

07000

-Bo6000

tiiii 0 5000 -

04000

0 3000

0 2000

01000

0 0000

0 05 25

ri1rt middot-middot-middot-middot-middot-

I 5 2

b

---11 middot151K(blsu~ccsI

- bull - bull - fk 130Q(bfU~lCSS)

lie bull134W(blsuccc]

lk-WD90A[blsultCS]

- - - lk middotMC397(bfUtlCS]

--o-- lk-MCJll(bbuctcsJ

--- ll (bl 6 out of 6 Mlttcs case]

_

( Likelihood Functions middotSuccess Cases - p - 1 c I

CATEGORIZATION OF ADDITIONAL PLATFORMS

I I I

bull CATEGORIES

Unexpected Survivals

Expected Survivals

bull APPROACH

Identify Corresponding Platform(s) Evaluated In the JIP

Determine Max Wave Height at Platform Location

Evaluate Whether Wave Hit the Deck

middot - Evaluate Likelihood of R and S to Differ from the Base Cases

Review Joint Information

Assess Special Features

bull IDENTIFY NUMBER OF PLATFORMS IN EACH CATEGORY

I

UPDATING OF CALIBRATION RESULTS

bull INCLUDE EFFECT OF ADDITIONAL PLATFORMS ON bullBbull

Consider Increased 1Uncertalntles In R and S

Develop bullNeW Ukellhood Functions for Different Categories

middotLet bullxbull Number of Ukely Unexpected Survival Cases and y Number of Ukely Expected Survival Cases

Determine Combined Likelihood Function lk (b Iadditional platforms) middot

bull REVISED LIKELIHOOD FUNCTION

lk(b I all cases) = lk(b I 9 platforms) middot x lk(b Ix platforms) x lk(b I y platforms)

bull ESTABLISH REVISED BIAS FACTOR B

f8 (b) u f8 (b) lk (b I all cases)

II

t t ~

shy

APPLICATIONS OF BIAS FACTORS

ii APPLICABLE TO PARTICULAR LOAD AND RESISTANCE RECIPE

bull USE BY THE API COMMITTEES AND MMS

Criteria Revisions

Recommended Design Guidelines Revisions bull

bullbull REASSESSMENT STUDIES AND DECISION MAKING

middot

bull RISK ASSESSMENT STUDIES BY THE OPERATORS

Individual Platforms

Fleet of Platforms

middot IMR Program Implementation

bullbull NOT TO BE USED IN DESIGN

I I fl J

EXAMPLE USE OF BIAS FACTORS

bull PHASE I EXAMPLE APPLICATION

bull EVALUATION OF ANNUAL PROBABILITY OF FAILURE

Establish Annual Maximum Seastate Data

Determine Coefficients to Define Base Shear

Determine Ultimate Capacity of Platform

Determine Probablllty of Failure Versus bullbbull

-bull DETERMINE UPDATED PROBABILITY OF FAILURE

Pf - P [S gt R]

Pf I P [~ gt Rb I b] f 8 (b) db

Pr - J P [S gt Rb I b] f bull 8 (b) db

bull TOTAL PROBABILITIES OF FAILURE

Total Pf - 1 bull Il [1 middotPf] - l Pf bull

Total Pf - 1 bull Il [1middot Pf] - l Pfmiddot

Total Pr = 1 - n [1middot PfbullJ = l Pfshy

I I I

EXAMPLE APPLICATION bull PHASE I

PLATFORM ST 151 K

bull GULF OF MEXICO

bull WATER DEPTH = 137 FT

bull ANNUAL MAXIMUM HS MEDIAN 175 ft COV 031

bull ASSUMED STORM DURATION 3 hoursmiddot

bull ASSUMED NUMBER OF WAVES 800

bull MEDIAN HS FOR STORM HOURS 1 2 AND 3

bull MEDIAN CURRENT VALUE bull

bull DETERMINE ULTIMATE CAPACITY FOR THREE DIRECTIONS

bull DETERMINE ANNUAL FAILURE PROBABILITIES

Broadsidemiddot Diagonal End-on

0037 0025 0007

0071 0044 0018

0022 0016 0004

DETERMINE TOTAL ANNUAL PROBABILITY OF FAILURE

Pf - 007

Pf - 013

Pr - 004

bull

EXAMPLE USE OF BIAS FACTORS - PHASE II middot

bull DETERMINE PREDICTIVE PROBABILITY OF FAILURE

Case - 1

Pf _ S S Pf (b b2) fa1a2 (b b2) db db2

g bull -1 (b1 R I S) - 1

92 - (b2 R2 I S) - 1

Case -2

Pf s ~ [bJ f11 (b) db

Pf [bJ p [g lt 0)

g - bull 1 (b1 R1 I S) - 1

1

bull1) l

I

BENEFITS OF ANDREW JIP bull PHASE II

lt

bull IMPROVED LOAD AND RESISTANCE RECIPE

bull IMPROVED CALIBRATION PROCEDURE

bull MULTIPLE BIAS FACTORS

bull PROCEDURE TO INCLUDE EFFECT OF OTHER PLATFORMS

bull PROCEDURE TO USE MULTIPLE BIAS FACTORS

  • Structure Bookmarks
    • ANDREW PHASE II JOINT1 INDUSTRY PROJECT
    • PROGRESS MEETING NUMBER 1
    • ANDREW PHASE II JOINT INDUSTRY PROJECT PROGRESS MEETING NUMBER 1 NOVEMBER 16 AGENDA
    • ANDREW PHASE II UST OF PARTICIPANTS
    • ANDREW PHASE I OBJECTIVES AND RESULTS
    • ANDREW PHASE II PROJECT OBJECTIVES
    • SUBSEQUENT PROJECTS
    • ANDREW PHASE II PROJECT TEAM
    • ANDREW PHASE II UPDATE MODELING middotAND ANALYSIS RECIPE
    • -ANDREW PHASE II CONDUCTOR RESISTANCE
    • ANDREW PHASE Il WA VE FORCE PROFILE
    • ANDREW PHASE II WA VE FORCE PROFILEmiddot
    • ANDREW PHASE II BRACE MODELING
    • ANDREW PHASE II FOUNDATIONS
    • ANDREW PHASE II middot middot JOINT MODELING
    • ANDREW PHASE Il JOINT MODELING OPTIONS DISCUSSION
    • ANDREW PHASE II JOINT MODEJJNG OPTIONS ILLUSTRATIONS
    • ANDREW PHASE II JOINT MODELING PHASE II METHOD
    • ANDREW PHASE II ANDREW STORM CONDmONS UPDATED
    • Platform Selection bullbullbull1111111111111
    • BasisCriteria for Selection bullbullbull1111111111111
    • CALIBRATION
    • WHERE TO INTRODUCE BIAS FACTOR (B
    • PHASE I -KEY CONCLUSIONS
    • bull CALIBRATION ISSUES PHASE II
    • PHASE II WORK SCOPE (CALIBRATION)
    • SOFTWARE
    • SENSITIVITY STUDIES POSTERIOR DISTRIBUTION OF B
    • WEIGHTING APPROACH bull
    • INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON
    • CATEGORIZATION OF ADDITIONAL PLATFORMS
    • UPDATING OF CALIBRATION RESULTS
    • APPLICATIONS OF BIAS FACTORS
    • EXAMPLE USE OF BIAS FACTORS
    • EXAMPLE APPLICATION bull PHASE I PLATFORM ST 151 K
    • EXAMPLE USE OF BIAS FACTORS -PHASE II middot
    • BENEFITS OF ANDREW JIP bull PHASE II
Page 32:  · . ' ANDREW PHASE II PROJECT OBJECTIVES . ' • ·UPDATE MODELING AND ANALYSIS RECIPE· JOINTS ' ' ·BRACES WAVE PROFILE FOUNDATIONS . ' ' CONDUCTORS· • ANALVZE PLATFORMS USING

ANDREW PHASE II

FOUNDATIONS

bull ~LICIT NONLINEAR API FOUNDATION USED INiPHASE I

bull FOUNDATIONS

JIP SELECTED MINI-VANE DATA FOR

STRENGTH PROFILE

bull COMMENTS ON BENCHMARK COMPARISONS

- SIGNIFICANT VARIABILITY DUE TO FOUNDATION MODELING METHODS

II Ii

z

ctV 4x SPlO caissonfy-42s soil-api(lf-097step-114)

Inelastic Events Legend Elastic Strut Buckling

Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield

--------shyBeam Clmn Fully Plastic Fracture

~--~~~~~~~~~--~----~__i~~---~~~~__--~

Pro)ect apipile Model South Pelto 10 Version Wed May 18072235 1994 - -

1bullt

+

t

f i bullbullbull

bullbull bullbull

I I

5

~x SPlO Caissonfy=42s soil=api(lf=lOOstep=137)

Inelastic Events Legend

Elastic Strut Bucklilg

Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield

earn Clmn Full~ Plas1c Fracture

ANDREW PHASE II middot middot

JOINT MODELING

middotbull API RIGID-PLASTIC USED IN PHASE I

a ELASTIC-BRITTLE USED IN PHASE IB

amiddot MSL WILL PROVIDE NONLINEAR FORCE-DEFORMATION

AND MOMENT-DEFORMATION RELATIONS FOR PHASE II

a MODELING OPTIONS FOR PHASE II -

l

- UNCO~LEDmiddotONE DIMENSIONAL MEMBER END JOINT MODELING

J Ii

ANDREW PHASE Il

JOINT MODELING OPTIONS DISCUSSION

bull JD COUPLED INTERACTION SURFACE

- IDEAL

- DATA NOT AVAILABLE TO CALIBRATE

bull 2D COUPLED INTERACTION SURFACE

- SIGNIFICANT MODELING IMPROVEMENT

- DATAAVAILABLE

- IMPLEMENTATION PRACTICAL BUT NEEDS DEVELOPMENT AND CALIBRATION

bull 2D UNCOUPLED INTERACTION SURFACE

- SIGNIFICANT MODELING IMPROVEMENT

- DATA AVAILABLE

- IMPLEMENTATION PRACTICAL AND AVAILABLE

ANDREW PHASE II

JOINT MODEJJNG OPTIONS ILLUSTRATIONS

P

e

ANDREW PHASE II

JOINT MODELING PHASE II METHOD

e

I I I

ANDREW PHASE II

ANDREW STORM CONDmONS UPDATED

bull -PHASE I USED OCEANWEATHER HINDCAST-(Nov 1992)

bull middotmiddotPHASE II WILL USE THE REVISED HINDCAST (Nov 1994)

bull lt c lt I lt

2-2

l---4-~~~+-~~-+-~~~1--~~-+-~~--l--~T--+-~~--i24

f-tJVtMfiz~ 1Cfl (r STvb(

Plollol bullbull 5middotNOYmiddot9214bull1J Irbullbull Illbull [ANDREVl~NDREVCUSTbullMAXll 5-NOY-1992 10bull54 - 9208

HEIGHT lml

I~ MAXI

I L 1~middotamp SZ~- ~A I 13c

I J) ( r I X~0iWJ~0gtJ gt L gt gt I 1c I 120

-- N

~ deg

I I I I I I e C I ~=-41 122 -94 -92 -90 -88 -86 -84 -82 -80

-

Figure 7 contours of maximum hindcast significant wave height

_

~

~tl ~

NOltNttIJ 17lt ltNS S iJJ)y

Maximum Significant Wave Height - Hurricane Andrew 1992

30 ~

26 ~ I ~-~~~- == -shy -~ 35-_

24deg1--shy --------~- middot-middot-------middotmiddotmiddot----middot------middot-middot~middot- - - ~-----1---middot

22middot ~~~~~~~-~ -94 -92 -90 -88 -84deg-86 -82 -somiddot

c

-

Platform Selection bullbullbull1111111111111

0 BasisCriteria for Selection middot

0 Review of Phase I Platforms middot 0 Alternative Platforms

0 PMB Recommendations

0 Participants Comments

gt

L

l I

BasisCriteria for Selection bullbullbull1111111111111

0 Benefit for Calibration - Classification (UnexpectedExpected)

- Failure Mode (StructuresFoundations)

- Damaged StructuresKnown response

- Better Representation of Full ~opulation

0 Availability of Data - Existing Models

- Inspection Data

- Damage Data

PREDICTED BEHAVIOR

-

OBSERVED

BEHAVIOR

-SURVIVED DAMAGED FAILED

SURVIVED

ST134W WD90A

-MC311 -

MC397 -

--

-

-

ST151K ST130Q

-

-

-

DAMAGED - -

-

-

--

-

-

ST52 SS139 ST177B -

ST161A

-_

-

FAILED

-

-

-

ST151H ST130A ST72

Operator - Platfonns - Platfonns Survived Damagedamp

- Falled -

bull

-

-

Jle eBDIGIPllDbl -Amocoshy 30 1 Chevron -

143 12 Exxon 72 Mobll 22 1 Phllllps 7 bull

-Shell 39 Unocal 25 1

Total Partlclnts Platfonns 338 15

Other Operators bull- 310 13

-

Total Platfonns - -

648 28

fbml

Platforms Considered 6 7

tie of Total Platfonns 1 25

OVERVIEW OF PLATFORMS SURVIVEDDAMAGEDFAILED

-

shy

v

Platforms Evaluated in Phase I

-

-Platfonn Water Year Number Analysis Avallabillty of Availability Platform

- Name Depth rl of Model DetaDed rl ClassUlcatlon

Installation legs Inspection Soll Report Based on Information Phase I Calibration - -ft

- -- Suaill flalfnrm ~BSi -STISIK 137 1963 8 CAP Yes Yes Un eel Sarvlval

-STIJOO 170 1964 4 CAPmiddotshy Yes ST134data u ST134W 137 1981 4 CAP Yes Yesmiddot E

-

edSarvlval -

middotsurvival

WD90A 184 - 1964 8 CAP Yes E -

- Sarvlvill

MC3ll 343 1978 8 KARMA

MC397 468 1991 4 KARMA -

bull ~reglmmiddot- -

Yes Sure Survival

bull Yes - Sare Sanlval-

-T23STS2 63 1969 4 CAP

T2SSS139 62 1969 4 CAP -

Yes

Yes

Yes Ex

Yes Une

bull SurvivalLikely to DamalI

middot bull SanlvalLlkely to DamalI

ST161A 118 1964 8 USFOS

Eail11a pound1al(bullllm Clss ST177B 142 --1965 8 CAP

STISIH 137 1964 - 8 CAP C

STIJOA 140 19S8 8 CAP - -

Yes

Yes

ollamed

Yes

- Yes E

-

at about 1 mUe

Yes

ST134data

bull SafvlvalLlkelv to Dam- bull- -

E middotrahre

Emeried Failure -

Likely FaUare

T21ST72 61 1969 4 CAP Collamed No Likely FaUure

Platforms Identified for Evaluating Their Likelihood of Falling Under UnexpeCted Survival Category

CllEllATOa

AllB

llLOCll

Sl1lvcnJllE

NAME

YEAR

INSTAUJCD

WATER

IJBPJB

middotmiddotDISTANCE

ftOMSllOllE

(MIUIS)

Ill

(M)

AMOCO BllW

STllI

WDI075

A

bull D

4

n uo 17J

44

32

bull 20

10

bull CllEVllOH STtllO c I 1bull 21 11

STl130 D 0 161 21 11 STllJ4 I 0 137 D - 11

STIUI G I 137 32 11

STIUI I 0 121 32 11

STIUI

J 0 140 32 11

STIUI It 4 137 32 11

STllSI PROl)I 0 137 32 11

STllSI PROD-2 0 137 32 11

STll7 A 0

140 35 11

WDl117 c 65 214 34 10

DXON STshy E 51 14 bull STl165 A t3 bull 34 WD0073

WD0073

WD0073

A _ c

4

62

4

IA IA

17J

22

22

22

bullbullbull WD003-shy E 65 161 Z7

A 0 IN 31 MOBD Bl12I

ssn A bull

50

50

31

7

bull 7

MIJllPllY SS114 B 7 50 7 7

PlllUIPS SSl14t A 55 33 7

SAMEDAN ST17J A 4 117 7 bull 7

SJIELL WD0103 A 65 223 27 WDllOJ bull 65 221 27 WDll04 c 65 221 27 WDOI04 D 67 2 27

UNOCAL

SSl20I

SSl20I

bull SSl20I

SS0209

SS0209

SS0209

SS0209

SS 0253

SS 0253

E

I

PtJllE

A

bull D

G A

c

0

4

4

7J

61

0

0

69

103

97

97

9S

100

95

100

165

175

34

34

34

34

34

34

34

54

54

bullbullbullbullbullbullbullbullbull

Platforms Recommended for Phase II

llalr y_ w- Aot Av- _ I 1 Doplla Model nm bull bull _ bull I I H11 s18-1 Irht lbwl~ bullbullbull(-illtdlel OM lllMtD

II ---

-STISIIC 137 1963 I CAP Yeo Yeo I Yeo

STtshy -170 1964 4 CAP Yeo ST134dlU I Yeo

STl34W 137 Yeo Yeo -1981 4 CAP D Yeo

WD90A 114 1964 I CAP Yeo - D

MC311 30 1971 I KARMA Yeo -sre-

MC397 middot 461 1991 4 ICARMA Yeo Sireshy - - r- -

__ T13ST5l 63 1969 4 CAP Yeo Yeobull middot- ~n- D Yeo _ Tl5SSl39 62 1969 4 CAP Yeo Yeo I Yeo

_ STIT711 142 1965 I CAP Yeo I Yeo middot-1shySTl61A Ill 1964 I USFOS Yeo Yeo - D bulloshy

- r-

STISIH 137 1964 I CAP y D y -_ STl31A 140 1951 I CAP Yeo STl34dlU I Yeo

T21Sll 61 1969 4 CAP No rar-middot I Yeo

~-rmiddot-middot SS209A 95 1967 16 CAP Yeo D -SSU4R

11 UI A 41 -in D SS 2741AIM CAP bull

_ _ -shy - -shy- -shy

CALIBRATION

bull PHASEIAPPROACH

bull CALIBRATION ISSUES

bull PHASE II WORK SCOPE

bull SOFlWARE AVAILABILITY

bull EFFECT OF ADDITIONAL PLATFORMS

bull WEIGHTING APPROACH

bull EXAMPLE APPLICATION middot

(l=Ofce Mode] Wind

~

fcurrent l -middot

~SW ~leldl4

ultimate resistance force I first

design load Illgt

i _

deflection

-

bull

4

Historical Hurrlcane Dariiage and Survival of Platforms

Components Required for Calibration (Petrauskas 1992)

PRIOR Distribution

PLATFORM EXPERIENCE

BAYESIAN Survivals and Failures

UPDATING

Waves Winds and Currents bull

POSTERIOR Distribution

- A_middotJA ~ Survival Data Failure Data

Result Result

Load Resistance

middotfl

-

Bayesian Updaing middotGeneral Description (Petrauskas 1992)

L Imiddot

WHERE TO INTRODUCE BIAS FACTOR (B

On Resistance (Rmue - B (RPREDICreD I

middotOn Loads (Smue middot - B (SPRE01cTEo

On Safety Margin (RSmue - B (RSPREDICTED

CALIBRATION METHODOLOGY

gt

r

FIELD OBSERVATIONS

Observed Behavior

survivals Damages

Failures

r

bull

CAPACITY ANALYSIS

Predicted Load and Resistance Predicted Failure Modes

Survival Damage Failure cases

RELIABILITY ANALYSIS

- Probability of Failure ~

Ukellhood Failure

BAYESIAN UPDATING

Prior Distnbution of B Combined Likelihood Function

Posterior Dlstnbution of B

BIAS FACTOR

Mean Value cov bull

Represents Modeling Uncertainties

Prior Distribution of B Likelihood Functionmiddot Observing Success case STISIK and Posterior Distribution of B

2 00

uo

160

140

120

e 100 L $ uoIshy

0 60 ~ ~ 040

020

0 00

0 00 0 50

_ I 00

b

I 50 2 00

-shy lk [bl STISIK)

--shy Pnor-f(b)

--shy Pos1enor r(bl ISIK)

Prior or B Mean bull 10 COV bull 030

Posterlor of Bbull Meanbull 136COV 0 16

2 50

Likelihood Function and Posterior Distribution Example Platform STISIK

- -

~

- Prior and Posterior Distributions or B Failure cases (ST177B STlSlH ST130A T21) Damage cases (T23 T25 ST161A) Success cases (STlSlK ST130Q ST134W WD90A MC311 MC397)

400

350

300

2 so 0-

I

bull bull

bull bull gt

- bullbull ~ bull bull I bull

_ bull bull bullbull I bull bull bullII V -

bull bull

I bull ~ bull

bullbull bullI bull bullI bull bull bull

middot -~

bull__ ~ middotI middotmiddot

L 2 00

e ~ I SO

-I 00

- 0 so

000 gt

0 00 020 0 40 060 0 80 I 00 120 140 160 110 200

b -

-

shy

Prior-r

- - - - Posterior fLlll failure cases

bull bull bull bull bull bull bull Posterior flall damage cases

- bull bull - bull middot Postenor flall success cases

Posterior flall success fulure damage cases

-

Posterior of B Mean= 119 COV 0101

Posterior Distribution or Blas Factor (8) bull Different Categories and All Cases Combined

shy

l

PHASE I - KEY CONCLUSIONS

bull GLOBAL CORRECTION FACTOR

bull 15 TO 20 CONSERVATISM MEAN)

bull UNEXPECTED SURVIVALS HAVE GREATEST IMPACT

bull DAMAGED CASES MORE INFORMATIVE

bull CALIBRATION ISSUES PHASE II

bull

EFFECT OF IMPROVED CAPACITY ANALYSIS RECIPE

bull APPLICABILITY TO MULTIPLE FAILURE MODES

bull NO FOUNDATION DAMAGE CASE INCLUDED

bull IMPROVED REPRESENTATION OF FULL ANDREW DISTRIBUTION

-bull NUMBER OF PLATFORMS CONSIDERED

I 1 I

I t ) bull ~

PHASE II WORK SCOPE (CALIBRATION)

bull

bull ESTABLISH MULTIPLE BIAS FACTORS

Use Nine Base Case Platforms

middot - Capacity Analysis for Revised recipe

Foundation Blas Factor from Foundation JIP

bull DEVELOP WEIGHTING APPROACH

bull UPDATE MULTIPLE BIAS FACTORS

bullEffect of Platforms Not In the Nine Base Cases

bull EXAMPLE APPLICATION OF BIAS FACTORS I

bullbull

PHASE I PHASE II

CAPACITYbull ANALYSISRu bull

---------------------1----middot

Jc1 - iJbullraquoJ dbull

FttlS frt[f~l ratdhJJ

-

LIKELIHOOD FUNCTION

DEVELOPMENT

lk Jb Jfall1N) bull Jb)

111 Ill Isurvtvoq bull 1 - bl

-middot -------------------shy ~---------------------middot-middot

r111ir8 Cb1 111 Cbl BAYESIAN

UPDATING

~cove

R1R2 bull middot---------------------~----

Jblbull P(g cOJ

1 bull JbRI (MBSI I bull 1

~middot

llcJllJfalluro) bull P(gcOJ Ill 111 b21fallun) bull P ((g1 CO) U 1112 c OD

middot--------------------shy ----------------------shy bull

r111111r8 Cbl 111 Ill i-1nro1

deg ~1b2)laquottfrr (b1b21-lnfo) re Ill bull 1-11middot b2) bull lodegamp

middot

i1 covamp i2 cov~i

-

shy

--

Likelihood Function vs b All Success Case STISI K STl30Q STl34W WD90A MC311 MC397

I 0000

0 9000

OIOOO

0 7000

~ 0amp000

ta osoOO

~ 0 4000

0 3000

0 2000

0 1000

bull 0 0000

bull

I

middot

middot bull I

Ibull

-sibull ~ -middot-middot-middot-middot -middot-

----7----+-----~ I

Ibull bull bull I

rbull

--- Jk ISIK[blsuccess)

- bull - bull - lk 130Q[blsuccess) bull

lk 134W[blsuccess)

Jk-WD90A(blsuccc~I

- - - lk MC397[blsucccss)

--ltgt-- lk middotMC31 l[blsucccss)

--- Jk [bl 6 out of6succcsHascs)

05 I 5 2 250

b

Likelihood Functionsmiddot Success Cases - Phbullsc L

shy

bullbull bull bull bull bull bull -------------------~

bull bull bullbull bullbull bullbull bullbull bullbull bullbull bull

bull bullbull

bullbullbullbull

bullbull

bullbullbullbullbullbullbull

bullbullbullbullbullbull

bullbullbullbullbullbullbull

z 0

t z ) LL

c 0 0c i ~ i

J

~ sect ~--------------~ e l5

I

I

SOFTWARE

bull PF PROGRAM

bull CALREL

bull PF +CALREL

bull RELACS

bull

gt i

l

SENSITIVITY STUDIES

POSTERIOR DISTRIBUTION OF B

bull EFFECT OF CHANGE IN DISTRIBUTION OF PRIOR OF B

bull EFFECT OF REMOVAL OF ONE PLATFORM FROM EACH CATEGORY

bull EFFECT OF REMOVAL OF THREE DAMAGE CASES

bull EFFECT OF ADDITIONAL 13 SIMILAR PLATFORMS

-I

Prior and Posterior Distributions of B Variation In COV of PriorofB fromOl to04 -- -

4500

-4000

-

- -

3500 --middot - bull bull - bull bull Prior-r ICOV =02

3000

- - - - Prior-r ICOV =03 -

bull -~ e 2soo -

bull bull bull bull bull bull bull Pnor-r ICOV =o 4t frJ ~ 2000 Posterior rlCOV =0 2

bull bull Postenor rlCOV =0 3 bull bull 1500 bull middot bull - shy -middot- Posterior rlCOV =04

middot~ I 000 - bull

-middotmiddot ~ ~middotmiddot M -~

- middot 0 500

wbullbullbull _ I middot~ middotmiddotmiddotmiddotmiddot middotmiddotmiddotmiddotmiddot - -~ ~ __ middot_ _ bull -shy0 000

0 000 0200 0 400 0 600 0 800 IOOO 1200 1400 1600 1800 2000 -

- - b

-

=shy

Posterior Distribution of Bias Factor (B) bull Effect of Variation In COV of Prior of B

bull

bullPrior and Posterior Distributions of B Evaluating Effect of Removal of One Platform from Each Category from the Base Case of 13 Platforms

400

~ ~ 350

j IA i

3 00 bull I bull I ---Prior-r I bull l

250 1 I 1

I bull bull ---Posterior r lall 13 casesshy- ~E 2 00 bull bull bull bull bull bull bull Posterior r without STISIK

- ~ 1 bull - bull I - bull bull - bull bull Posterior r Iwithout T2I I 50 I

J-tt~i - bull - bull - Posterior r Iwithout T2S

tOO [-

I I ~ ~ I 0 50

0 00

___ r J i

l~T

~--

0 00 0 20 0 40 0 60 080 I 00 120 140 160 110 2 00

b

zshy

Posterior Distribution of Bias Factor (8) bull Effect of Removal of middotPlatform from Any Category

bullPrior and Posterior Distributions or B Evaluating Effect or Removal or 3 Damage Cases from the Base Case or 13 Platrorms

4

35

3

25

e L 2 ~ bulle shy

I 15

bull I

05

0

0 02 04 06 I 2 I 4 16 IS 2

I I

I

I

I Ibull

J -tmiddot L--

bullI

~

I

fl

II ~

I I v I

middot bull bull

bull L

~ ll- I-~lt

OS I

b

I

bull

1middot bullbullbullbull

~ __

bull

- - - - Prior-r

- - - - Postenor flall failure cases

Postenor flall success cases

--- Posterior flall 13 cases

---Posterior fI success failure cases only

IPosterior or B Mean =t11 COV =0121

-

_I

Posterir Distribution or Bias Factor (B) bull Effect or Removal or 3-Damage Cases

-

bullPrior and Posterior Distributions or B Evaluating Effect or Additional 13 Similar Observations as

Used in Andrew JIP The POsterior or B from Andrew-JIP Considered as New Prior or B

- 500

450

400

350

bull bull

bull bull bull bull J bull

bull bull f ~ I

I ~ Ibull bull

bullbull bull

l bull I -bullbull bull bull

I

I bull bull

bull I I~~bull bull bull

bull ~

I bull bull

~ ~_ ~~~

- - New Prior r bull

Posterior 114 new failure cases ~ 300- bull bull bull bull bull bull bull Posterior f 13 new damage casesl 150 ~

- bull bull - bull bull Posterior 116 new success cases i 2 00

Postenor 1113 new success failure -150 damage cases

I 00

0 50 -

0 00 0 00 0 20 0 40 0 60 0 80 I 00 120 I 40 160 110 2 00

Posterior ol B Mean= 120 COV= 0071 b

shy

shy

_rshy

Posterior Distribution of Bias Factor (B) bull Effect of Additional 13-Similar Observations

1

1 I

WEIGHTING APPROACH bull

bull ASSESSMENT OF PLATFORMS DATA

Select Platforms Most Mappable to the Analyzed Platforms

Unlikely SurvivalDamage Cases Identification

Based on Cut-Off Criteria Agemiddot Hs etc

bull INFORMATION REQUIRED FOR COMPARISON

bull CATEGORIZATION OF ADDITIONALmiddot PLATFORMS

bull UPDATING OF CALIBRATION RESULTS

--

OVERVIEW OF PLATFORMS SURVIVEDDAMAGED -~

Operator -

- Platforms

Survived CS)

Platforms

Damaged CD)

Platforms Used In

Phase IJIP -

JI~ ~Bill~IRBDll -

Amoco -

Chevron -

Exxon

Mob II

Phllllps

Shell

Unocal -

30

143

72 22

7

39 25shy -

1

12

bull

1 bull

1

1(S)+1 (D)

3 (S) +3(D)

2 (S)

bull shy

bull

bull

bull

Total Partlclnants Platforms 338 - 15 10C6S+4D) -

Other Operators

310

13

3 (D)

Total Platforms 648 28 13C6S+7D) -

Summary of Andrew JIP bull Phase I Platforms

Plairshy w v- N-ber Lea-le 1 shy car Hind_ o-w N-bull Deplh fl fl IJpe la bull n Yeadon Jadtet lbelledtf liefalit -- _ -OD

IL H - -shy _ICllshy IL IL - - rQClto ~ _

K STUIK 137 1963 8 365 608S4 Cellardeck

-middot STl-shy 170 1964 4 -~ K 39 62269 Cellardeck - --ST134W 137 1911 4 43 608S4 Smvlnlshy- _ - -

WD90A 184 1964 8 K 267Sbull34 83 SOS31 SUboftllar deck SmvMI

MC311 343 1978 8 x S7 62 - MC397 468 1991 4 x shy S33bull6SO 65911 -shySmeSmvlvol -- gt -

r-- T23ST52 63 1969 4 ton- shyKshy 52594 S03 -T25SSl39 62 1969 4 Kshy SOS94 S063 - middot ton-shy__ -

bullbull a ton--shyST161A 118 1964 8 K - 347S 51495 Cellarclecti

r~

ST1771 142 1965 8 K 39 60167 CeDordeck failme

STISIH 137 1964 8 K 395 608S4 CeDardeck Failme

K ST130A 140 1958 a SS 6095 I lblv Pai1uno - T21ST72 61 1969 4 Kshy S4S94 4974 I lhlv Pai1uno -

- -

JE

INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON

bull WATER DEPTH

bull bull YEAR OF INSTALLATIONDESIGN

bull PLATFORM ORIENTATION

bull PLATFORM LATITUDE AND LONGITUDE

bull KEY SKETCHES OF PLATFORM STRUCTURAL FRAMING

Number of Legs Plies Leg Batter Number of Conductors bull Deck Elevations (Including sub-cellar deck If any) Jacket Framing Elevations Brace type In Vertical Frames Sizes of Primary Members (leg pile vertical frames braces primary horizontals) Joint Details (can gap etc) LegPiie Annulus Grouting Status

bull middotPOST ANDREW INSPECTION REPORT Significant Wave Height Hs (ft)

bull INFORMATION ON DAMAGES PRIOR TO ANDREW

bull RESULTS FROM IN-HOUSE STUDIES IF AVAILABLE

bull ~RIGINAL DESIGN CRITERIA INFORMATION IF AVAILABLE

bull SOIL SHEAR STRENGTH PROFILE IF AVAILABLE

Platforms Identified for Evaluating Therr Likelihood of Falling Under Unexpected Survival Category

OPlRATOR

middotADA

BLOClt

llmucnJllE

NAME

YEAll

INSTAUJD

WA1ER

DBPTll

DISTANCE

JlllOM SHOllE

(MILU)

Bl

(M)

AMOCO 111113

STl11

WDI075

A

bull D

St

17

120

1n

44

20

10 I

CllllVllQfl STlllO c 61 110 21 11 STIUO D Q 1110 21 11

STl134 p Q 137 29 11

ST0151 G 61 137 11

ST0151 I 3 121 11

ST0151 J Q 140 11

STl151 It 137 11

ST0151 PJl001 Q UT 11

ST0151 PROD-2 Q UT 11

ST017 A 3 140 35 11

WDIUT c 65 214 34 10

llXXON STI055 II 51 6T 14 I

STl15 A t3 34

WDI073

WD0073

Amiddotshy Q

1111 1111

22

22

I

I

WDI073 c 1n 22 I

- WDOCl93-shy II

65 1110 2T A 3 194 31 MOBU 111121

SS IOl7

A B

501

50

50

31

T

I

T 6

MtlllPllY SS114 B T 50 T T

PlllllIPS SS0149 A 5 55 33 T

SAMllDAN STln A 107 T T

SBllLL WDl103 A 65 223 2T WDl103 B 65 221 2T WD0104 c 65 221 2T WD0104 D 6T 20 2T

UNOCAL SS020I

SS 020I SSl20I

SS 0209

SS0209

II p

PURll

A

B

3

n 61

103

97

97

95

100

34

34

34

34 34

bullbull bullbullbull

- SS0209 D u 95 34 I

- SS 1209 SSl253

SS0253

G A

c

6T

2

100

165

1T5

34

54

54

bullbull I

I I

I I

SOUTH MARCH

~LAND

N

SOUTH PaTOI ENE ISWltD AREA

MMS Defined Limits ror Offshore Structures Requiring Inspections Following Andrew

t---Jbulln3 T25bull I bullTtl

bull1$1Htl1Abull bull1SIK1779bull

~

middot-DaTA AREA~T

CRANO rsLE

AREA bull1300

bullll4W

=shy

bullPath of Hurricane Andrew with Platforms Used In Calibration

Likelihood Function vs b All Success Case STISI K STIJOQ ST134W WD90A MCJll MCJ97

I 0000

0 9000

0 8000

07000

-Bo6000

tiiii 0 5000 -

04000

0 3000

0 2000

01000

0 0000

0 05 25

ri1rt middot-middot-middot-middot-middot-

I 5 2

b

---11 middot151K(blsu~ccsI

- bull - bull - fk 130Q(bfU~lCSS)

lie bull134W(blsuccc]

lk-WD90A[blsultCS]

- - - lk middotMC397(bfUtlCS]

--o-- lk-MCJll(bbuctcsJ

--- ll (bl 6 out of 6 Mlttcs case]

_

( Likelihood Functions middotSuccess Cases - p - 1 c I

CATEGORIZATION OF ADDITIONAL PLATFORMS

I I I

bull CATEGORIES

Unexpected Survivals

Expected Survivals

bull APPROACH

Identify Corresponding Platform(s) Evaluated In the JIP

Determine Max Wave Height at Platform Location

Evaluate Whether Wave Hit the Deck

middot - Evaluate Likelihood of R and S to Differ from the Base Cases

Review Joint Information

Assess Special Features

bull IDENTIFY NUMBER OF PLATFORMS IN EACH CATEGORY

I

UPDATING OF CALIBRATION RESULTS

bull INCLUDE EFFECT OF ADDITIONAL PLATFORMS ON bullBbull

Consider Increased 1Uncertalntles In R and S

Develop bullNeW Ukellhood Functions for Different Categories

middotLet bullxbull Number of Ukely Unexpected Survival Cases and y Number of Ukely Expected Survival Cases

Determine Combined Likelihood Function lk (b Iadditional platforms) middot

bull REVISED LIKELIHOOD FUNCTION

lk(b I all cases) = lk(b I 9 platforms) middot x lk(b Ix platforms) x lk(b I y platforms)

bull ESTABLISH REVISED BIAS FACTOR B

f8 (b) u f8 (b) lk (b I all cases)

II

t t ~

shy

APPLICATIONS OF BIAS FACTORS

ii APPLICABLE TO PARTICULAR LOAD AND RESISTANCE RECIPE

bull USE BY THE API COMMITTEES AND MMS

Criteria Revisions

Recommended Design Guidelines Revisions bull

bullbull REASSESSMENT STUDIES AND DECISION MAKING

middot

bull RISK ASSESSMENT STUDIES BY THE OPERATORS

Individual Platforms

Fleet of Platforms

middot IMR Program Implementation

bullbull NOT TO BE USED IN DESIGN

I I fl J

EXAMPLE USE OF BIAS FACTORS

bull PHASE I EXAMPLE APPLICATION

bull EVALUATION OF ANNUAL PROBABILITY OF FAILURE

Establish Annual Maximum Seastate Data

Determine Coefficients to Define Base Shear

Determine Ultimate Capacity of Platform

Determine Probablllty of Failure Versus bullbbull

-bull DETERMINE UPDATED PROBABILITY OF FAILURE

Pf - P [S gt R]

Pf I P [~ gt Rb I b] f 8 (b) db

Pr - J P [S gt Rb I b] f bull 8 (b) db

bull TOTAL PROBABILITIES OF FAILURE

Total Pf - 1 bull Il [1 middotPf] - l Pf bull

Total Pf - 1 bull Il [1middot Pf] - l Pfmiddot

Total Pr = 1 - n [1middot PfbullJ = l Pfshy

I I I

EXAMPLE APPLICATION bull PHASE I

PLATFORM ST 151 K

bull GULF OF MEXICO

bull WATER DEPTH = 137 FT

bull ANNUAL MAXIMUM HS MEDIAN 175 ft COV 031

bull ASSUMED STORM DURATION 3 hoursmiddot

bull ASSUMED NUMBER OF WAVES 800

bull MEDIAN HS FOR STORM HOURS 1 2 AND 3

bull MEDIAN CURRENT VALUE bull

bull DETERMINE ULTIMATE CAPACITY FOR THREE DIRECTIONS

bull DETERMINE ANNUAL FAILURE PROBABILITIES

Broadsidemiddot Diagonal End-on

0037 0025 0007

0071 0044 0018

0022 0016 0004

DETERMINE TOTAL ANNUAL PROBABILITY OF FAILURE

Pf - 007

Pf - 013

Pr - 004

bull

EXAMPLE USE OF BIAS FACTORS - PHASE II middot

bull DETERMINE PREDICTIVE PROBABILITY OF FAILURE

Case - 1

Pf _ S S Pf (b b2) fa1a2 (b b2) db db2

g bull -1 (b1 R I S) - 1

92 - (b2 R2 I S) - 1

Case -2

Pf s ~ [bJ f11 (b) db

Pf [bJ p [g lt 0)

g - bull 1 (b1 R1 I S) - 1

1

bull1) l

I

BENEFITS OF ANDREW JIP bull PHASE II

lt

bull IMPROVED LOAD AND RESISTANCE RECIPE

bull IMPROVED CALIBRATION PROCEDURE

bull MULTIPLE BIAS FACTORS

bull PROCEDURE TO INCLUDE EFFECT OF OTHER PLATFORMS

bull PROCEDURE TO USE MULTIPLE BIAS FACTORS

  • Structure Bookmarks
    • ANDREW PHASE II JOINT1 INDUSTRY PROJECT
    • PROGRESS MEETING NUMBER 1
    • ANDREW PHASE II JOINT INDUSTRY PROJECT PROGRESS MEETING NUMBER 1 NOVEMBER 16 AGENDA
    • ANDREW PHASE II UST OF PARTICIPANTS
    • ANDREW PHASE I OBJECTIVES AND RESULTS
    • ANDREW PHASE II PROJECT OBJECTIVES
    • SUBSEQUENT PROJECTS
    • ANDREW PHASE II PROJECT TEAM
    • ANDREW PHASE II UPDATE MODELING middotAND ANALYSIS RECIPE
    • -ANDREW PHASE II CONDUCTOR RESISTANCE
    • ANDREW PHASE Il WA VE FORCE PROFILE
    • ANDREW PHASE II WA VE FORCE PROFILEmiddot
    • ANDREW PHASE II BRACE MODELING
    • ANDREW PHASE II FOUNDATIONS
    • ANDREW PHASE II middot middot JOINT MODELING
    • ANDREW PHASE Il JOINT MODELING OPTIONS DISCUSSION
    • ANDREW PHASE II JOINT MODEJJNG OPTIONS ILLUSTRATIONS
    • ANDREW PHASE II JOINT MODELING PHASE II METHOD
    • ANDREW PHASE II ANDREW STORM CONDmONS UPDATED
    • Platform Selection bullbullbull1111111111111
    • BasisCriteria for Selection bullbullbull1111111111111
    • CALIBRATION
    • WHERE TO INTRODUCE BIAS FACTOR (B
    • PHASE I -KEY CONCLUSIONS
    • bull CALIBRATION ISSUES PHASE II
    • PHASE II WORK SCOPE (CALIBRATION)
    • SOFTWARE
    • SENSITIVITY STUDIES POSTERIOR DISTRIBUTION OF B
    • WEIGHTING APPROACH bull
    • INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON
    • CATEGORIZATION OF ADDITIONAL PLATFORMS
    • UPDATING OF CALIBRATION RESULTS
    • APPLICATIONS OF BIAS FACTORS
    • EXAMPLE USE OF BIAS FACTORS
    • EXAMPLE APPLICATION bull PHASE I PLATFORM ST 151 K
    • EXAMPLE USE OF BIAS FACTORS -PHASE II middot
    • BENEFITS OF ANDREW JIP bull PHASE II
Page 33:  · . ' ANDREW PHASE II PROJECT OBJECTIVES . ' • ·UPDATE MODELING AND ANALYSIS RECIPE· JOINTS ' ' ·BRACES WAVE PROFILE FOUNDATIONS . ' ' CONDUCTORS· • ANALVZE PLATFORMS USING

II Ii

z

ctV 4x SPlO caissonfy-42s soil-api(lf-097step-114)

Inelastic Events Legend Elastic Strut Buckling

Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield

--------shyBeam Clmn Fully Plastic Fracture

~--~~~~~~~~~--~----~__i~~---~~~~__--~

Pro)ect apipile Model South Pelto 10 Version Wed May 18072235 1994 - -

1bullt

+

t

f i bullbullbull

bullbull bullbull

I I

5

~x SPlO Caissonfy=42s soil=api(lf=lOOstep=137)

Inelastic Events Legend

Elastic Strut Bucklilg

Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield

earn Clmn Full~ Plas1c Fracture

ANDREW PHASE II middot middot

JOINT MODELING

middotbull API RIGID-PLASTIC USED IN PHASE I

a ELASTIC-BRITTLE USED IN PHASE IB

amiddot MSL WILL PROVIDE NONLINEAR FORCE-DEFORMATION

AND MOMENT-DEFORMATION RELATIONS FOR PHASE II

a MODELING OPTIONS FOR PHASE II -

l

- UNCO~LEDmiddotONE DIMENSIONAL MEMBER END JOINT MODELING

J Ii

ANDREW PHASE Il

JOINT MODELING OPTIONS DISCUSSION

bull JD COUPLED INTERACTION SURFACE

- IDEAL

- DATA NOT AVAILABLE TO CALIBRATE

bull 2D COUPLED INTERACTION SURFACE

- SIGNIFICANT MODELING IMPROVEMENT

- DATAAVAILABLE

- IMPLEMENTATION PRACTICAL BUT NEEDS DEVELOPMENT AND CALIBRATION

bull 2D UNCOUPLED INTERACTION SURFACE

- SIGNIFICANT MODELING IMPROVEMENT

- DATA AVAILABLE

- IMPLEMENTATION PRACTICAL AND AVAILABLE

ANDREW PHASE II

JOINT MODEJJNG OPTIONS ILLUSTRATIONS

P

e

ANDREW PHASE II

JOINT MODELING PHASE II METHOD

e

I I I

ANDREW PHASE II

ANDREW STORM CONDmONS UPDATED

bull -PHASE I USED OCEANWEATHER HINDCAST-(Nov 1992)

bull middotmiddotPHASE II WILL USE THE REVISED HINDCAST (Nov 1994)

bull lt c lt I lt

2-2

l---4-~~~+-~~-+-~~~1--~~-+-~~--l--~T--+-~~--i24

f-tJVtMfiz~ 1Cfl (r STvb(

Plollol bullbull 5middotNOYmiddot9214bull1J Irbullbull Illbull [ANDREVl~NDREVCUSTbullMAXll 5-NOY-1992 10bull54 - 9208

HEIGHT lml

I~ MAXI

I L 1~middotamp SZ~- ~A I 13c

I J) ( r I X~0iWJ~0gtJ gt L gt gt I 1c I 120

-- N

~ deg

I I I I I I e C I ~=-41 122 -94 -92 -90 -88 -86 -84 -82 -80

-

Figure 7 contours of maximum hindcast significant wave height

_

~

~tl ~

NOltNttIJ 17lt ltNS S iJJ)y

Maximum Significant Wave Height - Hurricane Andrew 1992

30 ~

26 ~ I ~-~~~- == -shy -~ 35-_

24deg1--shy --------~- middot-middot-------middotmiddotmiddot----middot------middot-middot~middot- - - ~-----1---middot

22middot ~~~~~~~-~ -94 -92 -90 -88 -84deg-86 -82 -somiddot

c

-

Platform Selection bullbullbull1111111111111

0 BasisCriteria for Selection middot

0 Review of Phase I Platforms middot 0 Alternative Platforms

0 PMB Recommendations

0 Participants Comments

gt

L

l I

BasisCriteria for Selection bullbullbull1111111111111

0 Benefit for Calibration - Classification (UnexpectedExpected)

- Failure Mode (StructuresFoundations)

- Damaged StructuresKnown response

- Better Representation of Full ~opulation

0 Availability of Data - Existing Models

- Inspection Data

- Damage Data

PREDICTED BEHAVIOR

-

OBSERVED

BEHAVIOR

-SURVIVED DAMAGED FAILED

SURVIVED

ST134W WD90A

-MC311 -

MC397 -

--

-

-

ST151K ST130Q

-

-

-

DAMAGED - -

-

-

--

-

-

ST52 SS139 ST177B -

ST161A

-_

-

FAILED

-

-

-

ST151H ST130A ST72

Operator - Platfonns - Platfonns Survived Damagedamp

- Falled -

bull

-

-

Jle eBDIGIPllDbl -Amocoshy 30 1 Chevron -

143 12 Exxon 72 Mobll 22 1 Phllllps 7 bull

-Shell 39 Unocal 25 1

Total Partlclnts Platfonns 338 15

Other Operators bull- 310 13

-

Total Platfonns - -

648 28

fbml

Platforms Considered 6 7

tie of Total Platfonns 1 25

OVERVIEW OF PLATFORMS SURVIVEDDAMAGEDFAILED

-

shy

v

Platforms Evaluated in Phase I

-

-Platfonn Water Year Number Analysis Avallabillty of Availability Platform

- Name Depth rl of Model DetaDed rl ClassUlcatlon

Installation legs Inspection Soll Report Based on Information Phase I Calibration - -ft

- -- Suaill flalfnrm ~BSi -STISIK 137 1963 8 CAP Yes Yes Un eel Sarvlval

-STIJOO 170 1964 4 CAPmiddotshy Yes ST134data u ST134W 137 1981 4 CAP Yes Yesmiddot E

-

edSarvlval -

middotsurvival

WD90A 184 - 1964 8 CAP Yes E -

- Sarvlvill

MC3ll 343 1978 8 KARMA

MC397 468 1991 4 KARMA -

bull ~reglmmiddot- -

Yes Sure Survival

bull Yes - Sare Sanlval-

-T23STS2 63 1969 4 CAP

T2SSS139 62 1969 4 CAP -

Yes

Yes

Yes Ex

Yes Une

bull SurvivalLikely to DamalI

middot bull SanlvalLlkely to DamalI

ST161A 118 1964 8 USFOS

Eail11a pound1al(bullllm Clss ST177B 142 --1965 8 CAP

STISIH 137 1964 - 8 CAP C

STIJOA 140 19S8 8 CAP - -

Yes

Yes

ollamed

Yes

- Yes E

-

at about 1 mUe

Yes

ST134data

bull SafvlvalLlkelv to Dam- bull- -

E middotrahre

Emeried Failure -

Likely FaUare

T21ST72 61 1969 4 CAP Collamed No Likely FaUure

Platforms Identified for Evaluating Their Likelihood of Falling Under UnexpeCted Survival Category

CllEllATOa

AllB

llLOCll

Sl1lvcnJllE

NAME

YEAR

INSTAUJCD

WATER

IJBPJB

middotmiddotDISTANCE

ftOMSllOllE

(MIUIS)

Ill

(M)

AMOCO BllW

STllI

WDI075

A

bull D

4

n uo 17J

44

32

bull 20

10

bull CllEVllOH STtllO c I 1bull 21 11

STl130 D 0 161 21 11 STllJ4 I 0 137 D - 11

STIUI G I 137 32 11

STIUI I 0 121 32 11

STIUI

J 0 140 32 11

STIUI It 4 137 32 11

STllSI PROl)I 0 137 32 11

STllSI PROD-2 0 137 32 11

STll7 A 0

140 35 11

WDl117 c 65 214 34 10

DXON STshy E 51 14 bull STl165 A t3 bull 34 WD0073

WD0073

WD0073

A _ c

4

62

4

IA IA

17J

22

22

22

bullbullbull WD003-shy E 65 161 Z7

A 0 IN 31 MOBD Bl12I

ssn A bull

50

50

31

7

bull 7

MIJllPllY SS114 B 7 50 7 7

PlllUIPS SSl14t A 55 33 7

SAMEDAN ST17J A 4 117 7 bull 7

SJIELL WD0103 A 65 223 27 WDllOJ bull 65 221 27 WDll04 c 65 221 27 WDOI04 D 67 2 27

UNOCAL

SSl20I

SSl20I

bull SSl20I

SS0209

SS0209

SS0209

SS0209

SS 0253

SS 0253

E

I

PtJllE

A

bull D

G A

c

0

4

4

7J

61

0

0

69

103

97

97

9S

100

95

100

165

175

34

34

34

34

34

34

34

54

54

bullbullbullbullbullbullbullbullbull

Platforms Recommended for Phase II

llalr y_ w- Aot Av- _ I 1 Doplla Model nm bull bull _ bull I I H11 s18-1 Irht lbwl~ bullbullbull(-illtdlel OM lllMtD

II ---

-STISIIC 137 1963 I CAP Yeo Yeo I Yeo

STtshy -170 1964 4 CAP Yeo ST134dlU I Yeo

STl34W 137 Yeo Yeo -1981 4 CAP D Yeo

WD90A 114 1964 I CAP Yeo - D

MC311 30 1971 I KARMA Yeo -sre-

MC397 middot 461 1991 4 ICARMA Yeo Sireshy - - r- -

__ T13ST5l 63 1969 4 CAP Yeo Yeobull middot- ~n- D Yeo _ Tl5SSl39 62 1969 4 CAP Yeo Yeo I Yeo

_ STIT711 142 1965 I CAP Yeo I Yeo middot-1shySTl61A Ill 1964 I USFOS Yeo Yeo - D bulloshy

- r-

STISIH 137 1964 I CAP y D y -_ STl31A 140 1951 I CAP Yeo STl34dlU I Yeo

T21Sll 61 1969 4 CAP No rar-middot I Yeo

~-rmiddot-middot SS209A 95 1967 16 CAP Yeo D -SSU4R

11 UI A 41 -in D SS 2741AIM CAP bull

_ _ -shy - -shy- -shy

CALIBRATION

bull PHASEIAPPROACH

bull CALIBRATION ISSUES

bull PHASE II WORK SCOPE

bull SOFlWARE AVAILABILITY

bull EFFECT OF ADDITIONAL PLATFORMS

bull WEIGHTING APPROACH

bull EXAMPLE APPLICATION middot

(l=Ofce Mode] Wind

~

fcurrent l -middot

~SW ~leldl4

ultimate resistance force I first

design load Illgt

i _

deflection

-

bull

4

Historical Hurrlcane Dariiage and Survival of Platforms

Components Required for Calibration (Petrauskas 1992)

PRIOR Distribution

PLATFORM EXPERIENCE

BAYESIAN Survivals and Failures

UPDATING

Waves Winds and Currents bull

POSTERIOR Distribution

- A_middotJA ~ Survival Data Failure Data

Result Result

Load Resistance

middotfl

-

Bayesian Updaing middotGeneral Description (Petrauskas 1992)

L Imiddot

WHERE TO INTRODUCE BIAS FACTOR (B

On Resistance (Rmue - B (RPREDICreD I

middotOn Loads (Smue middot - B (SPRE01cTEo

On Safety Margin (RSmue - B (RSPREDICTED

CALIBRATION METHODOLOGY

gt

r

FIELD OBSERVATIONS

Observed Behavior

survivals Damages

Failures

r

bull

CAPACITY ANALYSIS

Predicted Load and Resistance Predicted Failure Modes

Survival Damage Failure cases

RELIABILITY ANALYSIS

- Probability of Failure ~

Ukellhood Failure

BAYESIAN UPDATING

Prior Distnbution of B Combined Likelihood Function

Posterior Dlstnbution of B

BIAS FACTOR

Mean Value cov bull

Represents Modeling Uncertainties

Prior Distribution of B Likelihood Functionmiddot Observing Success case STISIK and Posterior Distribution of B

2 00

uo

160

140

120

e 100 L $ uoIshy

0 60 ~ ~ 040

020

0 00

0 00 0 50

_ I 00

b

I 50 2 00

-shy lk [bl STISIK)

--shy Pnor-f(b)

--shy Pos1enor r(bl ISIK)

Prior or B Mean bull 10 COV bull 030

Posterlor of Bbull Meanbull 136COV 0 16

2 50

Likelihood Function and Posterior Distribution Example Platform STISIK

- -

~

- Prior and Posterior Distributions or B Failure cases (ST177B STlSlH ST130A T21) Damage cases (T23 T25 ST161A) Success cases (STlSlK ST130Q ST134W WD90A MC311 MC397)

400

350

300

2 so 0-

I

bull bull

bull bull gt

- bullbull ~ bull bull I bull

_ bull bull bullbull I bull bull bullII V -

bull bull

I bull ~ bull

bullbull bullI bull bullI bull bull bull

middot -~

bull__ ~ middotI middotmiddot

L 2 00

e ~ I SO

-I 00

- 0 so

000 gt

0 00 020 0 40 060 0 80 I 00 120 140 160 110 200

b -

-

shy

Prior-r

- - - - Posterior fLlll failure cases

bull bull bull bull bull bull bull Posterior flall damage cases

- bull bull - bull middot Postenor flall success cases

Posterior flall success fulure damage cases

-

Posterior of B Mean= 119 COV 0101

Posterior Distribution or Blas Factor (8) bull Different Categories and All Cases Combined

shy

l

PHASE I - KEY CONCLUSIONS

bull GLOBAL CORRECTION FACTOR

bull 15 TO 20 CONSERVATISM MEAN)

bull UNEXPECTED SURVIVALS HAVE GREATEST IMPACT

bull DAMAGED CASES MORE INFORMATIVE

bull CALIBRATION ISSUES PHASE II

bull

EFFECT OF IMPROVED CAPACITY ANALYSIS RECIPE

bull APPLICABILITY TO MULTIPLE FAILURE MODES

bull NO FOUNDATION DAMAGE CASE INCLUDED

bull IMPROVED REPRESENTATION OF FULL ANDREW DISTRIBUTION

-bull NUMBER OF PLATFORMS CONSIDERED

I 1 I

I t ) bull ~

PHASE II WORK SCOPE (CALIBRATION)

bull

bull ESTABLISH MULTIPLE BIAS FACTORS

Use Nine Base Case Platforms

middot - Capacity Analysis for Revised recipe

Foundation Blas Factor from Foundation JIP

bull DEVELOP WEIGHTING APPROACH

bull UPDATE MULTIPLE BIAS FACTORS

bullEffect of Platforms Not In the Nine Base Cases

bull EXAMPLE APPLICATION OF BIAS FACTORS I

bullbull

PHASE I PHASE II

CAPACITYbull ANALYSISRu bull

---------------------1----middot

Jc1 - iJbullraquoJ dbull

FttlS frt[f~l ratdhJJ

-

LIKELIHOOD FUNCTION

DEVELOPMENT

lk Jb Jfall1N) bull Jb)

111 Ill Isurvtvoq bull 1 - bl

-middot -------------------shy ~---------------------middot-middot

r111ir8 Cb1 111 Cbl BAYESIAN

UPDATING

~cove

R1R2 bull middot---------------------~----

Jblbull P(g cOJ

1 bull JbRI (MBSI I bull 1

~middot

llcJllJfalluro) bull P(gcOJ Ill 111 b21fallun) bull P ((g1 CO) U 1112 c OD

middot--------------------shy ----------------------shy bull

r111111r8 Cbl 111 Ill i-1nro1

deg ~1b2)laquottfrr (b1b21-lnfo) re Ill bull 1-11middot b2) bull lodegamp

middot

i1 covamp i2 cov~i

-

shy

--

Likelihood Function vs b All Success Case STISI K STl30Q STl34W WD90A MC311 MC397

I 0000

0 9000

OIOOO

0 7000

~ 0amp000

ta osoOO

~ 0 4000

0 3000

0 2000

0 1000

bull 0 0000

bull

I

middot

middot bull I

Ibull

-sibull ~ -middot-middot-middot-middot -middot-

----7----+-----~ I

Ibull bull bull I

rbull

--- Jk ISIK[blsuccess)

- bull - bull - lk 130Q[blsuccess) bull

lk 134W[blsuccess)

Jk-WD90A(blsuccc~I

- - - lk MC397[blsucccss)

--ltgt-- lk middotMC31 l[blsucccss)

--- Jk [bl 6 out of6succcsHascs)

05 I 5 2 250

b

Likelihood Functionsmiddot Success Cases - Phbullsc L

shy

bullbull bull bull bull bull bull -------------------~

bull bull bullbull bullbull bullbull bullbull bullbull bullbull bull

bull bullbull

bullbullbullbull

bullbull

bullbullbullbullbullbullbull

bullbullbullbullbullbull

bullbullbullbullbullbullbull

z 0

t z ) LL

c 0 0c i ~ i

J

~ sect ~--------------~ e l5

I

I

SOFTWARE

bull PF PROGRAM

bull CALREL

bull PF +CALREL

bull RELACS

bull

gt i

l

SENSITIVITY STUDIES

POSTERIOR DISTRIBUTION OF B

bull EFFECT OF CHANGE IN DISTRIBUTION OF PRIOR OF B

bull EFFECT OF REMOVAL OF ONE PLATFORM FROM EACH CATEGORY

bull EFFECT OF REMOVAL OF THREE DAMAGE CASES

bull EFFECT OF ADDITIONAL 13 SIMILAR PLATFORMS

-I

Prior and Posterior Distributions of B Variation In COV of PriorofB fromOl to04 -- -

4500

-4000

-

- -

3500 --middot - bull bull - bull bull Prior-r ICOV =02

3000

- - - - Prior-r ICOV =03 -

bull -~ e 2soo -

bull bull bull bull bull bull bull Pnor-r ICOV =o 4t frJ ~ 2000 Posterior rlCOV =0 2

bull bull Postenor rlCOV =0 3 bull bull 1500 bull middot bull - shy -middot- Posterior rlCOV =04

middot~ I 000 - bull

-middotmiddot ~ ~middotmiddot M -~

- middot 0 500

wbullbullbull _ I middot~ middotmiddotmiddotmiddotmiddot middotmiddotmiddotmiddotmiddot - -~ ~ __ middot_ _ bull -shy0 000

0 000 0200 0 400 0 600 0 800 IOOO 1200 1400 1600 1800 2000 -

- - b

-

=shy

Posterior Distribution of Bias Factor (B) bull Effect of Variation In COV of Prior of B

bull

bullPrior and Posterior Distributions of B Evaluating Effect of Removal of One Platform from Each Category from the Base Case of 13 Platforms

400

~ ~ 350

j IA i

3 00 bull I bull I ---Prior-r I bull l

250 1 I 1

I bull bull ---Posterior r lall 13 casesshy- ~E 2 00 bull bull bull bull bull bull bull Posterior r without STISIK

- ~ 1 bull - bull I - bull bull - bull bull Posterior r Iwithout T2I I 50 I

J-tt~i - bull - bull - Posterior r Iwithout T2S

tOO [-

I I ~ ~ I 0 50

0 00

___ r J i

l~T

~--

0 00 0 20 0 40 0 60 080 I 00 120 140 160 110 2 00

b

zshy

Posterior Distribution of Bias Factor (8) bull Effect of Removal of middotPlatform from Any Category

bullPrior and Posterior Distributions or B Evaluating Effect or Removal or 3 Damage Cases from the Base Case or 13 Platrorms

4

35

3

25

e L 2 ~ bulle shy

I 15

bull I

05

0

0 02 04 06 I 2 I 4 16 IS 2

I I

I

I

I Ibull

J -tmiddot L--

bullI

~

I

fl

II ~

I I v I

middot bull bull

bull L

~ ll- I-~lt

OS I

b

I

bull

1middot bullbullbullbull

~ __

bull

- - - - Prior-r

- - - - Postenor flall failure cases

Postenor flall success cases

--- Posterior flall 13 cases

---Posterior fI success failure cases only

IPosterior or B Mean =t11 COV =0121

-

_I

Posterir Distribution or Bias Factor (B) bull Effect or Removal or 3-Damage Cases

-

bullPrior and Posterior Distributions or B Evaluating Effect or Additional 13 Similar Observations as

Used in Andrew JIP The POsterior or B from Andrew-JIP Considered as New Prior or B

- 500

450

400

350

bull bull

bull bull bull bull J bull

bull bull f ~ I

I ~ Ibull bull

bullbull bull

l bull I -bullbull bull bull

I

I bull bull

bull I I~~bull bull bull

bull ~

I bull bull

~ ~_ ~~~

- - New Prior r bull

Posterior 114 new failure cases ~ 300- bull bull bull bull bull bull bull Posterior f 13 new damage casesl 150 ~

- bull bull - bull bull Posterior 116 new success cases i 2 00

Postenor 1113 new success failure -150 damage cases

I 00

0 50 -

0 00 0 00 0 20 0 40 0 60 0 80 I 00 120 I 40 160 110 2 00

Posterior ol B Mean= 120 COV= 0071 b

shy

shy

_rshy

Posterior Distribution of Bias Factor (B) bull Effect of Additional 13-Similar Observations

1

1 I

WEIGHTING APPROACH bull

bull ASSESSMENT OF PLATFORMS DATA

Select Platforms Most Mappable to the Analyzed Platforms

Unlikely SurvivalDamage Cases Identification

Based on Cut-Off Criteria Agemiddot Hs etc

bull INFORMATION REQUIRED FOR COMPARISON

bull CATEGORIZATION OF ADDITIONALmiddot PLATFORMS

bull UPDATING OF CALIBRATION RESULTS

--

OVERVIEW OF PLATFORMS SURVIVEDDAMAGED -~

Operator -

- Platforms

Survived CS)

Platforms

Damaged CD)

Platforms Used In

Phase IJIP -

JI~ ~Bill~IRBDll -

Amoco -

Chevron -

Exxon

Mob II

Phllllps

Shell

Unocal -

30

143

72 22

7

39 25shy -

1

12

bull

1 bull

1

1(S)+1 (D)

3 (S) +3(D)

2 (S)

bull shy

bull

bull

bull

Total Partlclnants Platforms 338 - 15 10C6S+4D) -

Other Operators

310

13

3 (D)

Total Platforms 648 28 13C6S+7D) -

Summary of Andrew JIP bull Phase I Platforms

Plairshy w v- N-ber Lea-le 1 shy car Hind_ o-w N-bull Deplh fl fl IJpe la bull n Yeadon Jadtet lbelledtf liefalit -- _ -OD

IL H - -shy _ICllshy IL IL - - rQClto ~ _

K STUIK 137 1963 8 365 608S4 Cellardeck

-middot STl-shy 170 1964 4 -~ K 39 62269 Cellardeck - --ST134W 137 1911 4 43 608S4 Smvlnlshy- _ - -

WD90A 184 1964 8 K 267Sbull34 83 SOS31 SUboftllar deck SmvMI

MC311 343 1978 8 x S7 62 - MC397 468 1991 4 x shy S33bull6SO 65911 -shySmeSmvlvol -- gt -

r-- T23ST52 63 1969 4 ton- shyKshy 52594 S03 -T25SSl39 62 1969 4 Kshy SOS94 S063 - middot ton-shy__ -

bullbull a ton--shyST161A 118 1964 8 K - 347S 51495 Cellarclecti

r~

ST1771 142 1965 8 K 39 60167 CeDordeck failme

STISIH 137 1964 8 K 395 608S4 CeDardeck Failme

K ST130A 140 1958 a SS 6095 I lblv Pai1uno - T21ST72 61 1969 4 Kshy S4S94 4974 I lhlv Pai1uno -

- -

JE

INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON

bull WATER DEPTH

bull bull YEAR OF INSTALLATIONDESIGN

bull PLATFORM ORIENTATION

bull PLATFORM LATITUDE AND LONGITUDE

bull KEY SKETCHES OF PLATFORM STRUCTURAL FRAMING

Number of Legs Plies Leg Batter Number of Conductors bull Deck Elevations (Including sub-cellar deck If any) Jacket Framing Elevations Brace type In Vertical Frames Sizes of Primary Members (leg pile vertical frames braces primary horizontals) Joint Details (can gap etc) LegPiie Annulus Grouting Status

bull middotPOST ANDREW INSPECTION REPORT Significant Wave Height Hs (ft)

bull INFORMATION ON DAMAGES PRIOR TO ANDREW

bull RESULTS FROM IN-HOUSE STUDIES IF AVAILABLE

bull ~RIGINAL DESIGN CRITERIA INFORMATION IF AVAILABLE

bull SOIL SHEAR STRENGTH PROFILE IF AVAILABLE

Platforms Identified for Evaluating Therr Likelihood of Falling Under Unexpected Survival Category

OPlRATOR

middotADA

BLOClt

llmucnJllE

NAME

YEAll

INSTAUJD

WA1ER

DBPTll

DISTANCE

JlllOM SHOllE

(MILU)

Bl

(M)

AMOCO 111113

STl11

WDI075

A

bull D

St

17

120

1n

44

20

10 I

CllllVllQfl STlllO c 61 110 21 11 STIUO D Q 1110 21 11

STl134 p Q 137 29 11

ST0151 G 61 137 11

ST0151 I 3 121 11

ST0151 J Q 140 11

STl151 It 137 11

ST0151 PJl001 Q UT 11

ST0151 PROD-2 Q UT 11

ST017 A 3 140 35 11

WDIUT c 65 214 34 10

llXXON STI055 II 51 6T 14 I

STl15 A t3 34

WDI073

WD0073

Amiddotshy Q

1111 1111

22

22

I

I

WDI073 c 1n 22 I

- WDOCl93-shy II

65 1110 2T A 3 194 31 MOBU 111121

SS IOl7

A B

501

50

50

31

T

I

T 6

MtlllPllY SS114 B T 50 T T

PlllllIPS SS0149 A 5 55 33 T

SAMllDAN STln A 107 T T

SBllLL WDl103 A 65 223 2T WDl103 B 65 221 2T WD0104 c 65 221 2T WD0104 D 6T 20 2T

UNOCAL SS020I

SS 020I SSl20I

SS 0209

SS0209

II p

PURll

A

B

3

n 61

103

97

97

95

100

34

34

34

34 34

bullbull bullbullbull

- SS0209 D u 95 34 I

- SS 1209 SSl253

SS0253

G A

c

6T

2

100

165

1T5

34

54

54

bullbull I

I I

I I

SOUTH MARCH

~LAND

N

SOUTH PaTOI ENE ISWltD AREA

MMS Defined Limits ror Offshore Structures Requiring Inspections Following Andrew

t---Jbulln3 T25bull I bullTtl

bull1$1Htl1Abull bull1SIK1779bull

~

middot-DaTA AREA~T

CRANO rsLE

AREA bull1300

bullll4W

=shy

bullPath of Hurricane Andrew with Platforms Used In Calibration

Likelihood Function vs b All Success Case STISI K STIJOQ ST134W WD90A MCJll MCJ97

I 0000

0 9000

0 8000

07000

-Bo6000

tiiii 0 5000 -

04000

0 3000

0 2000

01000

0 0000

0 05 25

ri1rt middot-middot-middot-middot-middot-

I 5 2

b

---11 middot151K(blsu~ccsI

- bull - bull - fk 130Q(bfU~lCSS)

lie bull134W(blsuccc]

lk-WD90A[blsultCS]

- - - lk middotMC397(bfUtlCS]

--o-- lk-MCJll(bbuctcsJ

--- ll (bl 6 out of 6 Mlttcs case]

_

( Likelihood Functions middotSuccess Cases - p - 1 c I

CATEGORIZATION OF ADDITIONAL PLATFORMS

I I I

bull CATEGORIES

Unexpected Survivals

Expected Survivals

bull APPROACH

Identify Corresponding Platform(s) Evaluated In the JIP

Determine Max Wave Height at Platform Location

Evaluate Whether Wave Hit the Deck

middot - Evaluate Likelihood of R and S to Differ from the Base Cases

Review Joint Information

Assess Special Features

bull IDENTIFY NUMBER OF PLATFORMS IN EACH CATEGORY

I

UPDATING OF CALIBRATION RESULTS

bull INCLUDE EFFECT OF ADDITIONAL PLATFORMS ON bullBbull

Consider Increased 1Uncertalntles In R and S

Develop bullNeW Ukellhood Functions for Different Categories

middotLet bullxbull Number of Ukely Unexpected Survival Cases and y Number of Ukely Expected Survival Cases

Determine Combined Likelihood Function lk (b Iadditional platforms) middot

bull REVISED LIKELIHOOD FUNCTION

lk(b I all cases) = lk(b I 9 platforms) middot x lk(b Ix platforms) x lk(b I y platforms)

bull ESTABLISH REVISED BIAS FACTOR B

f8 (b) u f8 (b) lk (b I all cases)

II

t t ~

shy

APPLICATIONS OF BIAS FACTORS

ii APPLICABLE TO PARTICULAR LOAD AND RESISTANCE RECIPE

bull USE BY THE API COMMITTEES AND MMS

Criteria Revisions

Recommended Design Guidelines Revisions bull

bullbull REASSESSMENT STUDIES AND DECISION MAKING

middot

bull RISK ASSESSMENT STUDIES BY THE OPERATORS

Individual Platforms

Fleet of Platforms

middot IMR Program Implementation

bullbull NOT TO BE USED IN DESIGN

I I fl J

EXAMPLE USE OF BIAS FACTORS

bull PHASE I EXAMPLE APPLICATION

bull EVALUATION OF ANNUAL PROBABILITY OF FAILURE

Establish Annual Maximum Seastate Data

Determine Coefficients to Define Base Shear

Determine Ultimate Capacity of Platform

Determine Probablllty of Failure Versus bullbbull

-bull DETERMINE UPDATED PROBABILITY OF FAILURE

Pf - P [S gt R]

Pf I P [~ gt Rb I b] f 8 (b) db

Pr - J P [S gt Rb I b] f bull 8 (b) db

bull TOTAL PROBABILITIES OF FAILURE

Total Pf - 1 bull Il [1 middotPf] - l Pf bull

Total Pf - 1 bull Il [1middot Pf] - l Pfmiddot

Total Pr = 1 - n [1middot PfbullJ = l Pfshy

I I I

EXAMPLE APPLICATION bull PHASE I

PLATFORM ST 151 K

bull GULF OF MEXICO

bull WATER DEPTH = 137 FT

bull ANNUAL MAXIMUM HS MEDIAN 175 ft COV 031

bull ASSUMED STORM DURATION 3 hoursmiddot

bull ASSUMED NUMBER OF WAVES 800

bull MEDIAN HS FOR STORM HOURS 1 2 AND 3

bull MEDIAN CURRENT VALUE bull

bull DETERMINE ULTIMATE CAPACITY FOR THREE DIRECTIONS

bull DETERMINE ANNUAL FAILURE PROBABILITIES

Broadsidemiddot Diagonal End-on

0037 0025 0007

0071 0044 0018

0022 0016 0004

DETERMINE TOTAL ANNUAL PROBABILITY OF FAILURE

Pf - 007

Pf - 013

Pr - 004

bull

EXAMPLE USE OF BIAS FACTORS - PHASE II middot

bull DETERMINE PREDICTIVE PROBABILITY OF FAILURE

Case - 1

Pf _ S S Pf (b b2) fa1a2 (b b2) db db2

g bull -1 (b1 R I S) - 1

92 - (b2 R2 I S) - 1

Case -2

Pf s ~ [bJ f11 (b) db

Pf [bJ p [g lt 0)

g - bull 1 (b1 R1 I S) - 1

1

bull1) l

I

BENEFITS OF ANDREW JIP bull PHASE II

lt

bull IMPROVED LOAD AND RESISTANCE RECIPE

bull IMPROVED CALIBRATION PROCEDURE

bull MULTIPLE BIAS FACTORS

bull PROCEDURE TO INCLUDE EFFECT OF OTHER PLATFORMS

bull PROCEDURE TO USE MULTIPLE BIAS FACTORS

  • Structure Bookmarks
    • ANDREW PHASE II JOINT1 INDUSTRY PROJECT
    • PROGRESS MEETING NUMBER 1
    • ANDREW PHASE II JOINT INDUSTRY PROJECT PROGRESS MEETING NUMBER 1 NOVEMBER 16 AGENDA
    • ANDREW PHASE II UST OF PARTICIPANTS
    • ANDREW PHASE I OBJECTIVES AND RESULTS
    • ANDREW PHASE II PROJECT OBJECTIVES
    • SUBSEQUENT PROJECTS
    • ANDREW PHASE II PROJECT TEAM
    • ANDREW PHASE II UPDATE MODELING middotAND ANALYSIS RECIPE
    • -ANDREW PHASE II CONDUCTOR RESISTANCE
    • ANDREW PHASE Il WA VE FORCE PROFILE
    • ANDREW PHASE II WA VE FORCE PROFILEmiddot
    • ANDREW PHASE II BRACE MODELING
    • ANDREW PHASE II FOUNDATIONS
    • ANDREW PHASE II middot middot JOINT MODELING
    • ANDREW PHASE Il JOINT MODELING OPTIONS DISCUSSION
    • ANDREW PHASE II JOINT MODEJJNG OPTIONS ILLUSTRATIONS
    • ANDREW PHASE II JOINT MODELING PHASE II METHOD
    • ANDREW PHASE II ANDREW STORM CONDmONS UPDATED
    • Platform Selection bullbullbull1111111111111
    • BasisCriteria for Selection bullbullbull1111111111111
    • CALIBRATION
    • WHERE TO INTRODUCE BIAS FACTOR (B
    • PHASE I -KEY CONCLUSIONS
    • bull CALIBRATION ISSUES PHASE II
    • PHASE II WORK SCOPE (CALIBRATION)
    • SOFTWARE
    • SENSITIVITY STUDIES POSTERIOR DISTRIBUTION OF B
    • WEIGHTING APPROACH bull
    • INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON
    • CATEGORIZATION OF ADDITIONAL PLATFORMS
    • UPDATING OF CALIBRATION RESULTS
    • APPLICATIONS OF BIAS FACTORS
    • EXAMPLE USE OF BIAS FACTORS
    • EXAMPLE APPLICATION bull PHASE I PLATFORM ST 151 K
    • EXAMPLE USE OF BIAS FACTORS -PHASE II middot
    • BENEFITS OF ANDREW JIP bull PHASE II
Page 34:  · . ' ANDREW PHASE II PROJECT OBJECTIVES . ' • ·UPDATE MODELING AND ANALYSIS RECIPE· JOINTS ' ' ·BRACES WAVE PROFILE FOUNDATIONS . ' ' CONDUCTORS· • ANALVZE PLATFORMS USING

~--~~~~~~~~~--~----~__i~~---~~~~__--~

Pro)ect apipile Model South Pelto 10 Version Wed May 18072235 1994 - -

1bullt

+

t

f i bullbullbull

bullbull bullbull

I I

5

~x SPlO Caissonfy=42s soil=api(lf=lOOstep=137)

Inelastic Events Legend

Elastic Strut Bucklilg

Strut Residual Strut Reloading

Plastic StrutNLTruss Beam Clmn Initial Yield

earn Clmn Full~ Plas1c Fracture

ANDREW PHASE II middot middot

JOINT MODELING

middotbull API RIGID-PLASTIC USED IN PHASE I

a ELASTIC-BRITTLE USED IN PHASE IB

amiddot MSL WILL PROVIDE NONLINEAR FORCE-DEFORMATION

AND MOMENT-DEFORMATION RELATIONS FOR PHASE II

a MODELING OPTIONS FOR PHASE II -

l

- UNCO~LEDmiddotONE DIMENSIONAL MEMBER END JOINT MODELING

J Ii

ANDREW PHASE Il

JOINT MODELING OPTIONS DISCUSSION

bull JD COUPLED INTERACTION SURFACE

- IDEAL

- DATA NOT AVAILABLE TO CALIBRATE

bull 2D COUPLED INTERACTION SURFACE

- SIGNIFICANT MODELING IMPROVEMENT

- DATAAVAILABLE

- IMPLEMENTATION PRACTICAL BUT NEEDS DEVELOPMENT AND CALIBRATION

bull 2D UNCOUPLED INTERACTION SURFACE

- SIGNIFICANT MODELING IMPROVEMENT

- DATA AVAILABLE

- IMPLEMENTATION PRACTICAL AND AVAILABLE

ANDREW PHASE II

JOINT MODEJJNG OPTIONS ILLUSTRATIONS

P

e

ANDREW PHASE II

JOINT MODELING PHASE II METHOD

e

I I I

ANDREW PHASE II

ANDREW STORM CONDmONS UPDATED

bull -PHASE I USED OCEANWEATHER HINDCAST-(Nov 1992)

bull middotmiddotPHASE II WILL USE THE REVISED HINDCAST (Nov 1994)

bull lt c lt I lt

2-2

l---4-~~~+-~~-+-~~~1--~~-+-~~--l--~T--+-~~--i24

f-tJVtMfiz~ 1Cfl (r STvb(

Plollol bullbull 5middotNOYmiddot9214bull1J Irbullbull Illbull [ANDREVl~NDREVCUSTbullMAXll 5-NOY-1992 10bull54 - 9208

HEIGHT lml

I~ MAXI

I L 1~middotamp SZ~- ~A I 13c

I J) ( r I X~0iWJ~0gtJ gt L gt gt I 1c I 120

-- N

~ deg

I I I I I I e C I ~=-41 122 -94 -92 -90 -88 -86 -84 -82 -80

-

Figure 7 contours of maximum hindcast significant wave height

_

~

~tl ~

NOltNttIJ 17lt ltNS S iJJ)y

Maximum Significant Wave Height - Hurricane Andrew 1992

30 ~

26 ~ I ~-~~~- == -shy -~ 35-_

24deg1--shy --------~- middot-middot-------middotmiddotmiddot----middot------middot-middot~middot- - - ~-----1---middot

22middot ~~~~~~~-~ -94 -92 -90 -88 -84deg-86 -82 -somiddot

c

-

Platform Selection bullbullbull1111111111111

0 BasisCriteria for Selection middot

0 Review of Phase I Platforms middot 0 Alternative Platforms

0 PMB Recommendations

0 Participants Comments

gt

L

l I

BasisCriteria for Selection bullbullbull1111111111111

0 Benefit for Calibration - Classification (UnexpectedExpected)

- Failure Mode (StructuresFoundations)

- Damaged StructuresKnown response

- Better Representation of Full ~opulation

0 Availability of Data - Existing Models

- Inspection Data

- Damage Data

PREDICTED BEHAVIOR

-

OBSERVED

BEHAVIOR

-SURVIVED DAMAGED FAILED

SURVIVED

ST134W WD90A

-MC311 -

MC397 -

--

-

-

ST151K ST130Q

-

-

-

DAMAGED - -

-

-

--

-

-

ST52 SS139 ST177B -

ST161A

-_

-

FAILED

-

-

-

ST151H ST130A ST72

Operator - Platfonns - Platfonns Survived Damagedamp

- Falled -

bull

-

-

Jle eBDIGIPllDbl -Amocoshy 30 1 Chevron -

143 12 Exxon 72 Mobll 22 1 Phllllps 7 bull

-Shell 39 Unocal 25 1

Total Partlclnts Platfonns 338 15

Other Operators bull- 310 13

-

Total Platfonns - -

648 28

fbml

Platforms Considered 6 7

tie of Total Platfonns 1 25

OVERVIEW OF PLATFORMS SURVIVEDDAMAGEDFAILED

-

shy

v

Platforms Evaluated in Phase I

-

-Platfonn Water Year Number Analysis Avallabillty of Availability Platform

- Name Depth rl of Model DetaDed rl ClassUlcatlon

Installation legs Inspection Soll Report Based on Information Phase I Calibration - -ft

- -- Suaill flalfnrm ~BSi -STISIK 137 1963 8 CAP Yes Yes Un eel Sarvlval

-STIJOO 170 1964 4 CAPmiddotshy Yes ST134data u ST134W 137 1981 4 CAP Yes Yesmiddot E

-

edSarvlval -

middotsurvival

WD90A 184 - 1964 8 CAP Yes E -

- Sarvlvill

MC3ll 343 1978 8 KARMA

MC397 468 1991 4 KARMA -

bull ~reglmmiddot- -

Yes Sure Survival

bull Yes - Sare Sanlval-

-T23STS2 63 1969 4 CAP

T2SSS139 62 1969 4 CAP -

Yes

Yes

Yes Ex

Yes Une

bull SurvivalLikely to DamalI

middot bull SanlvalLlkely to DamalI

ST161A 118 1964 8 USFOS

Eail11a pound1al(bullllm Clss ST177B 142 --1965 8 CAP

STISIH 137 1964 - 8 CAP C

STIJOA 140 19S8 8 CAP - -

Yes

Yes

ollamed

Yes

- Yes E

-

at about 1 mUe

Yes

ST134data

bull SafvlvalLlkelv to Dam- bull- -

E middotrahre

Emeried Failure -

Likely FaUare

T21ST72 61 1969 4 CAP Collamed No Likely FaUure

Platforms Identified for Evaluating Their Likelihood of Falling Under UnexpeCted Survival Category

CllEllATOa

AllB

llLOCll

Sl1lvcnJllE

NAME

YEAR

INSTAUJCD

WATER

IJBPJB

middotmiddotDISTANCE

ftOMSllOllE

(MIUIS)

Ill

(M)

AMOCO BllW

STllI

WDI075

A

bull D

4

n uo 17J

44

32

bull 20

10

bull CllEVllOH STtllO c I 1bull 21 11

STl130 D 0 161 21 11 STllJ4 I 0 137 D - 11

STIUI G I 137 32 11

STIUI I 0 121 32 11

STIUI

J 0 140 32 11

STIUI It 4 137 32 11

STllSI PROl)I 0 137 32 11

STllSI PROD-2 0 137 32 11

STll7 A 0

140 35 11

WDl117 c 65 214 34 10

DXON STshy E 51 14 bull STl165 A t3 bull 34 WD0073

WD0073

WD0073

A _ c

4

62

4

IA IA

17J

22

22

22

bullbullbull WD003-shy E 65 161 Z7

A 0 IN 31 MOBD Bl12I

ssn A bull

50

50

31

7

bull 7

MIJllPllY SS114 B 7 50 7 7

PlllUIPS SSl14t A 55 33 7

SAMEDAN ST17J A 4 117 7 bull 7

SJIELL WD0103 A 65 223 27 WDllOJ bull 65 221 27 WDll04 c 65 221 27 WDOI04 D 67 2 27

UNOCAL

SSl20I

SSl20I

bull SSl20I

SS0209

SS0209

SS0209

SS0209

SS 0253

SS 0253

E

I

PtJllE

A

bull D

G A

c

0

4

4

7J

61

0

0

69

103

97

97

9S

100

95

100

165

175

34

34

34

34

34

34

34

54

54

bullbullbullbullbullbullbullbullbull

Platforms Recommended for Phase II

llalr y_ w- Aot Av- _ I 1 Doplla Model nm bull bull _ bull I I H11 s18-1 Irht lbwl~ bullbullbull(-illtdlel OM lllMtD

II ---

-STISIIC 137 1963 I CAP Yeo Yeo I Yeo

STtshy -170 1964 4 CAP Yeo ST134dlU I Yeo

STl34W 137 Yeo Yeo -1981 4 CAP D Yeo

WD90A 114 1964 I CAP Yeo - D

MC311 30 1971 I KARMA Yeo -sre-

MC397 middot 461 1991 4 ICARMA Yeo Sireshy - - r- -

__ T13ST5l 63 1969 4 CAP Yeo Yeobull middot- ~n- D Yeo _ Tl5SSl39 62 1969 4 CAP Yeo Yeo I Yeo

_ STIT711 142 1965 I CAP Yeo I Yeo middot-1shySTl61A Ill 1964 I USFOS Yeo Yeo - D bulloshy

- r-

STISIH 137 1964 I CAP y D y -_ STl31A 140 1951 I CAP Yeo STl34dlU I Yeo

T21Sll 61 1969 4 CAP No rar-middot I Yeo

~-rmiddot-middot SS209A 95 1967 16 CAP Yeo D -SSU4R

11 UI A 41 -in D SS 2741AIM CAP bull

_ _ -shy - -shy- -shy

CALIBRATION

bull PHASEIAPPROACH

bull CALIBRATION ISSUES

bull PHASE II WORK SCOPE

bull SOFlWARE AVAILABILITY

bull EFFECT OF ADDITIONAL PLATFORMS

bull WEIGHTING APPROACH

bull EXAMPLE APPLICATION middot

(l=Ofce Mode] Wind

~

fcurrent l -middot

~SW ~leldl4

ultimate resistance force I first

design load Illgt

i _

deflection

-

bull

4

Historical Hurrlcane Dariiage and Survival of Platforms

Components Required for Calibration (Petrauskas 1992)

PRIOR Distribution

PLATFORM EXPERIENCE

BAYESIAN Survivals and Failures

UPDATING

Waves Winds and Currents bull

POSTERIOR Distribution

- A_middotJA ~ Survival Data Failure Data

Result Result

Load Resistance

middotfl

-

Bayesian Updaing middotGeneral Description (Petrauskas 1992)

L Imiddot

WHERE TO INTRODUCE BIAS FACTOR (B

On Resistance (Rmue - B (RPREDICreD I

middotOn Loads (Smue middot - B (SPRE01cTEo

On Safety Margin (RSmue - B (RSPREDICTED

CALIBRATION METHODOLOGY

gt

r

FIELD OBSERVATIONS

Observed Behavior

survivals Damages

Failures

r

bull

CAPACITY ANALYSIS

Predicted Load and Resistance Predicted Failure Modes

Survival Damage Failure cases

RELIABILITY ANALYSIS

- Probability of Failure ~

Ukellhood Failure

BAYESIAN UPDATING

Prior Distnbution of B Combined Likelihood Function

Posterior Dlstnbution of B

BIAS FACTOR

Mean Value cov bull

Represents Modeling Uncertainties

Prior Distribution of B Likelihood Functionmiddot Observing Success case STISIK and Posterior Distribution of B

2 00

uo

160

140

120

e 100 L $ uoIshy

0 60 ~ ~ 040

020

0 00

0 00 0 50

_ I 00

b

I 50 2 00

-shy lk [bl STISIK)

--shy Pnor-f(b)

--shy Pos1enor r(bl ISIK)

Prior or B Mean bull 10 COV bull 030

Posterlor of Bbull Meanbull 136COV 0 16

2 50

Likelihood Function and Posterior Distribution Example Platform STISIK

- -

~

- Prior and Posterior Distributions or B Failure cases (ST177B STlSlH ST130A T21) Damage cases (T23 T25 ST161A) Success cases (STlSlK ST130Q ST134W WD90A MC311 MC397)

400

350

300

2 so 0-

I

bull bull

bull bull gt

- bullbull ~ bull bull I bull

_ bull bull bullbull I bull bull bullII V -

bull bull

I bull ~ bull

bullbull bullI bull bullI bull bull bull

middot -~

bull__ ~ middotI middotmiddot

L 2 00

e ~ I SO

-I 00

- 0 so

000 gt

0 00 020 0 40 060 0 80 I 00 120 140 160 110 200

b -

-

shy

Prior-r

- - - - Posterior fLlll failure cases

bull bull bull bull bull bull bull Posterior flall damage cases

- bull bull - bull middot Postenor flall success cases

Posterior flall success fulure damage cases

-

Posterior of B Mean= 119 COV 0101

Posterior Distribution or Blas Factor (8) bull Different Categories and All Cases Combined

shy

l

PHASE I - KEY CONCLUSIONS

bull GLOBAL CORRECTION FACTOR

bull 15 TO 20 CONSERVATISM MEAN)

bull UNEXPECTED SURVIVALS HAVE GREATEST IMPACT

bull DAMAGED CASES MORE INFORMATIVE

bull CALIBRATION ISSUES PHASE II

bull

EFFECT OF IMPROVED CAPACITY ANALYSIS RECIPE

bull APPLICABILITY TO MULTIPLE FAILURE MODES

bull NO FOUNDATION DAMAGE CASE INCLUDED

bull IMPROVED REPRESENTATION OF FULL ANDREW DISTRIBUTION

-bull NUMBER OF PLATFORMS CONSIDERED

I 1 I

I t ) bull ~

PHASE II WORK SCOPE (CALIBRATION)

bull

bull ESTABLISH MULTIPLE BIAS FACTORS

Use Nine Base Case Platforms

middot - Capacity Analysis for Revised recipe

Foundation Blas Factor from Foundation JIP

bull DEVELOP WEIGHTING APPROACH

bull UPDATE MULTIPLE BIAS FACTORS

bullEffect of Platforms Not In the Nine Base Cases

bull EXAMPLE APPLICATION OF BIAS FACTORS I

bullbull

PHASE I PHASE II

CAPACITYbull ANALYSISRu bull

---------------------1----middot

Jc1 - iJbullraquoJ dbull

FttlS frt[f~l ratdhJJ

-

LIKELIHOOD FUNCTION

DEVELOPMENT

lk Jb Jfall1N) bull Jb)

111 Ill Isurvtvoq bull 1 - bl

-middot -------------------shy ~---------------------middot-middot

r111ir8 Cb1 111 Cbl BAYESIAN

UPDATING

~cove

R1R2 bull middot---------------------~----

Jblbull P(g cOJ

1 bull JbRI (MBSI I bull 1

~middot

llcJllJfalluro) bull P(gcOJ Ill 111 b21fallun) bull P ((g1 CO) U 1112 c OD

middot--------------------shy ----------------------shy bull

r111111r8 Cbl 111 Ill i-1nro1

deg ~1b2)laquottfrr (b1b21-lnfo) re Ill bull 1-11middot b2) bull lodegamp

middot

i1 covamp i2 cov~i

-

shy

--

Likelihood Function vs b All Success Case STISI K STl30Q STl34W WD90A MC311 MC397

I 0000

0 9000

OIOOO

0 7000

~ 0amp000

ta osoOO

~ 0 4000

0 3000

0 2000

0 1000

bull 0 0000

bull

I

middot

middot bull I

Ibull

-sibull ~ -middot-middot-middot-middot -middot-

----7----+-----~ I

Ibull bull bull I

rbull

--- Jk ISIK[blsuccess)

- bull - bull - lk 130Q[blsuccess) bull

lk 134W[blsuccess)

Jk-WD90A(blsuccc~I

- - - lk MC397[blsucccss)

--ltgt-- lk middotMC31 l[blsucccss)

--- Jk [bl 6 out of6succcsHascs)

05 I 5 2 250

b

Likelihood Functionsmiddot Success Cases - Phbullsc L

shy

bullbull bull bull bull bull bull -------------------~

bull bull bullbull bullbull bullbull bullbull bullbull bullbull bull

bull bullbull

bullbullbullbull

bullbull

bullbullbullbullbullbullbull

bullbullbullbullbullbull

bullbullbullbullbullbullbull

z 0

t z ) LL

c 0 0c i ~ i

J

~ sect ~--------------~ e l5

I

I

SOFTWARE

bull PF PROGRAM

bull CALREL

bull PF +CALREL

bull RELACS

bull

gt i

l

SENSITIVITY STUDIES

POSTERIOR DISTRIBUTION OF B

bull EFFECT OF CHANGE IN DISTRIBUTION OF PRIOR OF B

bull EFFECT OF REMOVAL OF ONE PLATFORM FROM EACH CATEGORY

bull EFFECT OF REMOVAL OF THREE DAMAGE CASES

bull EFFECT OF ADDITIONAL 13 SIMILAR PLATFORMS

-I

Prior and Posterior Distributions of B Variation In COV of PriorofB fromOl to04 -- -

4500

-4000

-

- -

3500 --middot - bull bull - bull bull Prior-r ICOV =02

3000

- - - - Prior-r ICOV =03 -

bull -~ e 2soo -

bull bull bull bull bull bull bull Pnor-r ICOV =o 4t frJ ~ 2000 Posterior rlCOV =0 2

bull bull Postenor rlCOV =0 3 bull bull 1500 bull middot bull - shy -middot- Posterior rlCOV =04

middot~ I 000 - bull

-middotmiddot ~ ~middotmiddot M -~

- middot 0 500

wbullbullbull _ I middot~ middotmiddotmiddotmiddotmiddot middotmiddotmiddotmiddotmiddot - -~ ~ __ middot_ _ bull -shy0 000

0 000 0200 0 400 0 600 0 800 IOOO 1200 1400 1600 1800 2000 -

- - b

-

=shy

Posterior Distribution of Bias Factor (B) bull Effect of Variation In COV of Prior of B

bull

bullPrior and Posterior Distributions of B Evaluating Effect of Removal of One Platform from Each Category from the Base Case of 13 Platforms

400

~ ~ 350

j IA i

3 00 bull I bull I ---Prior-r I bull l

250 1 I 1

I bull bull ---Posterior r lall 13 casesshy- ~E 2 00 bull bull bull bull bull bull bull Posterior r without STISIK

- ~ 1 bull - bull I - bull bull - bull bull Posterior r Iwithout T2I I 50 I

J-tt~i - bull - bull - Posterior r Iwithout T2S

tOO [-

I I ~ ~ I 0 50

0 00

___ r J i

l~T

~--

0 00 0 20 0 40 0 60 080 I 00 120 140 160 110 2 00

b

zshy

Posterior Distribution of Bias Factor (8) bull Effect of Removal of middotPlatform from Any Category

bullPrior and Posterior Distributions or B Evaluating Effect or Removal or 3 Damage Cases from the Base Case or 13 Platrorms

4

35

3

25

e L 2 ~ bulle shy

I 15

bull I

05

0

0 02 04 06 I 2 I 4 16 IS 2

I I

I

I

I Ibull

J -tmiddot L--

bullI

~

I

fl

II ~

I I v I

middot bull bull

bull L

~ ll- I-~lt

OS I

b

I

bull

1middot bullbullbullbull

~ __

bull

- - - - Prior-r

- - - - Postenor flall failure cases

Postenor flall success cases

--- Posterior flall 13 cases

---Posterior fI success failure cases only

IPosterior or B Mean =t11 COV =0121

-

_I

Posterir Distribution or Bias Factor (B) bull Effect or Removal or 3-Damage Cases

-

bullPrior and Posterior Distributions or B Evaluating Effect or Additional 13 Similar Observations as

Used in Andrew JIP The POsterior or B from Andrew-JIP Considered as New Prior or B

- 500

450

400

350

bull bull

bull bull bull bull J bull

bull bull f ~ I

I ~ Ibull bull

bullbull bull

l bull I -bullbull bull bull

I

I bull bull

bull I I~~bull bull bull

bull ~

I bull bull

~ ~_ ~~~

- - New Prior r bull

Posterior 114 new failure cases ~ 300- bull bull bull bull bull bull bull Posterior f 13 new damage casesl 150 ~

- bull bull - bull bull Posterior 116 new success cases i 2 00

Postenor 1113 new success failure -150 damage cases

I 00

0 50 -

0 00 0 00 0 20 0 40 0 60 0 80 I 00 120 I 40 160 110 2 00

Posterior ol B Mean= 120 COV= 0071 b

shy

shy

_rshy

Posterior Distribution of Bias Factor (B) bull Effect of Additional 13-Similar Observations

1

1 I

WEIGHTING APPROACH bull

bull ASSESSMENT OF PLATFORMS DATA

Select Platforms Most Mappable to the Analyzed Platforms

Unlikely SurvivalDamage Cases Identification

Based on Cut-Off Criteria Agemiddot Hs etc

bull INFORMATION REQUIRED FOR COMPARISON

bull CATEGORIZATION OF ADDITIONALmiddot PLATFORMS

bull UPDATING OF CALIBRATION RESULTS

--

OVERVIEW OF PLATFORMS SURVIVEDDAMAGED -~

Operator -

- Platforms

Survived CS)

Platforms

Damaged CD)

Platforms Used In

Phase IJIP -

JI~ ~Bill~IRBDll -

Amoco -

Chevron -

Exxon

Mob II

Phllllps

Shell

Unocal -

30

143

72 22

7

39 25shy -

1

12

bull

1 bull

1

1(S)+1 (D)

3 (S) +3(D)

2 (S)

bull shy

bull

bull

bull

Total Partlclnants Platforms 338 - 15 10C6S+4D) -

Other Operators

310

13

3 (D)

Total Platforms 648 28 13C6S+7D) -

Summary of Andrew JIP bull Phase I Platforms

Plairshy w v- N-ber Lea-le 1 shy car Hind_ o-w N-bull Deplh fl fl IJpe la bull n Yeadon Jadtet lbelledtf liefalit -- _ -OD

IL H - -shy _ICllshy IL IL - - rQClto ~ _

K STUIK 137 1963 8 365 608S4 Cellardeck

-middot STl-shy 170 1964 4 -~ K 39 62269 Cellardeck - --ST134W 137 1911 4 43 608S4 Smvlnlshy- _ - -

WD90A 184 1964 8 K 267Sbull34 83 SOS31 SUboftllar deck SmvMI

MC311 343 1978 8 x S7 62 - MC397 468 1991 4 x shy S33bull6SO 65911 -shySmeSmvlvol -- gt -

r-- T23ST52 63 1969 4 ton- shyKshy 52594 S03 -T25SSl39 62 1969 4 Kshy SOS94 S063 - middot ton-shy__ -

bullbull a ton--shyST161A 118 1964 8 K - 347S 51495 Cellarclecti

r~

ST1771 142 1965 8 K 39 60167 CeDordeck failme

STISIH 137 1964 8 K 395 608S4 CeDardeck Failme

K ST130A 140 1958 a SS 6095 I lblv Pai1uno - T21ST72 61 1969 4 Kshy S4S94 4974 I lhlv Pai1uno -

- -

JE

INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON

bull WATER DEPTH

bull bull YEAR OF INSTALLATIONDESIGN

bull PLATFORM ORIENTATION

bull PLATFORM LATITUDE AND LONGITUDE

bull KEY SKETCHES OF PLATFORM STRUCTURAL FRAMING

Number of Legs Plies Leg Batter Number of Conductors bull Deck Elevations (Including sub-cellar deck If any) Jacket Framing Elevations Brace type In Vertical Frames Sizes of Primary Members (leg pile vertical frames braces primary horizontals) Joint Details (can gap etc) LegPiie Annulus Grouting Status

bull middotPOST ANDREW INSPECTION REPORT Significant Wave Height Hs (ft)

bull INFORMATION ON DAMAGES PRIOR TO ANDREW

bull RESULTS FROM IN-HOUSE STUDIES IF AVAILABLE

bull ~RIGINAL DESIGN CRITERIA INFORMATION IF AVAILABLE

bull SOIL SHEAR STRENGTH PROFILE IF AVAILABLE

Platforms Identified for Evaluating Therr Likelihood of Falling Under Unexpected Survival Category

OPlRATOR

middotADA

BLOClt

llmucnJllE

NAME

YEAll

INSTAUJD

WA1ER

DBPTll

DISTANCE

JlllOM SHOllE

(MILU)

Bl

(M)

AMOCO 111113

STl11

WDI075

A

bull D

St

17

120

1n

44

20

10 I

CllllVllQfl STlllO c 61 110 21 11 STIUO D Q 1110 21 11

STl134 p Q 137 29 11

ST0151 G 61 137 11

ST0151 I 3 121 11

ST0151 J Q 140 11

STl151 It 137 11

ST0151 PJl001 Q UT 11

ST0151 PROD-2 Q UT 11

ST017 A 3 140 35 11

WDIUT c 65 214 34 10

llXXON STI055 II 51 6T 14 I

STl15 A t3 34

WDI073

WD0073

Amiddotshy Q

1111 1111

22

22

I

I

WDI073 c 1n 22 I

- WDOCl93-shy II

65 1110 2T A 3 194 31 MOBU 111121

SS IOl7

A B

501

50

50

31

T

I

T 6

MtlllPllY SS114 B T 50 T T

PlllllIPS SS0149 A 5 55 33 T

SAMllDAN STln A 107 T T

SBllLL WDl103 A 65 223 2T WDl103 B 65 221 2T WD0104 c 65 221 2T WD0104 D 6T 20 2T

UNOCAL SS020I

SS 020I SSl20I

SS 0209

SS0209

II p

PURll

A

B

3

n 61

103

97

97

95

100

34

34

34

34 34

bullbull bullbullbull

- SS0209 D u 95 34 I

- SS 1209 SSl253

SS0253

G A

c

6T

2

100

165

1T5

34

54

54

bullbull I

I I

I I

SOUTH MARCH

~LAND

N

SOUTH PaTOI ENE ISWltD AREA

MMS Defined Limits ror Offshore Structures Requiring Inspections Following Andrew

t---Jbulln3 T25bull I bullTtl

bull1$1Htl1Abull bull1SIK1779bull

~

middot-DaTA AREA~T

CRANO rsLE

AREA bull1300

bullll4W

=shy

bullPath of Hurricane Andrew with Platforms Used In Calibration

Likelihood Function vs b All Success Case STISI K STIJOQ ST134W WD90A MCJll MCJ97

I 0000

0 9000

0 8000

07000

-Bo6000

tiiii 0 5000 -

04000

0 3000

0 2000

01000

0 0000

0 05 25

ri1rt middot-middot-middot-middot-middot-

I 5 2

b

---11 middot151K(blsu~ccsI

- bull - bull - fk 130Q(bfU~lCSS)

lie bull134W(blsuccc]

lk-WD90A[blsultCS]

- - - lk middotMC397(bfUtlCS]

--o-- lk-MCJll(bbuctcsJ

--- ll (bl 6 out of 6 Mlttcs case]

_

( Likelihood Functions middotSuccess Cases - p - 1 c I

CATEGORIZATION OF ADDITIONAL PLATFORMS

I I I

bull CATEGORIES

Unexpected Survivals

Expected Survivals

bull APPROACH

Identify Corresponding Platform(s) Evaluated In the JIP

Determine Max Wave Height at Platform Location

Evaluate Whether Wave Hit the Deck

middot - Evaluate Likelihood of R and S to Differ from the Base Cases

Review Joint Information

Assess Special Features

bull IDENTIFY NUMBER OF PLATFORMS IN EACH CATEGORY

I

UPDATING OF CALIBRATION RESULTS

bull INCLUDE EFFECT OF ADDITIONAL PLATFORMS ON bullBbull

Consider Increased 1Uncertalntles In R and S

Develop bullNeW Ukellhood Functions for Different Categories

middotLet bullxbull Number of Ukely Unexpected Survival Cases and y Number of Ukely Expected Survival Cases

Determine Combined Likelihood Function lk (b Iadditional platforms) middot

bull REVISED LIKELIHOOD FUNCTION

lk(b I all cases) = lk(b I 9 platforms) middot x lk(b Ix platforms) x lk(b I y platforms)

bull ESTABLISH REVISED BIAS FACTOR B

f8 (b) u f8 (b) lk (b I all cases)

II

t t ~

shy

APPLICATIONS OF BIAS FACTORS

ii APPLICABLE TO PARTICULAR LOAD AND RESISTANCE RECIPE

bull USE BY THE API COMMITTEES AND MMS

Criteria Revisions

Recommended Design Guidelines Revisions bull

bullbull REASSESSMENT STUDIES AND DECISION MAKING

middot

bull RISK ASSESSMENT STUDIES BY THE OPERATORS

Individual Platforms

Fleet of Platforms

middot IMR Program Implementation

bullbull NOT TO BE USED IN DESIGN

I I fl J

EXAMPLE USE OF BIAS FACTORS

bull PHASE I EXAMPLE APPLICATION

bull EVALUATION OF ANNUAL PROBABILITY OF FAILURE

Establish Annual Maximum Seastate Data

Determine Coefficients to Define Base Shear

Determine Ultimate Capacity of Platform

Determine Probablllty of Failure Versus bullbbull

-bull DETERMINE UPDATED PROBABILITY OF FAILURE

Pf - P [S gt R]

Pf I P [~ gt Rb I b] f 8 (b) db

Pr - J P [S gt Rb I b] f bull 8 (b) db

bull TOTAL PROBABILITIES OF FAILURE

Total Pf - 1 bull Il [1 middotPf] - l Pf bull

Total Pf - 1 bull Il [1middot Pf] - l Pfmiddot

Total Pr = 1 - n [1middot PfbullJ = l Pfshy

I I I

EXAMPLE APPLICATION bull PHASE I

PLATFORM ST 151 K

bull GULF OF MEXICO

bull WATER DEPTH = 137 FT

bull ANNUAL MAXIMUM HS MEDIAN 175 ft COV 031

bull ASSUMED STORM DURATION 3 hoursmiddot

bull ASSUMED NUMBER OF WAVES 800

bull MEDIAN HS FOR STORM HOURS 1 2 AND 3

bull MEDIAN CURRENT VALUE bull

bull DETERMINE ULTIMATE CAPACITY FOR THREE DIRECTIONS

bull DETERMINE ANNUAL FAILURE PROBABILITIES

Broadsidemiddot Diagonal End-on

0037 0025 0007

0071 0044 0018

0022 0016 0004

DETERMINE TOTAL ANNUAL PROBABILITY OF FAILURE

Pf - 007

Pf - 013

Pr - 004

bull

EXAMPLE USE OF BIAS FACTORS - PHASE II middot

bull DETERMINE PREDICTIVE PROBABILITY OF FAILURE

Case - 1

Pf _ S S Pf (b b2) fa1a2 (b b2) db db2

g bull -1 (b1 R I S) - 1

92 - (b2 R2 I S) - 1

Case -2

Pf s ~ [bJ f11 (b) db

Pf [bJ p [g lt 0)

g - bull 1 (b1 R1 I S) - 1

1

bull1) l

I

BENEFITS OF ANDREW JIP bull PHASE II

lt

bull IMPROVED LOAD AND RESISTANCE RECIPE

bull IMPROVED CALIBRATION PROCEDURE

bull MULTIPLE BIAS FACTORS

bull PROCEDURE TO INCLUDE EFFECT OF OTHER PLATFORMS

bull PROCEDURE TO USE MULTIPLE BIAS FACTORS

  • Structure Bookmarks
    • ANDREW PHASE II JOINT1 INDUSTRY PROJECT
    • PROGRESS MEETING NUMBER 1
    • ANDREW PHASE II JOINT INDUSTRY PROJECT PROGRESS MEETING NUMBER 1 NOVEMBER 16 AGENDA
    • ANDREW PHASE II UST OF PARTICIPANTS
    • ANDREW PHASE I OBJECTIVES AND RESULTS
    • ANDREW PHASE II PROJECT OBJECTIVES
    • SUBSEQUENT PROJECTS
    • ANDREW PHASE II PROJECT TEAM
    • ANDREW PHASE II UPDATE MODELING middotAND ANALYSIS RECIPE
    • -ANDREW PHASE II CONDUCTOR RESISTANCE
    • ANDREW PHASE Il WA VE FORCE PROFILE
    • ANDREW PHASE II WA VE FORCE PROFILEmiddot
    • ANDREW PHASE II BRACE MODELING
    • ANDREW PHASE II FOUNDATIONS
    • ANDREW PHASE II middot middot JOINT MODELING
    • ANDREW PHASE Il JOINT MODELING OPTIONS DISCUSSION
    • ANDREW PHASE II JOINT MODEJJNG OPTIONS ILLUSTRATIONS
    • ANDREW PHASE II JOINT MODELING PHASE II METHOD
    • ANDREW PHASE II ANDREW STORM CONDmONS UPDATED
    • Platform Selection bullbullbull1111111111111
    • BasisCriteria for Selection bullbullbull1111111111111
    • CALIBRATION
    • WHERE TO INTRODUCE BIAS FACTOR (B
    • PHASE I -KEY CONCLUSIONS
    • bull CALIBRATION ISSUES PHASE II
    • PHASE II WORK SCOPE (CALIBRATION)
    • SOFTWARE
    • SENSITIVITY STUDIES POSTERIOR DISTRIBUTION OF B
    • WEIGHTING APPROACH bull
    • INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON
    • CATEGORIZATION OF ADDITIONAL PLATFORMS
    • UPDATING OF CALIBRATION RESULTS
    • APPLICATIONS OF BIAS FACTORS
    • EXAMPLE USE OF BIAS FACTORS
    • EXAMPLE APPLICATION bull PHASE I PLATFORM ST 151 K
    • EXAMPLE USE OF BIAS FACTORS -PHASE II middot
    • BENEFITS OF ANDREW JIP bull PHASE II
Page 35:  · . ' ANDREW PHASE II PROJECT OBJECTIVES . ' • ·UPDATE MODELING AND ANALYSIS RECIPE· JOINTS ' ' ·BRACES WAVE PROFILE FOUNDATIONS . ' ' CONDUCTORS· • ANALVZE PLATFORMS USING

ANDREW PHASE II middot middot

JOINT MODELING

middotbull API RIGID-PLASTIC USED IN PHASE I

a ELASTIC-BRITTLE USED IN PHASE IB

amiddot MSL WILL PROVIDE NONLINEAR FORCE-DEFORMATION

AND MOMENT-DEFORMATION RELATIONS FOR PHASE II

a MODELING OPTIONS FOR PHASE II -

l

- UNCO~LEDmiddotONE DIMENSIONAL MEMBER END JOINT MODELING

J Ii

ANDREW PHASE Il

JOINT MODELING OPTIONS DISCUSSION

bull JD COUPLED INTERACTION SURFACE

- IDEAL

- DATA NOT AVAILABLE TO CALIBRATE

bull 2D COUPLED INTERACTION SURFACE

- SIGNIFICANT MODELING IMPROVEMENT

- DATAAVAILABLE

- IMPLEMENTATION PRACTICAL BUT NEEDS DEVELOPMENT AND CALIBRATION

bull 2D UNCOUPLED INTERACTION SURFACE

- SIGNIFICANT MODELING IMPROVEMENT

- DATA AVAILABLE

- IMPLEMENTATION PRACTICAL AND AVAILABLE

ANDREW PHASE II

JOINT MODEJJNG OPTIONS ILLUSTRATIONS

P

e

ANDREW PHASE II

JOINT MODELING PHASE II METHOD

e

I I I

ANDREW PHASE II

ANDREW STORM CONDmONS UPDATED

bull -PHASE I USED OCEANWEATHER HINDCAST-(Nov 1992)

bull middotmiddotPHASE II WILL USE THE REVISED HINDCAST (Nov 1994)

bull lt c lt I lt

2-2

l---4-~~~+-~~-+-~~~1--~~-+-~~--l--~T--+-~~--i24

f-tJVtMfiz~ 1Cfl (r STvb(

Plollol bullbull 5middotNOYmiddot9214bull1J Irbullbull Illbull [ANDREVl~NDREVCUSTbullMAXll 5-NOY-1992 10bull54 - 9208

HEIGHT lml

I~ MAXI

I L 1~middotamp SZ~- ~A I 13c

I J) ( r I X~0iWJ~0gtJ gt L gt gt I 1c I 120

-- N

~ deg

I I I I I I e C I ~=-41 122 -94 -92 -90 -88 -86 -84 -82 -80

-

Figure 7 contours of maximum hindcast significant wave height

_

~

~tl ~

NOltNttIJ 17lt ltNS S iJJ)y

Maximum Significant Wave Height - Hurricane Andrew 1992

30 ~

26 ~ I ~-~~~- == -shy -~ 35-_

24deg1--shy --------~- middot-middot-------middotmiddotmiddot----middot------middot-middot~middot- - - ~-----1---middot

22middot ~~~~~~~-~ -94 -92 -90 -88 -84deg-86 -82 -somiddot

c

-

Platform Selection bullbullbull1111111111111

0 BasisCriteria for Selection middot

0 Review of Phase I Platforms middot 0 Alternative Platforms

0 PMB Recommendations

0 Participants Comments

gt

L

l I

BasisCriteria for Selection bullbullbull1111111111111

0 Benefit for Calibration - Classification (UnexpectedExpected)

- Failure Mode (StructuresFoundations)

- Damaged StructuresKnown response

- Better Representation of Full ~opulation

0 Availability of Data - Existing Models

- Inspection Data

- Damage Data

PREDICTED BEHAVIOR

-

OBSERVED

BEHAVIOR

-SURVIVED DAMAGED FAILED

SURVIVED

ST134W WD90A

-MC311 -

MC397 -

--

-

-

ST151K ST130Q

-

-

-

DAMAGED - -

-

-

--

-

-

ST52 SS139 ST177B -

ST161A

-_

-

FAILED

-

-

-

ST151H ST130A ST72

Operator - Platfonns - Platfonns Survived Damagedamp

- Falled -

bull

-

-

Jle eBDIGIPllDbl -Amocoshy 30 1 Chevron -

143 12 Exxon 72 Mobll 22 1 Phllllps 7 bull

-Shell 39 Unocal 25 1

Total Partlclnts Platfonns 338 15

Other Operators bull- 310 13

-

Total Platfonns - -

648 28

fbml

Platforms Considered 6 7

tie of Total Platfonns 1 25

OVERVIEW OF PLATFORMS SURVIVEDDAMAGEDFAILED

-

shy

v

Platforms Evaluated in Phase I

-

-Platfonn Water Year Number Analysis Avallabillty of Availability Platform

- Name Depth rl of Model DetaDed rl ClassUlcatlon

Installation legs Inspection Soll Report Based on Information Phase I Calibration - -ft

- -- Suaill flalfnrm ~BSi -STISIK 137 1963 8 CAP Yes Yes Un eel Sarvlval

-STIJOO 170 1964 4 CAPmiddotshy Yes ST134data u ST134W 137 1981 4 CAP Yes Yesmiddot E

-

edSarvlval -

middotsurvival

WD90A 184 - 1964 8 CAP Yes E -

- Sarvlvill

MC3ll 343 1978 8 KARMA

MC397 468 1991 4 KARMA -

bull ~reglmmiddot- -

Yes Sure Survival

bull Yes - Sare Sanlval-

-T23STS2 63 1969 4 CAP

T2SSS139 62 1969 4 CAP -

Yes

Yes

Yes Ex

Yes Une

bull SurvivalLikely to DamalI

middot bull SanlvalLlkely to DamalI

ST161A 118 1964 8 USFOS

Eail11a pound1al(bullllm Clss ST177B 142 --1965 8 CAP

STISIH 137 1964 - 8 CAP C

STIJOA 140 19S8 8 CAP - -

Yes

Yes

ollamed

Yes

- Yes E

-

at about 1 mUe

Yes

ST134data

bull SafvlvalLlkelv to Dam- bull- -

E middotrahre

Emeried Failure -

Likely FaUare

T21ST72 61 1969 4 CAP Collamed No Likely FaUure

Platforms Identified for Evaluating Their Likelihood of Falling Under UnexpeCted Survival Category

CllEllATOa

AllB

llLOCll

Sl1lvcnJllE

NAME

YEAR

INSTAUJCD

WATER

IJBPJB

middotmiddotDISTANCE

ftOMSllOllE

(MIUIS)

Ill

(M)

AMOCO BllW

STllI

WDI075

A

bull D

4

n uo 17J

44

32

bull 20

10

bull CllEVllOH STtllO c I 1bull 21 11

STl130 D 0 161 21 11 STllJ4 I 0 137 D - 11

STIUI G I 137 32 11

STIUI I 0 121 32 11

STIUI

J 0 140 32 11

STIUI It 4 137 32 11

STllSI PROl)I 0 137 32 11

STllSI PROD-2 0 137 32 11

STll7 A 0

140 35 11

WDl117 c 65 214 34 10

DXON STshy E 51 14 bull STl165 A t3 bull 34 WD0073

WD0073

WD0073

A _ c

4

62

4

IA IA

17J

22

22

22

bullbullbull WD003-shy E 65 161 Z7

A 0 IN 31 MOBD Bl12I

ssn A bull

50

50

31

7

bull 7

MIJllPllY SS114 B 7 50 7 7

PlllUIPS SSl14t A 55 33 7

SAMEDAN ST17J A 4 117 7 bull 7

SJIELL WD0103 A 65 223 27 WDllOJ bull 65 221 27 WDll04 c 65 221 27 WDOI04 D 67 2 27

UNOCAL

SSl20I

SSl20I

bull SSl20I

SS0209

SS0209

SS0209

SS0209

SS 0253

SS 0253

E

I

PtJllE

A

bull D

G A

c

0

4

4

7J

61

0

0

69

103

97

97

9S

100

95

100

165

175

34

34

34

34

34

34

34

54

54

bullbullbullbullbullbullbullbullbull

Platforms Recommended for Phase II

llalr y_ w- Aot Av- _ I 1 Doplla Model nm bull bull _ bull I I H11 s18-1 Irht lbwl~ bullbullbull(-illtdlel OM lllMtD

II ---

-STISIIC 137 1963 I CAP Yeo Yeo I Yeo

STtshy -170 1964 4 CAP Yeo ST134dlU I Yeo

STl34W 137 Yeo Yeo -1981 4 CAP D Yeo

WD90A 114 1964 I CAP Yeo - D

MC311 30 1971 I KARMA Yeo -sre-

MC397 middot 461 1991 4 ICARMA Yeo Sireshy - - r- -

__ T13ST5l 63 1969 4 CAP Yeo Yeobull middot- ~n- D Yeo _ Tl5SSl39 62 1969 4 CAP Yeo Yeo I Yeo

_ STIT711 142 1965 I CAP Yeo I Yeo middot-1shySTl61A Ill 1964 I USFOS Yeo Yeo - D bulloshy

- r-

STISIH 137 1964 I CAP y D y -_ STl31A 140 1951 I CAP Yeo STl34dlU I Yeo

T21Sll 61 1969 4 CAP No rar-middot I Yeo

~-rmiddot-middot SS209A 95 1967 16 CAP Yeo D -SSU4R

11 UI A 41 -in D SS 2741AIM CAP bull

_ _ -shy - -shy- -shy

CALIBRATION

bull PHASEIAPPROACH

bull CALIBRATION ISSUES

bull PHASE II WORK SCOPE

bull SOFlWARE AVAILABILITY

bull EFFECT OF ADDITIONAL PLATFORMS

bull WEIGHTING APPROACH

bull EXAMPLE APPLICATION middot

(l=Ofce Mode] Wind

~

fcurrent l -middot

~SW ~leldl4

ultimate resistance force I first

design load Illgt

i _

deflection

-

bull

4

Historical Hurrlcane Dariiage and Survival of Platforms

Components Required for Calibration (Petrauskas 1992)

PRIOR Distribution

PLATFORM EXPERIENCE

BAYESIAN Survivals and Failures

UPDATING

Waves Winds and Currents bull

POSTERIOR Distribution

- A_middotJA ~ Survival Data Failure Data

Result Result

Load Resistance

middotfl

-

Bayesian Updaing middotGeneral Description (Petrauskas 1992)

L Imiddot

WHERE TO INTRODUCE BIAS FACTOR (B

On Resistance (Rmue - B (RPREDICreD I

middotOn Loads (Smue middot - B (SPRE01cTEo

On Safety Margin (RSmue - B (RSPREDICTED

CALIBRATION METHODOLOGY

gt

r

FIELD OBSERVATIONS

Observed Behavior

survivals Damages

Failures

r

bull

CAPACITY ANALYSIS

Predicted Load and Resistance Predicted Failure Modes

Survival Damage Failure cases

RELIABILITY ANALYSIS

- Probability of Failure ~

Ukellhood Failure

BAYESIAN UPDATING

Prior Distnbution of B Combined Likelihood Function

Posterior Dlstnbution of B

BIAS FACTOR

Mean Value cov bull

Represents Modeling Uncertainties

Prior Distribution of B Likelihood Functionmiddot Observing Success case STISIK and Posterior Distribution of B

2 00

uo

160

140

120

e 100 L $ uoIshy

0 60 ~ ~ 040

020

0 00

0 00 0 50

_ I 00

b

I 50 2 00

-shy lk [bl STISIK)

--shy Pnor-f(b)

--shy Pos1enor r(bl ISIK)

Prior or B Mean bull 10 COV bull 030

Posterlor of Bbull Meanbull 136COV 0 16

2 50

Likelihood Function and Posterior Distribution Example Platform STISIK

- -

~

- Prior and Posterior Distributions or B Failure cases (ST177B STlSlH ST130A T21) Damage cases (T23 T25 ST161A) Success cases (STlSlK ST130Q ST134W WD90A MC311 MC397)

400

350

300

2 so 0-

I

bull bull

bull bull gt

- bullbull ~ bull bull I bull

_ bull bull bullbull I bull bull bullII V -

bull bull

I bull ~ bull

bullbull bullI bull bullI bull bull bull

middot -~

bull__ ~ middotI middotmiddot

L 2 00

e ~ I SO

-I 00

- 0 so

000 gt

0 00 020 0 40 060 0 80 I 00 120 140 160 110 200

b -

-

shy

Prior-r

- - - - Posterior fLlll failure cases

bull bull bull bull bull bull bull Posterior flall damage cases

- bull bull - bull middot Postenor flall success cases

Posterior flall success fulure damage cases

-

Posterior of B Mean= 119 COV 0101

Posterior Distribution or Blas Factor (8) bull Different Categories and All Cases Combined

shy

l

PHASE I - KEY CONCLUSIONS

bull GLOBAL CORRECTION FACTOR

bull 15 TO 20 CONSERVATISM MEAN)

bull UNEXPECTED SURVIVALS HAVE GREATEST IMPACT

bull DAMAGED CASES MORE INFORMATIVE

bull CALIBRATION ISSUES PHASE II

bull

EFFECT OF IMPROVED CAPACITY ANALYSIS RECIPE

bull APPLICABILITY TO MULTIPLE FAILURE MODES

bull NO FOUNDATION DAMAGE CASE INCLUDED

bull IMPROVED REPRESENTATION OF FULL ANDREW DISTRIBUTION

-bull NUMBER OF PLATFORMS CONSIDERED

I 1 I

I t ) bull ~

PHASE II WORK SCOPE (CALIBRATION)

bull

bull ESTABLISH MULTIPLE BIAS FACTORS

Use Nine Base Case Platforms

middot - Capacity Analysis for Revised recipe

Foundation Blas Factor from Foundation JIP

bull DEVELOP WEIGHTING APPROACH

bull UPDATE MULTIPLE BIAS FACTORS

bullEffect of Platforms Not In the Nine Base Cases

bull EXAMPLE APPLICATION OF BIAS FACTORS I

bullbull

PHASE I PHASE II

CAPACITYbull ANALYSISRu bull

---------------------1----middot

Jc1 - iJbullraquoJ dbull

FttlS frt[f~l ratdhJJ

-

LIKELIHOOD FUNCTION

DEVELOPMENT

lk Jb Jfall1N) bull Jb)

111 Ill Isurvtvoq bull 1 - bl

-middot -------------------shy ~---------------------middot-middot

r111ir8 Cb1 111 Cbl BAYESIAN

UPDATING

~cove

R1R2 bull middot---------------------~----

Jblbull P(g cOJ

1 bull JbRI (MBSI I bull 1

~middot

llcJllJfalluro) bull P(gcOJ Ill 111 b21fallun) bull P ((g1 CO) U 1112 c OD

middot--------------------shy ----------------------shy bull

r111111r8 Cbl 111 Ill i-1nro1

deg ~1b2)laquottfrr (b1b21-lnfo) re Ill bull 1-11middot b2) bull lodegamp

middot

i1 covamp i2 cov~i

-

shy

--

Likelihood Function vs b All Success Case STISI K STl30Q STl34W WD90A MC311 MC397

I 0000

0 9000

OIOOO

0 7000

~ 0amp000

ta osoOO

~ 0 4000

0 3000

0 2000

0 1000

bull 0 0000

bull

I

middot

middot bull I

Ibull

-sibull ~ -middot-middot-middot-middot -middot-

----7----+-----~ I

Ibull bull bull I

rbull

--- Jk ISIK[blsuccess)

- bull - bull - lk 130Q[blsuccess) bull

lk 134W[blsuccess)

Jk-WD90A(blsuccc~I

- - - lk MC397[blsucccss)

--ltgt-- lk middotMC31 l[blsucccss)

--- Jk [bl 6 out of6succcsHascs)

05 I 5 2 250

b

Likelihood Functionsmiddot Success Cases - Phbullsc L

shy

bullbull bull bull bull bull bull -------------------~

bull bull bullbull bullbull bullbull bullbull bullbull bullbull bull

bull bullbull

bullbullbullbull

bullbull

bullbullbullbullbullbullbull

bullbullbullbullbullbull

bullbullbullbullbullbullbull

z 0

t z ) LL

c 0 0c i ~ i

J

~ sect ~--------------~ e l5

I

I

SOFTWARE

bull PF PROGRAM

bull CALREL

bull PF +CALREL

bull RELACS

bull

gt i

l

SENSITIVITY STUDIES

POSTERIOR DISTRIBUTION OF B

bull EFFECT OF CHANGE IN DISTRIBUTION OF PRIOR OF B

bull EFFECT OF REMOVAL OF ONE PLATFORM FROM EACH CATEGORY

bull EFFECT OF REMOVAL OF THREE DAMAGE CASES

bull EFFECT OF ADDITIONAL 13 SIMILAR PLATFORMS

-I

Prior and Posterior Distributions of B Variation In COV of PriorofB fromOl to04 -- -

4500

-4000

-

- -

3500 --middot - bull bull - bull bull Prior-r ICOV =02

3000

- - - - Prior-r ICOV =03 -

bull -~ e 2soo -

bull bull bull bull bull bull bull Pnor-r ICOV =o 4t frJ ~ 2000 Posterior rlCOV =0 2

bull bull Postenor rlCOV =0 3 bull bull 1500 bull middot bull - shy -middot- Posterior rlCOV =04

middot~ I 000 - bull

-middotmiddot ~ ~middotmiddot M -~

- middot 0 500

wbullbullbull _ I middot~ middotmiddotmiddotmiddotmiddot middotmiddotmiddotmiddotmiddot - -~ ~ __ middot_ _ bull -shy0 000

0 000 0200 0 400 0 600 0 800 IOOO 1200 1400 1600 1800 2000 -

- - b

-

=shy

Posterior Distribution of Bias Factor (B) bull Effect of Variation In COV of Prior of B

bull

bullPrior and Posterior Distributions of B Evaluating Effect of Removal of One Platform from Each Category from the Base Case of 13 Platforms

400

~ ~ 350

j IA i

3 00 bull I bull I ---Prior-r I bull l

250 1 I 1

I bull bull ---Posterior r lall 13 casesshy- ~E 2 00 bull bull bull bull bull bull bull Posterior r without STISIK

- ~ 1 bull - bull I - bull bull - bull bull Posterior r Iwithout T2I I 50 I

J-tt~i - bull - bull - Posterior r Iwithout T2S

tOO [-

I I ~ ~ I 0 50

0 00

___ r J i

l~T

~--

0 00 0 20 0 40 0 60 080 I 00 120 140 160 110 2 00

b

zshy

Posterior Distribution of Bias Factor (8) bull Effect of Removal of middotPlatform from Any Category

bullPrior and Posterior Distributions or B Evaluating Effect or Removal or 3 Damage Cases from the Base Case or 13 Platrorms

4

35

3

25

e L 2 ~ bulle shy

I 15

bull I

05

0

0 02 04 06 I 2 I 4 16 IS 2

I I

I

I

I Ibull

J -tmiddot L--

bullI

~

I

fl

II ~

I I v I

middot bull bull

bull L

~ ll- I-~lt

OS I

b

I

bull

1middot bullbullbullbull

~ __

bull

- - - - Prior-r

- - - - Postenor flall failure cases

Postenor flall success cases

--- Posterior flall 13 cases

---Posterior fI success failure cases only

IPosterior or B Mean =t11 COV =0121

-

_I

Posterir Distribution or Bias Factor (B) bull Effect or Removal or 3-Damage Cases

-

bullPrior and Posterior Distributions or B Evaluating Effect or Additional 13 Similar Observations as

Used in Andrew JIP The POsterior or B from Andrew-JIP Considered as New Prior or B

- 500

450

400

350

bull bull

bull bull bull bull J bull

bull bull f ~ I

I ~ Ibull bull

bullbull bull

l bull I -bullbull bull bull

I

I bull bull

bull I I~~bull bull bull

bull ~

I bull bull

~ ~_ ~~~

- - New Prior r bull

Posterior 114 new failure cases ~ 300- bull bull bull bull bull bull bull Posterior f 13 new damage casesl 150 ~

- bull bull - bull bull Posterior 116 new success cases i 2 00

Postenor 1113 new success failure -150 damage cases

I 00

0 50 -

0 00 0 00 0 20 0 40 0 60 0 80 I 00 120 I 40 160 110 2 00

Posterior ol B Mean= 120 COV= 0071 b

shy

shy

_rshy

Posterior Distribution of Bias Factor (B) bull Effect of Additional 13-Similar Observations

1

1 I

WEIGHTING APPROACH bull

bull ASSESSMENT OF PLATFORMS DATA

Select Platforms Most Mappable to the Analyzed Platforms

Unlikely SurvivalDamage Cases Identification

Based on Cut-Off Criteria Agemiddot Hs etc

bull INFORMATION REQUIRED FOR COMPARISON

bull CATEGORIZATION OF ADDITIONALmiddot PLATFORMS

bull UPDATING OF CALIBRATION RESULTS

--

OVERVIEW OF PLATFORMS SURVIVEDDAMAGED -~

Operator -

- Platforms

Survived CS)

Platforms

Damaged CD)

Platforms Used In

Phase IJIP -

JI~ ~Bill~IRBDll -

Amoco -

Chevron -

Exxon

Mob II

Phllllps

Shell

Unocal -

30

143

72 22

7

39 25shy -

1

12

bull

1 bull

1

1(S)+1 (D)

3 (S) +3(D)

2 (S)

bull shy

bull

bull

bull

Total Partlclnants Platforms 338 - 15 10C6S+4D) -

Other Operators

310

13

3 (D)

Total Platforms 648 28 13C6S+7D) -

Summary of Andrew JIP bull Phase I Platforms

Plairshy w v- N-ber Lea-le 1 shy car Hind_ o-w N-bull Deplh fl fl IJpe la bull n Yeadon Jadtet lbelledtf liefalit -- _ -OD

IL H - -shy _ICllshy IL IL - - rQClto ~ _

K STUIK 137 1963 8 365 608S4 Cellardeck

-middot STl-shy 170 1964 4 -~ K 39 62269 Cellardeck - --ST134W 137 1911 4 43 608S4 Smvlnlshy- _ - -

WD90A 184 1964 8 K 267Sbull34 83 SOS31 SUboftllar deck SmvMI

MC311 343 1978 8 x S7 62 - MC397 468 1991 4 x shy S33bull6SO 65911 -shySmeSmvlvol -- gt -

r-- T23ST52 63 1969 4 ton- shyKshy 52594 S03 -T25SSl39 62 1969 4 Kshy SOS94 S063 - middot ton-shy__ -

bullbull a ton--shyST161A 118 1964 8 K - 347S 51495 Cellarclecti

r~

ST1771 142 1965 8 K 39 60167 CeDordeck failme

STISIH 137 1964 8 K 395 608S4 CeDardeck Failme

K ST130A 140 1958 a SS 6095 I lblv Pai1uno - T21ST72 61 1969 4 Kshy S4S94 4974 I lhlv Pai1uno -

- -

JE

INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON

bull WATER DEPTH

bull bull YEAR OF INSTALLATIONDESIGN

bull PLATFORM ORIENTATION

bull PLATFORM LATITUDE AND LONGITUDE

bull KEY SKETCHES OF PLATFORM STRUCTURAL FRAMING

Number of Legs Plies Leg Batter Number of Conductors bull Deck Elevations (Including sub-cellar deck If any) Jacket Framing Elevations Brace type In Vertical Frames Sizes of Primary Members (leg pile vertical frames braces primary horizontals) Joint Details (can gap etc) LegPiie Annulus Grouting Status

bull middotPOST ANDREW INSPECTION REPORT Significant Wave Height Hs (ft)

bull INFORMATION ON DAMAGES PRIOR TO ANDREW

bull RESULTS FROM IN-HOUSE STUDIES IF AVAILABLE

bull ~RIGINAL DESIGN CRITERIA INFORMATION IF AVAILABLE

bull SOIL SHEAR STRENGTH PROFILE IF AVAILABLE

Platforms Identified for Evaluating Therr Likelihood of Falling Under Unexpected Survival Category

OPlRATOR

middotADA

BLOClt

llmucnJllE

NAME

YEAll

INSTAUJD

WA1ER

DBPTll

DISTANCE

JlllOM SHOllE

(MILU)

Bl

(M)

AMOCO 111113

STl11

WDI075

A

bull D

St

17

120

1n

44

20

10 I

CllllVllQfl STlllO c 61 110 21 11 STIUO D Q 1110 21 11

STl134 p Q 137 29 11

ST0151 G 61 137 11

ST0151 I 3 121 11

ST0151 J Q 140 11

STl151 It 137 11

ST0151 PJl001 Q UT 11

ST0151 PROD-2 Q UT 11

ST017 A 3 140 35 11

WDIUT c 65 214 34 10

llXXON STI055 II 51 6T 14 I

STl15 A t3 34

WDI073

WD0073

Amiddotshy Q

1111 1111

22

22

I

I

WDI073 c 1n 22 I

- WDOCl93-shy II

65 1110 2T A 3 194 31 MOBU 111121

SS IOl7

A B

501

50

50

31

T

I

T 6

MtlllPllY SS114 B T 50 T T

PlllllIPS SS0149 A 5 55 33 T

SAMllDAN STln A 107 T T

SBllLL WDl103 A 65 223 2T WDl103 B 65 221 2T WD0104 c 65 221 2T WD0104 D 6T 20 2T

UNOCAL SS020I

SS 020I SSl20I

SS 0209

SS0209

II p

PURll

A

B

3

n 61

103

97

97

95

100

34

34

34

34 34

bullbull bullbullbull

- SS0209 D u 95 34 I

- SS 1209 SSl253

SS0253

G A

c

6T

2

100

165

1T5

34

54

54

bullbull I

I I

I I

SOUTH MARCH

~LAND

N

SOUTH PaTOI ENE ISWltD AREA

MMS Defined Limits ror Offshore Structures Requiring Inspections Following Andrew

t---Jbulln3 T25bull I bullTtl

bull1$1Htl1Abull bull1SIK1779bull

~

middot-DaTA AREA~T

CRANO rsLE

AREA bull1300

bullll4W

=shy

bullPath of Hurricane Andrew with Platforms Used In Calibration

Likelihood Function vs b All Success Case STISI K STIJOQ ST134W WD90A MCJll MCJ97

I 0000

0 9000

0 8000

07000

-Bo6000

tiiii 0 5000 -

04000

0 3000

0 2000

01000

0 0000

0 05 25

ri1rt middot-middot-middot-middot-middot-

I 5 2

b

---11 middot151K(blsu~ccsI

- bull - bull - fk 130Q(bfU~lCSS)

lie bull134W(blsuccc]

lk-WD90A[blsultCS]

- - - lk middotMC397(bfUtlCS]

--o-- lk-MCJll(bbuctcsJ

--- ll (bl 6 out of 6 Mlttcs case]

_

( Likelihood Functions middotSuccess Cases - p - 1 c I

CATEGORIZATION OF ADDITIONAL PLATFORMS

I I I

bull CATEGORIES

Unexpected Survivals

Expected Survivals

bull APPROACH

Identify Corresponding Platform(s) Evaluated In the JIP

Determine Max Wave Height at Platform Location

Evaluate Whether Wave Hit the Deck

middot - Evaluate Likelihood of R and S to Differ from the Base Cases

Review Joint Information

Assess Special Features

bull IDENTIFY NUMBER OF PLATFORMS IN EACH CATEGORY

I

UPDATING OF CALIBRATION RESULTS

bull INCLUDE EFFECT OF ADDITIONAL PLATFORMS ON bullBbull

Consider Increased 1Uncertalntles In R and S

Develop bullNeW Ukellhood Functions for Different Categories

middotLet bullxbull Number of Ukely Unexpected Survival Cases and y Number of Ukely Expected Survival Cases

Determine Combined Likelihood Function lk (b Iadditional platforms) middot

bull REVISED LIKELIHOOD FUNCTION

lk(b I all cases) = lk(b I 9 platforms) middot x lk(b Ix platforms) x lk(b I y platforms)

bull ESTABLISH REVISED BIAS FACTOR B

f8 (b) u f8 (b) lk (b I all cases)

II

t t ~

shy

APPLICATIONS OF BIAS FACTORS

ii APPLICABLE TO PARTICULAR LOAD AND RESISTANCE RECIPE

bull USE BY THE API COMMITTEES AND MMS

Criteria Revisions

Recommended Design Guidelines Revisions bull

bullbull REASSESSMENT STUDIES AND DECISION MAKING

middot

bull RISK ASSESSMENT STUDIES BY THE OPERATORS

Individual Platforms

Fleet of Platforms

middot IMR Program Implementation

bullbull NOT TO BE USED IN DESIGN

I I fl J

EXAMPLE USE OF BIAS FACTORS

bull PHASE I EXAMPLE APPLICATION

bull EVALUATION OF ANNUAL PROBABILITY OF FAILURE

Establish Annual Maximum Seastate Data

Determine Coefficients to Define Base Shear

Determine Ultimate Capacity of Platform

Determine Probablllty of Failure Versus bullbbull

-bull DETERMINE UPDATED PROBABILITY OF FAILURE

Pf - P [S gt R]

Pf I P [~ gt Rb I b] f 8 (b) db

Pr - J P [S gt Rb I b] f bull 8 (b) db

bull TOTAL PROBABILITIES OF FAILURE

Total Pf - 1 bull Il [1 middotPf] - l Pf bull

Total Pf - 1 bull Il [1middot Pf] - l Pfmiddot

Total Pr = 1 - n [1middot PfbullJ = l Pfshy

I I I

EXAMPLE APPLICATION bull PHASE I

PLATFORM ST 151 K

bull GULF OF MEXICO

bull WATER DEPTH = 137 FT

bull ANNUAL MAXIMUM HS MEDIAN 175 ft COV 031

bull ASSUMED STORM DURATION 3 hoursmiddot

bull ASSUMED NUMBER OF WAVES 800

bull MEDIAN HS FOR STORM HOURS 1 2 AND 3

bull MEDIAN CURRENT VALUE bull

bull DETERMINE ULTIMATE CAPACITY FOR THREE DIRECTIONS

bull DETERMINE ANNUAL FAILURE PROBABILITIES

Broadsidemiddot Diagonal End-on

0037 0025 0007

0071 0044 0018

0022 0016 0004

DETERMINE TOTAL ANNUAL PROBABILITY OF FAILURE

Pf - 007

Pf - 013

Pr - 004

bull

EXAMPLE USE OF BIAS FACTORS - PHASE II middot

bull DETERMINE PREDICTIVE PROBABILITY OF FAILURE

Case - 1

Pf _ S S Pf (b b2) fa1a2 (b b2) db db2

g bull -1 (b1 R I S) - 1

92 - (b2 R2 I S) - 1

Case -2

Pf s ~ [bJ f11 (b) db

Pf [bJ p [g lt 0)

g - bull 1 (b1 R1 I S) - 1

1

bull1) l

I

BENEFITS OF ANDREW JIP bull PHASE II

lt

bull IMPROVED LOAD AND RESISTANCE RECIPE

bull IMPROVED CALIBRATION PROCEDURE

bull MULTIPLE BIAS FACTORS

bull PROCEDURE TO INCLUDE EFFECT OF OTHER PLATFORMS

bull PROCEDURE TO USE MULTIPLE BIAS FACTORS

  • Structure Bookmarks
    • ANDREW PHASE II JOINT1 INDUSTRY PROJECT
    • PROGRESS MEETING NUMBER 1
    • ANDREW PHASE II JOINT INDUSTRY PROJECT PROGRESS MEETING NUMBER 1 NOVEMBER 16 AGENDA
    • ANDREW PHASE II UST OF PARTICIPANTS
    • ANDREW PHASE I OBJECTIVES AND RESULTS
    • ANDREW PHASE II PROJECT OBJECTIVES
    • SUBSEQUENT PROJECTS
    • ANDREW PHASE II PROJECT TEAM
    • ANDREW PHASE II UPDATE MODELING middotAND ANALYSIS RECIPE
    • -ANDREW PHASE II CONDUCTOR RESISTANCE
    • ANDREW PHASE Il WA VE FORCE PROFILE
    • ANDREW PHASE II WA VE FORCE PROFILEmiddot
    • ANDREW PHASE II BRACE MODELING
    • ANDREW PHASE II FOUNDATIONS
    • ANDREW PHASE II middot middot JOINT MODELING
    • ANDREW PHASE Il JOINT MODELING OPTIONS DISCUSSION
    • ANDREW PHASE II JOINT MODEJJNG OPTIONS ILLUSTRATIONS
    • ANDREW PHASE II JOINT MODELING PHASE II METHOD
    • ANDREW PHASE II ANDREW STORM CONDmONS UPDATED
    • Platform Selection bullbullbull1111111111111
    • BasisCriteria for Selection bullbullbull1111111111111
    • CALIBRATION
    • WHERE TO INTRODUCE BIAS FACTOR (B
    • PHASE I -KEY CONCLUSIONS
    • bull CALIBRATION ISSUES PHASE II
    • PHASE II WORK SCOPE (CALIBRATION)
    • SOFTWARE
    • SENSITIVITY STUDIES POSTERIOR DISTRIBUTION OF B
    • WEIGHTING APPROACH bull
    • INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON
    • CATEGORIZATION OF ADDITIONAL PLATFORMS
    • UPDATING OF CALIBRATION RESULTS
    • APPLICATIONS OF BIAS FACTORS
    • EXAMPLE USE OF BIAS FACTORS
    • EXAMPLE APPLICATION bull PHASE I PLATFORM ST 151 K
    • EXAMPLE USE OF BIAS FACTORS -PHASE II middot
    • BENEFITS OF ANDREW JIP bull PHASE II
Page 36:  · . ' ANDREW PHASE II PROJECT OBJECTIVES . ' • ·UPDATE MODELING AND ANALYSIS RECIPE· JOINTS ' ' ·BRACES WAVE PROFILE FOUNDATIONS . ' ' CONDUCTORS· • ANALVZE PLATFORMS USING

J Ii

ANDREW PHASE Il

JOINT MODELING OPTIONS DISCUSSION

bull JD COUPLED INTERACTION SURFACE

- IDEAL

- DATA NOT AVAILABLE TO CALIBRATE

bull 2D COUPLED INTERACTION SURFACE

- SIGNIFICANT MODELING IMPROVEMENT

- DATAAVAILABLE

- IMPLEMENTATION PRACTICAL BUT NEEDS DEVELOPMENT AND CALIBRATION

bull 2D UNCOUPLED INTERACTION SURFACE

- SIGNIFICANT MODELING IMPROVEMENT

- DATA AVAILABLE

- IMPLEMENTATION PRACTICAL AND AVAILABLE

ANDREW PHASE II

JOINT MODEJJNG OPTIONS ILLUSTRATIONS

P

e

ANDREW PHASE II

JOINT MODELING PHASE II METHOD

e

I I I

ANDREW PHASE II

ANDREW STORM CONDmONS UPDATED

bull -PHASE I USED OCEANWEATHER HINDCAST-(Nov 1992)

bull middotmiddotPHASE II WILL USE THE REVISED HINDCAST (Nov 1994)

bull lt c lt I lt

2-2

l---4-~~~+-~~-+-~~~1--~~-+-~~--l--~T--+-~~--i24

f-tJVtMfiz~ 1Cfl (r STvb(

Plollol bullbull 5middotNOYmiddot9214bull1J Irbullbull Illbull [ANDREVl~NDREVCUSTbullMAXll 5-NOY-1992 10bull54 - 9208

HEIGHT lml

I~ MAXI

I L 1~middotamp SZ~- ~A I 13c

I J) ( r I X~0iWJ~0gtJ gt L gt gt I 1c I 120

-- N

~ deg

I I I I I I e C I ~=-41 122 -94 -92 -90 -88 -86 -84 -82 -80

-

Figure 7 contours of maximum hindcast significant wave height

_

~

~tl ~

NOltNttIJ 17lt ltNS S iJJ)y

Maximum Significant Wave Height - Hurricane Andrew 1992

30 ~

26 ~ I ~-~~~- == -shy -~ 35-_

24deg1--shy --------~- middot-middot-------middotmiddotmiddot----middot------middot-middot~middot- - - ~-----1---middot

22middot ~~~~~~~-~ -94 -92 -90 -88 -84deg-86 -82 -somiddot

c

-

Platform Selection bullbullbull1111111111111

0 BasisCriteria for Selection middot

0 Review of Phase I Platforms middot 0 Alternative Platforms

0 PMB Recommendations

0 Participants Comments

gt

L

l I

BasisCriteria for Selection bullbullbull1111111111111

0 Benefit for Calibration - Classification (UnexpectedExpected)

- Failure Mode (StructuresFoundations)

- Damaged StructuresKnown response

- Better Representation of Full ~opulation

0 Availability of Data - Existing Models

- Inspection Data

- Damage Data

PREDICTED BEHAVIOR

-

OBSERVED

BEHAVIOR

-SURVIVED DAMAGED FAILED

SURVIVED

ST134W WD90A

-MC311 -

MC397 -

--

-

-

ST151K ST130Q

-

-

-

DAMAGED - -

-

-

--

-

-

ST52 SS139 ST177B -

ST161A

-_

-

FAILED

-

-

-

ST151H ST130A ST72

Operator - Platfonns - Platfonns Survived Damagedamp

- Falled -

bull

-

-

Jle eBDIGIPllDbl -Amocoshy 30 1 Chevron -

143 12 Exxon 72 Mobll 22 1 Phllllps 7 bull

-Shell 39 Unocal 25 1

Total Partlclnts Platfonns 338 15

Other Operators bull- 310 13

-

Total Platfonns - -

648 28

fbml

Platforms Considered 6 7

tie of Total Platfonns 1 25

OVERVIEW OF PLATFORMS SURVIVEDDAMAGEDFAILED

-

shy

v

Platforms Evaluated in Phase I

-

-Platfonn Water Year Number Analysis Avallabillty of Availability Platform

- Name Depth rl of Model DetaDed rl ClassUlcatlon

Installation legs Inspection Soll Report Based on Information Phase I Calibration - -ft

- -- Suaill flalfnrm ~BSi -STISIK 137 1963 8 CAP Yes Yes Un eel Sarvlval

-STIJOO 170 1964 4 CAPmiddotshy Yes ST134data u ST134W 137 1981 4 CAP Yes Yesmiddot E

-

edSarvlval -

middotsurvival

WD90A 184 - 1964 8 CAP Yes E -

- Sarvlvill

MC3ll 343 1978 8 KARMA

MC397 468 1991 4 KARMA -

bull ~reglmmiddot- -

Yes Sure Survival

bull Yes - Sare Sanlval-

-T23STS2 63 1969 4 CAP

T2SSS139 62 1969 4 CAP -

Yes

Yes

Yes Ex

Yes Une

bull SurvivalLikely to DamalI

middot bull SanlvalLlkely to DamalI

ST161A 118 1964 8 USFOS

Eail11a pound1al(bullllm Clss ST177B 142 --1965 8 CAP

STISIH 137 1964 - 8 CAP C

STIJOA 140 19S8 8 CAP - -

Yes

Yes

ollamed

Yes

- Yes E

-

at about 1 mUe

Yes

ST134data

bull SafvlvalLlkelv to Dam- bull- -

E middotrahre

Emeried Failure -

Likely FaUare

T21ST72 61 1969 4 CAP Collamed No Likely FaUure

Platforms Identified for Evaluating Their Likelihood of Falling Under UnexpeCted Survival Category

CllEllATOa

AllB

llLOCll

Sl1lvcnJllE

NAME

YEAR

INSTAUJCD

WATER

IJBPJB

middotmiddotDISTANCE

ftOMSllOllE

(MIUIS)

Ill

(M)

AMOCO BllW

STllI

WDI075

A

bull D

4

n uo 17J

44

32

bull 20

10

bull CllEVllOH STtllO c I 1bull 21 11

STl130 D 0 161 21 11 STllJ4 I 0 137 D - 11

STIUI G I 137 32 11

STIUI I 0 121 32 11

STIUI

J 0 140 32 11

STIUI It 4 137 32 11

STllSI PROl)I 0 137 32 11

STllSI PROD-2 0 137 32 11

STll7 A 0

140 35 11

WDl117 c 65 214 34 10

DXON STshy E 51 14 bull STl165 A t3 bull 34 WD0073

WD0073

WD0073

A _ c

4

62

4

IA IA

17J

22

22

22

bullbullbull WD003-shy E 65 161 Z7

A 0 IN 31 MOBD Bl12I

ssn A bull

50

50

31

7

bull 7

MIJllPllY SS114 B 7 50 7 7

PlllUIPS SSl14t A 55 33 7

SAMEDAN ST17J A 4 117 7 bull 7

SJIELL WD0103 A 65 223 27 WDllOJ bull 65 221 27 WDll04 c 65 221 27 WDOI04 D 67 2 27

UNOCAL

SSl20I

SSl20I

bull SSl20I

SS0209

SS0209

SS0209

SS0209

SS 0253

SS 0253

E

I

PtJllE

A

bull D

G A

c

0

4

4

7J

61

0

0

69

103

97

97

9S

100

95

100

165

175

34

34

34

34

34

34

34

54

54

bullbullbullbullbullbullbullbullbull

Platforms Recommended for Phase II

llalr y_ w- Aot Av- _ I 1 Doplla Model nm bull bull _ bull I I H11 s18-1 Irht lbwl~ bullbullbull(-illtdlel OM lllMtD

II ---

-STISIIC 137 1963 I CAP Yeo Yeo I Yeo

STtshy -170 1964 4 CAP Yeo ST134dlU I Yeo

STl34W 137 Yeo Yeo -1981 4 CAP D Yeo

WD90A 114 1964 I CAP Yeo - D

MC311 30 1971 I KARMA Yeo -sre-

MC397 middot 461 1991 4 ICARMA Yeo Sireshy - - r- -

__ T13ST5l 63 1969 4 CAP Yeo Yeobull middot- ~n- D Yeo _ Tl5SSl39 62 1969 4 CAP Yeo Yeo I Yeo

_ STIT711 142 1965 I CAP Yeo I Yeo middot-1shySTl61A Ill 1964 I USFOS Yeo Yeo - D bulloshy

- r-

STISIH 137 1964 I CAP y D y -_ STl31A 140 1951 I CAP Yeo STl34dlU I Yeo

T21Sll 61 1969 4 CAP No rar-middot I Yeo

~-rmiddot-middot SS209A 95 1967 16 CAP Yeo D -SSU4R

11 UI A 41 -in D SS 2741AIM CAP bull

_ _ -shy - -shy- -shy

CALIBRATION

bull PHASEIAPPROACH

bull CALIBRATION ISSUES

bull PHASE II WORK SCOPE

bull SOFlWARE AVAILABILITY

bull EFFECT OF ADDITIONAL PLATFORMS

bull WEIGHTING APPROACH

bull EXAMPLE APPLICATION middot

(l=Ofce Mode] Wind

~

fcurrent l -middot

~SW ~leldl4

ultimate resistance force I first

design load Illgt

i _

deflection

-

bull

4

Historical Hurrlcane Dariiage and Survival of Platforms

Components Required for Calibration (Petrauskas 1992)

PRIOR Distribution

PLATFORM EXPERIENCE

BAYESIAN Survivals and Failures

UPDATING

Waves Winds and Currents bull

POSTERIOR Distribution

- A_middotJA ~ Survival Data Failure Data

Result Result

Load Resistance

middotfl

-

Bayesian Updaing middotGeneral Description (Petrauskas 1992)

L Imiddot

WHERE TO INTRODUCE BIAS FACTOR (B

On Resistance (Rmue - B (RPREDICreD I

middotOn Loads (Smue middot - B (SPRE01cTEo

On Safety Margin (RSmue - B (RSPREDICTED

CALIBRATION METHODOLOGY

gt

r

FIELD OBSERVATIONS

Observed Behavior

survivals Damages

Failures

r

bull

CAPACITY ANALYSIS

Predicted Load and Resistance Predicted Failure Modes

Survival Damage Failure cases

RELIABILITY ANALYSIS

- Probability of Failure ~

Ukellhood Failure

BAYESIAN UPDATING

Prior Distnbution of B Combined Likelihood Function

Posterior Dlstnbution of B

BIAS FACTOR

Mean Value cov bull

Represents Modeling Uncertainties

Prior Distribution of B Likelihood Functionmiddot Observing Success case STISIK and Posterior Distribution of B

2 00

uo

160

140

120

e 100 L $ uoIshy

0 60 ~ ~ 040

020

0 00

0 00 0 50

_ I 00

b

I 50 2 00

-shy lk [bl STISIK)

--shy Pnor-f(b)

--shy Pos1enor r(bl ISIK)

Prior or B Mean bull 10 COV bull 030

Posterlor of Bbull Meanbull 136COV 0 16

2 50

Likelihood Function and Posterior Distribution Example Platform STISIK

- -

~

- Prior and Posterior Distributions or B Failure cases (ST177B STlSlH ST130A T21) Damage cases (T23 T25 ST161A) Success cases (STlSlK ST130Q ST134W WD90A MC311 MC397)

400

350

300

2 so 0-

I

bull bull

bull bull gt

- bullbull ~ bull bull I bull

_ bull bull bullbull I bull bull bullII V -

bull bull

I bull ~ bull

bullbull bullI bull bullI bull bull bull

middot -~

bull__ ~ middotI middotmiddot

L 2 00

e ~ I SO

-I 00

- 0 so

000 gt

0 00 020 0 40 060 0 80 I 00 120 140 160 110 200

b -

-

shy

Prior-r

- - - - Posterior fLlll failure cases

bull bull bull bull bull bull bull Posterior flall damage cases

- bull bull - bull middot Postenor flall success cases

Posterior flall success fulure damage cases

-

Posterior of B Mean= 119 COV 0101

Posterior Distribution or Blas Factor (8) bull Different Categories and All Cases Combined

shy

l

PHASE I - KEY CONCLUSIONS

bull GLOBAL CORRECTION FACTOR

bull 15 TO 20 CONSERVATISM MEAN)

bull UNEXPECTED SURVIVALS HAVE GREATEST IMPACT

bull DAMAGED CASES MORE INFORMATIVE

bull CALIBRATION ISSUES PHASE II

bull

EFFECT OF IMPROVED CAPACITY ANALYSIS RECIPE

bull APPLICABILITY TO MULTIPLE FAILURE MODES

bull NO FOUNDATION DAMAGE CASE INCLUDED

bull IMPROVED REPRESENTATION OF FULL ANDREW DISTRIBUTION

-bull NUMBER OF PLATFORMS CONSIDERED

I 1 I

I t ) bull ~

PHASE II WORK SCOPE (CALIBRATION)

bull

bull ESTABLISH MULTIPLE BIAS FACTORS

Use Nine Base Case Platforms

middot - Capacity Analysis for Revised recipe

Foundation Blas Factor from Foundation JIP

bull DEVELOP WEIGHTING APPROACH

bull UPDATE MULTIPLE BIAS FACTORS

bullEffect of Platforms Not In the Nine Base Cases

bull EXAMPLE APPLICATION OF BIAS FACTORS I

bullbull

PHASE I PHASE II

CAPACITYbull ANALYSISRu bull

---------------------1----middot

Jc1 - iJbullraquoJ dbull

FttlS frt[f~l ratdhJJ

-

LIKELIHOOD FUNCTION

DEVELOPMENT

lk Jb Jfall1N) bull Jb)

111 Ill Isurvtvoq bull 1 - bl

-middot -------------------shy ~---------------------middot-middot

r111ir8 Cb1 111 Cbl BAYESIAN

UPDATING

~cove

R1R2 bull middot---------------------~----

Jblbull P(g cOJ

1 bull JbRI (MBSI I bull 1

~middot

llcJllJfalluro) bull P(gcOJ Ill 111 b21fallun) bull P ((g1 CO) U 1112 c OD

middot--------------------shy ----------------------shy bull

r111111r8 Cbl 111 Ill i-1nro1

deg ~1b2)laquottfrr (b1b21-lnfo) re Ill bull 1-11middot b2) bull lodegamp

middot

i1 covamp i2 cov~i

-

shy

--

Likelihood Function vs b All Success Case STISI K STl30Q STl34W WD90A MC311 MC397

I 0000

0 9000

OIOOO

0 7000

~ 0amp000

ta osoOO

~ 0 4000

0 3000

0 2000

0 1000

bull 0 0000

bull

I

middot

middot bull I

Ibull

-sibull ~ -middot-middot-middot-middot -middot-

----7----+-----~ I

Ibull bull bull I

rbull

--- Jk ISIK[blsuccess)

- bull - bull - lk 130Q[blsuccess) bull

lk 134W[blsuccess)

Jk-WD90A(blsuccc~I

- - - lk MC397[blsucccss)

--ltgt-- lk middotMC31 l[blsucccss)

--- Jk [bl 6 out of6succcsHascs)

05 I 5 2 250

b

Likelihood Functionsmiddot Success Cases - Phbullsc L

shy

bullbull bull bull bull bull bull -------------------~

bull bull bullbull bullbull bullbull bullbull bullbull bullbull bull

bull bullbull

bullbullbullbull

bullbull

bullbullbullbullbullbullbull

bullbullbullbullbullbull

bullbullbullbullbullbullbull

z 0

t z ) LL

c 0 0c i ~ i

J

~ sect ~--------------~ e l5

I

I

SOFTWARE

bull PF PROGRAM

bull CALREL

bull PF +CALREL

bull RELACS

bull

gt i

l

SENSITIVITY STUDIES

POSTERIOR DISTRIBUTION OF B

bull EFFECT OF CHANGE IN DISTRIBUTION OF PRIOR OF B

bull EFFECT OF REMOVAL OF ONE PLATFORM FROM EACH CATEGORY

bull EFFECT OF REMOVAL OF THREE DAMAGE CASES

bull EFFECT OF ADDITIONAL 13 SIMILAR PLATFORMS

-I

Prior and Posterior Distributions of B Variation In COV of PriorofB fromOl to04 -- -

4500

-4000

-

- -

3500 --middot - bull bull - bull bull Prior-r ICOV =02

3000

- - - - Prior-r ICOV =03 -

bull -~ e 2soo -

bull bull bull bull bull bull bull Pnor-r ICOV =o 4t frJ ~ 2000 Posterior rlCOV =0 2

bull bull Postenor rlCOV =0 3 bull bull 1500 bull middot bull - shy -middot- Posterior rlCOV =04

middot~ I 000 - bull

-middotmiddot ~ ~middotmiddot M -~

- middot 0 500

wbullbullbull _ I middot~ middotmiddotmiddotmiddotmiddot middotmiddotmiddotmiddotmiddot - -~ ~ __ middot_ _ bull -shy0 000

0 000 0200 0 400 0 600 0 800 IOOO 1200 1400 1600 1800 2000 -

- - b

-

=shy

Posterior Distribution of Bias Factor (B) bull Effect of Variation In COV of Prior of B

bull

bullPrior and Posterior Distributions of B Evaluating Effect of Removal of One Platform from Each Category from the Base Case of 13 Platforms

400

~ ~ 350

j IA i

3 00 bull I bull I ---Prior-r I bull l

250 1 I 1

I bull bull ---Posterior r lall 13 casesshy- ~E 2 00 bull bull bull bull bull bull bull Posterior r without STISIK

- ~ 1 bull - bull I - bull bull - bull bull Posterior r Iwithout T2I I 50 I

J-tt~i - bull - bull - Posterior r Iwithout T2S

tOO [-

I I ~ ~ I 0 50

0 00

___ r J i

l~T

~--

0 00 0 20 0 40 0 60 080 I 00 120 140 160 110 2 00

b

zshy

Posterior Distribution of Bias Factor (8) bull Effect of Removal of middotPlatform from Any Category

bullPrior and Posterior Distributions or B Evaluating Effect or Removal or 3 Damage Cases from the Base Case or 13 Platrorms

4

35

3

25

e L 2 ~ bulle shy

I 15

bull I

05

0

0 02 04 06 I 2 I 4 16 IS 2

I I

I

I

I Ibull

J -tmiddot L--

bullI

~

I

fl

II ~

I I v I

middot bull bull

bull L

~ ll- I-~lt

OS I

b

I

bull

1middot bullbullbullbull

~ __

bull

- - - - Prior-r

- - - - Postenor flall failure cases

Postenor flall success cases

--- Posterior flall 13 cases

---Posterior fI success failure cases only

IPosterior or B Mean =t11 COV =0121

-

_I

Posterir Distribution or Bias Factor (B) bull Effect or Removal or 3-Damage Cases

-

bullPrior and Posterior Distributions or B Evaluating Effect or Additional 13 Similar Observations as

Used in Andrew JIP The POsterior or B from Andrew-JIP Considered as New Prior or B

- 500

450

400

350

bull bull

bull bull bull bull J bull

bull bull f ~ I

I ~ Ibull bull

bullbull bull

l bull I -bullbull bull bull

I

I bull bull

bull I I~~bull bull bull

bull ~

I bull bull

~ ~_ ~~~

- - New Prior r bull

Posterior 114 new failure cases ~ 300- bull bull bull bull bull bull bull Posterior f 13 new damage casesl 150 ~

- bull bull - bull bull Posterior 116 new success cases i 2 00

Postenor 1113 new success failure -150 damage cases

I 00

0 50 -

0 00 0 00 0 20 0 40 0 60 0 80 I 00 120 I 40 160 110 2 00

Posterior ol B Mean= 120 COV= 0071 b

shy

shy

_rshy

Posterior Distribution of Bias Factor (B) bull Effect of Additional 13-Similar Observations

1

1 I

WEIGHTING APPROACH bull

bull ASSESSMENT OF PLATFORMS DATA

Select Platforms Most Mappable to the Analyzed Platforms

Unlikely SurvivalDamage Cases Identification

Based on Cut-Off Criteria Agemiddot Hs etc

bull INFORMATION REQUIRED FOR COMPARISON

bull CATEGORIZATION OF ADDITIONALmiddot PLATFORMS

bull UPDATING OF CALIBRATION RESULTS

--

OVERVIEW OF PLATFORMS SURVIVEDDAMAGED -~

Operator -

- Platforms

Survived CS)

Platforms

Damaged CD)

Platforms Used In

Phase IJIP -

JI~ ~Bill~IRBDll -

Amoco -

Chevron -

Exxon

Mob II

Phllllps

Shell

Unocal -

30

143

72 22

7

39 25shy -

1

12

bull

1 bull

1

1(S)+1 (D)

3 (S) +3(D)

2 (S)

bull shy

bull

bull

bull

Total Partlclnants Platforms 338 - 15 10C6S+4D) -

Other Operators

310

13

3 (D)

Total Platforms 648 28 13C6S+7D) -

Summary of Andrew JIP bull Phase I Platforms

Plairshy w v- N-ber Lea-le 1 shy car Hind_ o-w N-bull Deplh fl fl IJpe la bull n Yeadon Jadtet lbelledtf liefalit -- _ -OD

IL H - -shy _ICllshy IL IL - - rQClto ~ _

K STUIK 137 1963 8 365 608S4 Cellardeck

-middot STl-shy 170 1964 4 -~ K 39 62269 Cellardeck - --ST134W 137 1911 4 43 608S4 Smvlnlshy- _ - -

WD90A 184 1964 8 K 267Sbull34 83 SOS31 SUboftllar deck SmvMI

MC311 343 1978 8 x S7 62 - MC397 468 1991 4 x shy S33bull6SO 65911 -shySmeSmvlvol -- gt -

r-- T23ST52 63 1969 4 ton- shyKshy 52594 S03 -T25SSl39 62 1969 4 Kshy SOS94 S063 - middot ton-shy__ -

bullbull a ton--shyST161A 118 1964 8 K - 347S 51495 Cellarclecti

r~

ST1771 142 1965 8 K 39 60167 CeDordeck failme

STISIH 137 1964 8 K 395 608S4 CeDardeck Failme

K ST130A 140 1958 a SS 6095 I lblv Pai1uno - T21ST72 61 1969 4 Kshy S4S94 4974 I lhlv Pai1uno -

- -

JE

INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON

bull WATER DEPTH

bull bull YEAR OF INSTALLATIONDESIGN

bull PLATFORM ORIENTATION

bull PLATFORM LATITUDE AND LONGITUDE

bull KEY SKETCHES OF PLATFORM STRUCTURAL FRAMING

Number of Legs Plies Leg Batter Number of Conductors bull Deck Elevations (Including sub-cellar deck If any) Jacket Framing Elevations Brace type In Vertical Frames Sizes of Primary Members (leg pile vertical frames braces primary horizontals) Joint Details (can gap etc) LegPiie Annulus Grouting Status

bull middotPOST ANDREW INSPECTION REPORT Significant Wave Height Hs (ft)

bull INFORMATION ON DAMAGES PRIOR TO ANDREW

bull RESULTS FROM IN-HOUSE STUDIES IF AVAILABLE

bull ~RIGINAL DESIGN CRITERIA INFORMATION IF AVAILABLE

bull SOIL SHEAR STRENGTH PROFILE IF AVAILABLE

Platforms Identified for Evaluating Therr Likelihood of Falling Under Unexpected Survival Category

OPlRATOR

middotADA

BLOClt

llmucnJllE

NAME

YEAll

INSTAUJD

WA1ER

DBPTll

DISTANCE

JlllOM SHOllE

(MILU)

Bl

(M)

AMOCO 111113

STl11

WDI075

A

bull D

St

17

120

1n

44

20

10 I

CllllVllQfl STlllO c 61 110 21 11 STIUO D Q 1110 21 11

STl134 p Q 137 29 11

ST0151 G 61 137 11

ST0151 I 3 121 11

ST0151 J Q 140 11

STl151 It 137 11

ST0151 PJl001 Q UT 11

ST0151 PROD-2 Q UT 11

ST017 A 3 140 35 11

WDIUT c 65 214 34 10

llXXON STI055 II 51 6T 14 I

STl15 A t3 34

WDI073

WD0073

Amiddotshy Q

1111 1111

22

22

I

I

WDI073 c 1n 22 I

- WDOCl93-shy II

65 1110 2T A 3 194 31 MOBU 111121

SS IOl7

A B

501

50

50

31

T

I

T 6

MtlllPllY SS114 B T 50 T T

PlllllIPS SS0149 A 5 55 33 T

SAMllDAN STln A 107 T T

SBllLL WDl103 A 65 223 2T WDl103 B 65 221 2T WD0104 c 65 221 2T WD0104 D 6T 20 2T

UNOCAL SS020I

SS 020I SSl20I

SS 0209

SS0209

II p

PURll

A

B

3

n 61

103

97

97

95

100

34

34

34

34 34

bullbull bullbullbull

- SS0209 D u 95 34 I

- SS 1209 SSl253

SS0253

G A

c

6T

2

100

165

1T5

34

54

54

bullbull I

I I

I I

SOUTH MARCH

~LAND

N

SOUTH PaTOI ENE ISWltD AREA

MMS Defined Limits ror Offshore Structures Requiring Inspections Following Andrew

t---Jbulln3 T25bull I bullTtl

bull1$1Htl1Abull bull1SIK1779bull

~

middot-DaTA AREA~T

CRANO rsLE

AREA bull1300

bullll4W

=shy

bullPath of Hurricane Andrew with Platforms Used In Calibration

Likelihood Function vs b All Success Case STISI K STIJOQ ST134W WD90A MCJll MCJ97

I 0000

0 9000

0 8000

07000

-Bo6000

tiiii 0 5000 -

04000

0 3000

0 2000

01000

0 0000

0 05 25

ri1rt middot-middot-middot-middot-middot-

I 5 2

b

---11 middot151K(blsu~ccsI

- bull - bull - fk 130Q(bfU~lCSS)

lie bull134W(blsuccc]

lk-WD90A[blsultCS]

- - - lk middotMC397(bfUtlCS]

--o-- lk-MCJll(bbuctcsJ

--- ll (bl 6 out of 6 Mlttcs case]

_

( Likelihood Functions middotSuccess Cases - p - 1 c I

CATEGORIZATION OF ADDITIONAL PLATFORMS

I I I

bull CATEGORIES

Unexpected Survivals

Expected Survivals

bull APPROACH

Identify Corresponding Platform(s) Evaluated In the JIP

Determine Max Wave Height at Platform Location

Evaluate Whether Wave Hit the Deck

middot - Evaluate Likelihood of R and S to Differ from the Base Cases

Review Joint Information

Assess Special Features

bull IDENTIFY NUMBER OF PLATFORMS IN EACH CATEGORY

I

UPDATING OF CALIBRATION RESULTS

bull INCLUDE EFFECT OF ADDITIONAL PLATFORMS ON bullBbull

Consider Increased 1Uncertalntles In R and S

Develop bullNeW Ukellhood Functions for Different Categories

middotLet bullxbull Number of Ukely Unexpected Survival Cases and y Number of Ukely Expected Survival Cases

Determine Combined Likelihood Function lk (b Iadditional platforms) middot

bull REVISED LIKELIHOOD FUNCTION

lk(b I all cases) = lk(b I 9 platforms) middot x lk(b Ix platforms) x lk(b I y platforms)

bull ESTABLISH REVISED BIAS FACTOR B

f8 (b) u f8 (b) lk (b I all cases)

II

t t ~

shy

APPLICATIONS OF BIAS FACTORS

ii APPLICABLE TO PARTICULAR LOAD AND RESISTANCE RECIPE

bull USE BY THE API COMMITTEES AND MMS

Criteria Revisions

Recommended Design Guidelines Revisions bull

bullbull REASSESSMENT STUDIES AND DECISION MAKING

middot

bull RISK ASSESSMENT STUDIES BY THE OPERATORS

Individual Platforms

Fleet of Platforms

middot IMR Program Implementation

bullbull NOT TO BE USED IN DESIGN

I I fl J

EXAMPLE USE OF BIAS FACTORS

bull PHASE I EXAMPLE APPLICATION

bull EVALUATION OF ANNUAL PROBABILITY OF FAILURE

Establish Annual Maximum Seastate Data

Determine Coefficients to Define Base Shear

Determine Ultimate Capacity of Platform

Determine Probablllty of Failure Versus bullbbull

-bull DETERMINE UPDATED PROBABILITY OF FAILURE

Pf - P [S gt R]

Pf I P [~ gt Rb I b] f 8 (b) db

Pr - J P [S gt Rb I b] f bull 8 (b) db

bull TOTAL PROBABILITIES OF FAILURE

Total Pf - 1 bull Il [1 middotPf] - l Pf bull

Total Pf - 1 bull Il [1middot Pf] - l Pfmiddot

Total Pr = 1 - n [1middot PfbullJ = l Pfshy

I I I

EXAMPLE APPLICATION bull PHASE I

PLATFORM ST 151 K

bull GULF OF MEXICO

bull WATER DEPTH = 137 FT

bull ANNUAL MAXIMUM HS MEDIAN 175 ft COV 031

bull ASSUMED STORM DURATION 3 hoursmiddot

bull ASSUMED NUMBER OF WAVES 800

bull MEDIAN HS FOR STORM HOURS 1 2 AND 3

bull MEDIAN CURRENT VALUE bull

bull DETERMINE ULTIMATE CAPACITY FOR THREE DIRECTIONS

bull DETERMINE ANNUAL FAILURE PROBABILITIES

Broadsidemiddot Diagonal End-on

0037 0025 0007

0071 0044 0018

0022 0016 0004

DETERMINE TOTAL ANNUAL PROBABILITY OF FAILURE

Pf - 007

Pf - 013

Pr - 004

bull

EXAMPLE USE OF BIAS FACTORS - PHASE II middot

bull DETERMINE PREDICTIVE PROBABILITY OF FAILURE

Case - 1

Pf _ S S Pf (b b2) fa1a2 (b b2) db db2

g bull -1 (b1 R I S) - 1

92 - (b2 R2 I S) - 1

Case -2

Pf s ~ [bJ f11 (b) db

Pf [bJ p [g lt 0)

g - bull 1 (b1 R1 I S) - 1

1

bull1) l

I

BENEFITS OF ANDREW JIP bull PHASE II

lt

bull IMPROVED LOAD AND RESISTANCE RECIPE

bull IMPROVED CALIBRATION PROCEDURE

bull MULTIPLE BIAS FACTORS

bull PROCEDURE TO INCLUDE EFFECT OF OTHER PLATFORMS

bull PROCEDURE TO USE MULTIPLE BIAS FACTORS

  • Structure Bookmarks
    • ANDREW PHASE II JOINT1 INDUSTRY PROJECT
    • PROGRESS MEETING NUMBER 1
    • ANDREW PHASE II JOINT INDUSTRY PROJECT PROGRESS MEETING NUMBER 1 NOVEMBER 16 AGENDA
    • ANDREW PHASE II UST OF PARTICIPANTS
    • ANDREW PHASE I OBJECTIVES AND RESULTS
    • ANDREW PHASE II PROJECT OBJECTIVES
    • SUBSEQUENT PROJECTS
    • ANDREW PHASE II PROJECT TEAM
    • ANDREW PHASE II UPDATE MODELING middotAND ANALYSIS RECIPE
    • -ANDREW PHASE II CONDUCTOR RESISTANCE
    • ANDREW PHASE Il WA VE FORCE PROFILE
    • ANDREW PHASE II WA VE FORCE PROFILEmiddot
    • ANDREW PHASE II BRACE MODELING
    • ANDREW PHASE II FOUNDATIONS
    • ANDREW PHASE II middot middot JOINT MODELING
    • ANDREW PHASE Il JOINT MODELING OPTIONS DISCUSSION
    • ANDREW PHASE II JOINT MODEJJNG OPTIONS ILLUSTRATIONS
    • ANDREW PHASE II JOINT MODELING PHASE II METHOD
    • ANDREW PHASE II ANDREW STORM CONDmONS UPDATED
    • Platform Selection bullbullbull1111111111111
    • BasisCriteria for Selection bullbullbull1111111111111
    • CALIBRATION
    • WHERE TO INTRODUCE BIAS FACTOR (B
    • PHASE I -KEY CONCLUSIONS
    • bull CALIBRATION ISSUES PHASE II
    • PHASE II WORK SCOPE (CALIBRATION)
    • SOFTWARE
    • SENSITIVITY STUDIES POSTERIOR DISTRIBUTION OF B
    • WEIGHTING APPROACH bull
    • INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON
    • CATEGORIZATION OF ADDITIONAL PLATFORMS
    • UPDATING OF CALIBRATION RESULTS
    • APPLICATIONS OF BIAS FACTORS
    • EXAMPLE USE OF BIAS FACTORS
    • EXAMPLE APPLICATION bull PHASE I PLATFORM ST 151 K
    • EXAMPLE USE OF BIAS FACTORS -PHASE II middot
    • BENEFITS OF ANDREW JIP bull PHASE II
Page 37:  · . ' ANDREW PHASE II PROJECT OBJECTIVES . ' • ·UPDATE MODELING AND ANALYSIS RECIPE· JOINTS ' ' ·BRACES WAVE PROFILE FOUNDATIONS . ' ' CONDUCTORS· • ANALVZE PLATFORMS USING

ANDREW PHASE II

JOINT MODEJJNG OPTIONS ILLUSTRATIONS

P

e

ANDREW PHASE II

JOINT MODELING PHASE II METHOD

e

I I I

ANDREW PHASE II

ANDREW STORM CONDmONS UPDATED

bull -PHASE I USED OCEANWEATHER HINDCAST-(Nov 1992)

bull middotmiddotPHASE II WILL USE THE REVISED HINDCAST (Nov 1994)

bull lt c lt I lt

2-2

l---4-~~~+-~~-+-~~~1--~~-+-~~--l--~T--+-~~--i24

f-tJVtMfiz~ 1Cfl (r STvb(

Plollol bullbull 5middotNOYmiddot9214bull1J Irbullbull Illbull [ANDREVl~NDREVCUSTbullMAXll 5-NOY-1992 10bull54 - 9208

HEIGHT lml

I~ MAXI

I L 1~middotamp SZ~- ~A I 13c

I J) ( r I X~0iWJ~0gtJ gt L gt gt I 1c I 120

-- N

~ deg

I I I I I I e C I ~=-41 122 -94 -92 -90 -88 -86 -84 -82 -80

-

Figure 7 contours of maximum hindcast significant wave height

_

~

~tl ~

NOltNttIJ 17lt ltNS S iJJ)y

Maximum Significant Wave Height - Hurricane Andrew 1992

30 ~

26 ~ I ~-~~~- == -shy -~ 35-_

24deg1--shy --------~- middot-middot-------middotmiddotmiddot----middot------middot-middot~middot- - - ~-----1---middot

22middot ~~~~~~~-~ -94 -92 -90 -88 -84deg-86 -82 -somiddot

c

-

Platform Selection bullbullbull1111111111111

0 BasisCriteria for Selection middot

0 Review of Phase I Platforms middot 0 Alternative Platforms

0 PMB Recommendations

0 Participants Comments

gt

L

l I

BasisCriteria for Selection bullbullbull1111111111111

0 Benefit for Calibration - Classification (UnexpectedExpected)

- Failure Mode (StructuresFoundations)

- Damaged StructuresKnown response

- Better Representation of Full ~opulation

0 Availability of Data - Existing Models

- Inspection Data

- Damage Data

PREDICTED BEHAVIOR

-

OBSERVED

BEHAVIOR

-SURVIVED DAMAGED FAILED

SURVIVED

ST134W WD90A

-MC311 -

MC397 -

--

-

-

ST151K ST130Q

-

-

-

DAMAGED - -

-

-

--

-

-

ST52 SS139 ST177B -

ST161A

-_

-

FAILED

-

-

-

ST151H ST130A ST72

Operator - Platfonns - Platfonns Survived Damagedamp

- Falled -

bull

-

-

Jle eBDIGIPllDbl -Amocoshy 30 1 Chevron -

143 12 Exxon 72 Mobll 22 1 Phllllps 7 bull

-Shell 39 Unocal 25 1

Total Partlclnts Platfonns 338 15

Other Operators bull- 310 13

-

Total Platfonns - -

648 28

fbml

Platforms Considered 6 7

tie of Total Platfonns 1 25

OVERVIEW OF PLATFORMS SURVIVEDDAMAGEDFAILED

-

shy

v

Platforms Evaluated in Phase I

-

-Platfonn Water Year Number Analysis Avallabillty of Availability Platform

- Name Depth rl of Model DetaDed rl ClassUlcatlon

Installation legs Inspection Soll Report Based on Information Phase I Calibration - -ft

- -- Suaill flalfnrm ~BSi -STISIK 137 1963 8 CAP Yes Yes Un eel Sarvlval

-STIJOO 170 1964 4 CAPmiddotshy Yes ST134data u ST134W 137 1981 4 CAP Yes Yesmiddot E

-

edSarvlval -

middotsurvival

WD90A 184 - 1964 8 CAP Yes E -

- Sarvlvill

MC3ll 343 1978 8 KARMA

MC397 468 1991 4 KARMA -

bull ~reglmmiddot- -

Yes Sure Survival

bull Yes - Sare Sanlval-

-T23STS2 63 1969 4 CAP

T2SSS139 62 1969 4 CAP -

Yes

Yes

Yes Ex

Yes Une

bull SurvivalLikely to DamalI

middot bull SanlvalLlkely to DamalI

ST161A 118 1964 8 USFOS

Eail11a pound1al(bullllm Clss ST177B 142 --1965 8 CAP

STISIH 137 1964 - 8 CAP C

STIJOA 140 19S8 8 CAP - -

Yes

Yes

ollamed

Yes

- Yes E

-

at about 1 mUe

Yes

ST134data

bull SafvlvalLlkelv to Dam- bull- -

E middotrahre

Emeried Failure -

Likely FaUare

T21ST72 61 1969 4 CAP Collamed No Likely FaUure

Platforms Identified for Evaluating Their Likelihood of Falling Under UnexpeCted Survival Category

CllEllATOa

AllB

llLOCll

Sl1lvcnJllE

NAME

YEAR

INSTAUJCD

WATER

IJBPJB

middotmiddotDISTANCE

ftOMSllOllE

(MIUIS)

Ill

(M)

AMOCO BllW

STllI

WDI075

A

bull D

4

n uo 17J

44

32

bull 20

10

bull CllEVllOH STtllO c I 1bull 21 11

STl130 D 0 161 21 11 STllJ4 I 0 137 D - 11

STIUI G I 137 32 11

STIUI I 0 121 32 11

STIUI

J 0 140 32 11

STIUI It 4 137 32 11

STllSI PROl)I 0 137 32 11

STllSI PROD-2 0 137 32 11

STll7 A 0

140 35 11

WDl117 c 65 214 34 10

DXON STshy E 51 14 bull STl165 A t3 bull 34 WD0073

WD0073

WD0073

A _ c

4

62

4

IA IA

17J

22

22

22

bullbullbull WD003-shy E 65 161 Z7

A 0 IN 31 MOBD Bl12I

ssn A bull

50

50

31

7

bull 7

MIJllPllY SS114 B 7 50 7 7

PlllUIPS SSl14t A 55 33 7

SAMEDAN ST17J A 4 117 7 bull 7

SJIELL WD0103 A 65 223 27 WDllOJ bull 65 221 27 WDll04 c 65 221 27 WDOI04 D 67 2 27

UNOCAL

SSl20I

SSl20I

bull SSl20I

SS0209

SS0209

SS0209

SS0209

SS 0253

SS 0253

E

I

PtJllE

A

bull D

G A

c

0

4

4

7J

61

0

0

69

103

97

97

9S

100

95

100

165

175

34

34

34

34

34

34

34

54

54

bullbullbullbullbullbullbullbullbull

Platforms Recommended for Phase II

llalr y_ w- Aot Av- _ I 1 Doplla Model nm bull bull _ bull I I H11 s18-1 Irht lbwl~ bullbullbull(-illtdlel OM lllMtD

II ---

-STISIIC 137 1963 I CAP Yeo Yeo I Yeo

STtshy -170 1964 4 CAP Yeo ST134dlU I Yeo

STl34W 137 Yeo Yeo -1981 4 CAP D Yeo

WD90A 114 1964 I CAP Yeo - D

MC311 30 1971 I KARMA Yeo -sre-

MC397 middot 461 1991 4 ICARMA Yeo Sireshy - - r- -

__ T13ST5l 63 1969 4 CAP Yeo Yeobull middot- ~n- D Yeo _ Tl5SSl39 62 1969 4 CAP Yeo Yeo I Yeo

_ STIT711 142 1965 I CAP Yeo I Yeo middot-1shySTl61A Ill 1964 I USFOS Yeo Yeo - D bulloshy

- r-

STISIH 137 1964 I CAP y D y -_ STl31A 140 1951 I CAP Yeo STl34dlU I Yeo

T21Sll 61 1969 4 CAP No rar-middot I Yeo

~-rmiddot-middot SS209A 95 1967 16 CAP Yeo D -SSU4R

11 UI A 41 -in D SS 2741AIM CAP bull

_ _ -shy - -shy- -shy

CALIBRATION

bull PHASEIAPPROACH

bull CALIBRATION ISSUES

bull PHASE II WORK SCOPE

bull SOFlWARE AVAILABILITY

bull EFFECT OF ADDITIONAL PLATFORMS

bull WEIGHTING APPROACH

bull EXAMPLE APPLICATION middot

(l=Ofce Mode] Wind

~

fcurrent l -middot

~SW ~leldl4

ultimate resistance force I first

design load Illgt

i _

deflection

-

bull

4

Historical Hurrlcane Dariiage and Survival of Platforms

Components Required for Calibration (Petrauskas 1992)

PRIOR Distribution

PLATFORM EXPERIENCE

BAYESIAN Survivals and Failures

UPDATING

Waves Winds and Currents bull

POSTERIOR Distribution

- A_middotJA ~ Survival Data Failure Data

Result Result

Load Resistance

middotfl

-

Bayesian Updaing middotGeneral Description (Petrauskas 1992)

L Imiddot

WHERE TO INTRODUCE BIAS FACTOR (B

On Resistance (Rmue - B (RPREDICreD I

middotOn Loads (Smue middot - B (SPRE01cTEo

On Safety Margin (RSmue - B (RSPREDICTED

CALIBRATION METHODOLOGY

gt

r

FIELD OBSERVATIONS

Observed Behavior

survivals Damages

Failures

r

bull

CAPACITY ANALYSIS

Predicted Load and Resistance Predicted Failure Modes

Survival Damage Failure cases

RELIABILITY ANALYSIS

- Probability of Failure ~

Ukellhood Failure

BAYESIAN UPDATING

Prior Distnbution of B Combined Likelihood Function

Posterior Dlstnbution of B

BIAS FACTOR

Mean Value cov bull

Represents Modeling Uncertainties

Prior Distribution of B Likelihood Functionmiddot Observing Success case STISIK and Posterior Distribution of B

2 00

uo

160

140

120

e 100 L $ uoIshy

0 60 ~ ~ 040

020

0 00

0 00 0 50

_ I 00

b

I 50 2 00

-shy lk [bl STISIK)

--shy Pnor-f(b)

--shy Pos1enor r(bl ISIK)

Prior or B Mean bull 10 COV bull 030

Posterlor of Bbull Meanbull 136COV 0 16

2 50

Likelihood Function and Posterior Distribution Example Platform STISIK

- -

~

- Prior and Posterior Distributions or B Failure cases (ST177B STlSlH ST130A T21) Damage cases (T23 T25 ST161A) Success cases (STlSlK ST130Q ST134W WD90A MC311 MC397)

400

350

300

2 so 0-

I

bull bull

bull bull gt

- bullbull ~ bull bull I bull

_ bull bull bullbull I bull bull bullII V -

bull bull

I bull ~ bull

bullbull bullI bull bullI bull bull bull

middot -~

bull__ ~ middotI middotmiddot

L 2 00

e ~ I SO

-I 00

- 0 so

000 gt

0 00 020 0 40 060 0 80 I 00 120 140 160 110 200

b -

-

shy

Prior-r

- - - - Posterior fLlll failure cases

bull bull bull bull bull bull bull Posterior flall damage cases

- bull bull - bull middot Postenor flall success cases

Posterior flall success fulure damage cases

-

Posterior of B Mean= 119 COV 0101

Posterior Distribution or Blas Factor (8) bull Different Categories and All Cases Combined

shy

l

PHASE I - KEY CONCLUSIONS

bull GLOBAL CORRECTION FACTOR

bull 15 TO 20 CONSERVATISM MEAN)

bull UNEXPECTED SURVIVALS HAVE GREATEST IMPACT

bull DAMAGED CASES MORE INFORMATIVE

bull CALIBRATION ISSUES PHASE II

bull

EFFECT OF IMPROVED CAPACITY ANALYSIS RECIPE

bull APPLICABILITY TO MULTIPLE FAILURE MODES

bull NO FOUNDATION DAMAGE CASE INCLUDED

bull IMPROVED REPRESENTATION OF FULL ANDREW DISTRIBUTION

-bull NUMBER OF PLATFORMS CONSIDERED

I 1 I

I t ) bull ~

PHASE II WORK SCOPE (CALIBRATION)

bull

bull ESTABLISH MULTIPLE BIAS FACTORS

Use Nine Base Case Platforms

middot - Capacity Analysis for Revised recipe

Foundation Blas Factor from Foundation JIP

bull DEVELOP WEIGHTING APPROACH

bull UPDATE MULTIPLE BIAS FACTORS

bullEffect of Platforms Not In the Nine Base Cases

bull EXAMPLE APPLICATION OF BIAS FACTORS I

bullbull

PHASE I PHASE II

CAPACITYbull ANALYSISRu bull

---------------------1----middot

Jc1 - iJbullraquoJ dbull

FttlS frt[f~l ratdhJJ

-

LIKELIHOOD FUNCTION

DEVELOPMENT

lk Jb Jfall1N) bull Jb)

111 Ill Isurvtvoq bull 1 - bl

-middot -------------------shy ~---------------------middot-middot

r111ir8 Cb1 111 Cbl BAYESIAN

UPDATING

~cove

R1R2 bull middot---------------------~----

Jblbull P(g cOJ

1 bull JbRI (MBSI I bull 1

~middot

llcJllJfalluro) bull P(gcOJ Ill 111 b21fallun) bull P ((g1 CO) U 1112 c OD

middot--------------------shy ----------------------shy bull

r111111r8 Cbl 111 Ill i-1nro1

deg ~1b2)laquottfrr (b1b21-lnfo) re Ill bull 1-11middot b2) bull lodegamp

middot

i1 covamp i2 cov~i

-

shy

--

Likelihood Function vs b All Success Case STISI K STl30Q STl34W WD90A MC311 MC397

I 0000

0 9000

OIOOO

0 7000

~ 0amp000

ta osoOO

~ 0 4000

0 3000

0 2000

0 1000

bull 0 0000

bull

I

middot

middot bull I

Ibull

-sibull ~ -middot-middot-middot-middot -middot-

----7----+-----~ I

Ibull bull bull I

rbull

--- Jk ISIK[blsuccess)

- bull - bull - lk 130Q[blsuccess) bull

lk 134W[blsuccess)

Jk-WD90A(blsuccc~I

- - - lk MC397[blsucccss)

--ltgt-- lk middotMC31 l[blsucccss)

--- Jk [bl 6 out of6succcsHascs)

05 I 5 2 250

b

Likelihood Functionsmiddot Success Cases - Phbullsc L

shy

bullbull bull bull bull bull bull -------------------~

bull bull bullbull bullbull bullbull bullbull bullbull bullbull bull

bull bullbull

bullbullbullbull

bullbull

bullbullbullbullbullbullbull

bullbullbullbullbullbull

bullbullbullbullbullbullbull

z 0

t z ) LL

c 0 0c i ~ i

J

~ sect ~--------------~ e l5

I

I

SOFTWARE

bull PF PROGRAM

bull CALREL

bull PF +CALREL

bull RELACS

bull

gt i

l

SENSITIVITY STUDIES

POSTERIOR DISTRIBUTION OF B

bull EFFECT OF CHANGE IN DISTRIBUTION OF PRIOR OF B

bull EFFECT OF REMOVAL OF ONE PLATFORM FROM EACH CATEGORY

bull EFFECT OF REMOVAL OF THREE DAMAGE CASES

bull EFFECT OF ADDITIONAL 13 SIMILAR PLATFORMS

-I

Prior and Posterior Distributions of B Variation In COV of PriorofB fromOl to04 -- -

4500

-4000

-

- -

3500 --middot - bull bull - bull bull Prior-r ICOV =02

3000

- - - - Prior-r ICOV =03 -

bull -~ e 2soo -

bull bull bull bull bull bull bull Pnor-r ICOV =o 4t frJ ~ 2000 Posterior rlCOV =0 2

bull bull Postenor rlCOV =0 3 bull bull 1500 bull middot bull - shy -middot- Posterior rlCOV =04

middot~ I 000 - bull

-middotmiddot ~ ~middotmiddot M -~

- middot 0 500

wbullbullbull _ I middot~ middotmiddotmiddotmiddotmiddot middotmiddotmiddotmiddotmiddot - -~ ~ __ middot_ _ bull -shy0 000

0 000 0200 0 400 0 600 0 800 IOOO 1200 1400 1600 1800 2000 -

- - b

-

=shy

Posterior Distribution of Bias Factor (B) bull Effect of Variation In COV of Prior of B

bull

bullPrior and Posterior Distributions of B Evaluating Effect of Removal of One Platform from Each Category from the Base Case of 13 Platforms

400

~ ~ 350

j IA i

3 00 bull I bull I ---Prior-r I bull l

250 1 I 1

I bull bull ---Posterior r lall 13 casesshy- ~E 2 00 bull bull bull bull bull bull bull Posterior r without STISIK

- ~ 1 bull - bull I - bull bull - bull bull Posterior r Iwithout T2I I 50 I

J-tt~i - bull - bull - Posterior r Iwithout T2S

tOO [-

I I ~ ~ I 0 50

0 00

___ r J i

l~T

~--

0 00 0 20 0 40 0 60 080 I 00 120 140 160 110 2 00

b

zshy

Posterior Distribution of Bias Factor (8) bull Effect of Removal of middotPlatform from Any Category

bullPrior and Posterior Distributions or B Evaluating Effect or Removal or 3 Damage Cases from the Base Case or 13 Platrorms

4

35

3

25

e L 2 ~ bulle shy

I 15

bull I

05

0

0 02 04 06 I 2 I 4 16 IS 2

I I

I

I

I Ibull

J -tmiddot L--

bullI

~

I

fl

II ~

I I v I

middot bull bull

bull L

~ ll- I-~lt

OS I

b

I

bull

1middot bullbullbullbull

~ __

bull

- - - - Prior-r

- - - - Postenor flall failure cases

Postenor flall success cases

--- Posterior flall 13 cases

---Posterior fI success failure cases only

IPosterior or B Mean =t11 COV =0121

-

_I

Posterir Distribution or Bias Factor (B) bull Effect or Removal or 3-Damage Cases

-

bullPrior and Posterior Distributions or B Evaluating Effect or Additional 13 Similar Observations as

Used in Andrew JIP The POsterior or B from Andrew-JIP Considered as New Prior or B

- 500

450

400

350

bull bull

bull bull bull bull J bull

bull bull f ~ I

I ~ Ibull bull

bullbull bull

l bull I -bullbull bull bull

I

I bull bull

bull I I~~bull bull bull

bull ~

I bull bull

~ ~_ ~~~

- - New Prior r bull

Posterior 114 new failure cases ~ 300- bull bull bull bull bull bull bull Posterior f 13 new damage casesl 150 ~

- bull bull - bull bull Posterior 116 new success cases i 2 00

Postenor 1113 new success failure -150 damage cases

I 00

0 50 -

0 00 0 00 0 20 0 40 0 60 0 80 I 00 120 I 40 160 110 2 00

Posterior ol B Mean= 120 COV= 0071 b

shy

shy

_rshy

Posterior Distribution of Bias Factor (B) bull Effect of Additional 13-Similar Observations

1

1 I

WEIGHTING APPROACH bull

bull ASSESSMENT OF PLATFORMS DATA

Select Platforms Most Mappable to the Analyzed Platforms

Unlikely SurvivalDamage Cases Identification

Based on Cut-Off Criteria Agemiddot Hs etc

bull INFORMATION REQUIRED FOR COMPARISON

bull CATEGORIZATION OF ADDITIONALmiddot PLATFORMS

bull UPDATING OF CALIBRATION RESULTS

--

OVERVIEW OF PLATFORMS SURVIVEDDAMAGED -~

Operator -

- Platforms

Survived CS)

Platforms

Damaged CD)

Platforms Used In

Phase IJIP -

JI~ ~Bill~IRBDll -

Amoco -

Chevron -

Exxon

Mob II

Phllllps

Shell

Unocal -

30

143

72 22

7

39 25shy -

1

12

bull

1 bull

1

1(S)+1 (D)

3 (S) +3(D)

2 (S)

bull shy

bull

bull

bull

Total Partlclnants Platforms 338 - 15 10C6S+4D) -

Other Operators

310

13

3 (D)

Total Platforms 648 28 13C6S+7D) -

Summary of Andrew JIP bull Phase I Platforms

Plairshy w v- N-ber Lea-le 1 shy car Hind_ o-w N-bull Deplh fl fl IJpe la bull n Yeadon Jadtet lbelledtf liefalit -- _ -OD

IL H - -shy _ICllshy IL IL - - rQClto ~ _

K STUIK 137 1963 8 365 608S4 Cellardeck

-middot STl-shy 170 1964 4 -~ K 39 62269 Cellardeck - --ST134W 137 1911 4 43 608S4 Smvlnlshy- _ - -

WD90A 184 1964 8 K 267Sbull34 83 SOS31 SUboftllar deck SmvMI

MC311 343 1978 8 x S7 62 - MC397 468 1991 4 x shy S33bull6SO 65911 -shySmeSmvlvol -- gt -

r-- T23ST52 63 1969 4 ton- shyKshy 52594 S03 -T25SSl39 62 1969 4 Kshy SOS94 S063 - middot ton-shy__ -

bullbull a ton--shyST161A 118 1964 8 K - 347S 51495 Cellarclecti

r~

ST1771 142 1965 8 K 39 60167 CeDordeck failme

STISIH 137 1964 8 K 395 608S4 CeDardeck Failme

K ST130A 140 1958 a SS 6095 I lblv Pai1uno - T21ST72 61 1969 4 Kshy S4S94 4974 I lhlv Pai1uno -

- -

JE

INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON

bull WATER DEPTH

bull bull YEAR OF INSTALLATIONDESIGN

bull PLATFORM ORIENTATION

bull PLATFORM LATITUDE AND LONGITUDE

bull KEY SKETCHES OF PLATFORM STRUCTURAL FRAMING

Number of Legs Plies Leg Batter Number of Conductors bull Deck Elevations (Including sub-cellar deck If any) Jacket Framing Elevations Brace type In Vertical Frames Sizes of Primary Members (leg pile vertical frames braces primary horizontals) Joint Details (can gap etc) LegPiie Annulus Grouting Status

bull middotPOST ANDREW INSPECTION REPORT Significant Wave Height Hs (ft)

bull INFORMATION ON DAMAGES PRIOR TO ANDREW

bull RESULTS FROM IN-HOUSE STUDIES IF AVAILABLE

bull ~RIGINAL DESIGN CRITERIA INFORMATION IF AVAILABLE

bull SOIL SHEAR STRENGTH PROFILE IF AVAILABLE

Platforms Identified for Evaluating Therr Likelihood of Falling Under Unexpected Survival Category

OPlRATOR

middotADA

BLOClt

llmucnJllE

NAME

YEAll

INSTAUJD

WA1ER

DBPTll

DISTANCE

JlllOM SHOllE

(MILU)

Bl

(M)

AMOCO 111113

STl11

WDI075

A

bull D

St

17

120

1n

44

20

10 I

CllllVllQfl STlllO c 61 110 21 11 STIUO D Q 1110 21 11

STl134 p Q 137 29 11

ST0151 G 61 137 11

ST0151 I 3 121 11

ST0151 J Q 140 11

STl151 It 137 11

ST0151 PJl001 Q UT 11

ST0151 PROD-2 Q UT 11

ST017 A 3 140 35 11

WDIUT c 65 214 34 10

llXXON STI055 II 51 6T 14 I

STl15 A t3 34

WDI073

WD0073

Amiddotshy Q

1111 1111

22

22

I

I

WDI073 c 1n 22 I

- WDOCl93-shy II

65 1110 2T A 3 194 31 MOBU 111121

SS IOl7

A B

501

50

50

31

T

I

T 6

MtlllPllY SS114 B T 50 T T

PlllllIPS SS0149 A 5 55 33 T

SAMllDAN STln A 107 T T

SBllLL WDl103 A 65 223 2T WDl103 B 65 221 2T WD0104 c 65 221 2T WD0104 D 6T 20 2T

UNOCAL SS020I

SS 020I SSl20I

SS 0209

SS0209

II p

PURll

A

B

3

n 61

103

97

97

95

100

34

34

34

34 34

bullbull bullbullbull

- SS0209 D u 95 34 I

- SS 1209 SSl253

SS0253

G A

c

6T

2

100

165

1T5

34

54

54

bullbull I

I I

I I

SOUTH MARCH

~LAND

N

SOUTH PaTOI ENE ISWltD AREA

MMS Defined Limits ror Offshore Structures Requiring Inspections Following Andrew

t---Jbulln3 T25bull I bullTtl

bull1$1Htl1Abull bull1SIK1779bull

~

middot-DaTA AREA~T

CRANO rsLE

AREA bull1300

bullll4W

=shy

bullPath of Hurricane Andrew with Platforms Used In Calibration

Likelihood Function vs b All Success Case STISI K STIJOQ ST134W WD90A MCJll MCJ97

I 0000

0 9000

0 8000

07000

-Bo6000

tiiii 0 5000 -

04000

0 3000

0 2000

01000

0 0000

0 05 25

ri1rt middot-middot-middot-middot-middot-

I 5 2

b

---11 middot151K(blsu~ccsI

- bull - bull - fk 130Q(bfU~lCSS)

lie bull134W(blsuccc]

lk-WD90A[blsultCS]

- - - lk middotMC397(bfUtlCS]

--o-- lk-MCJll(bbuctcsJ

--- ll (bl 6 out of 6 Mlttcs case]

_

( Likelihood Functions middotSuccess Cases - p - 1 c I

CATEGORIZATION OF ADDITIONAL PLATFORMS

I I I

bull CATEGORIES

Unexpected Survivals

Expected Survivals

bull APPROACH

Identify Corresponding Platform(s) Evaluated In the JIP

Determine Max Wave Height at Platform Location

Evaluate Whether Wave Hit the Deck

middot - Evaluate Likelihood of R and S to Differ from the Base Cases

Review Joint Information

Assess Special Features

bull IDENTIFY NUMBER OF PLATFORMS IN EACH CATEGORY

I

UPDATING OF CALIBRATION RESULTS

bull INCLUDE EFFECT OF ADDITIONAL PLATFORMS ON bullBbull

Consider Increased 1Uncertalntles In R and S

Develop bullNeW Ukellhood Functions for Different Categories

middotLet bullxbull Number of Ukely Unexpected Survival Cases and y Number of Ukely Expected Survival Cases

Determine Combined Likelihood Function lk (b Iadditional platforms) middot

bull REVISED LIKELIHOOD FUNCTION

lk(b I all cases) = lk(b I 9 platforms) middot x lk(b Ix platforms) x lk(b I y platforms)

bull ESTABLISH REVISED BIAS FACTOR B

f8 (b) u f8 (b) lk (b I all cases)

II

t t ~

shy

APPLICATIONS OF BIAS FACTORS

ii APPLICABLE TO PARTICULAR LOAD AND RESISTANCE RECIPE

bull USE BY THE API COMMITTEES AND MMS

Criteria Revisions

Recommended Design Guidelines Revisions bull

bullbull REASSESSMENT STUDIES AND DECISION MAKING

middot

bull RISK ASSESSMENT STUDIES BY THE OPERATORS

Individual Platforms

Fleet of Platforms

middot IMR Program Implementation

bullbull NOT TO BE USED IN DESIGN

I I fl J

EXAMPLE USE OF BIAS FACTORS

bull PHASE I EXAMPLE APPLICATION

bull EVALUATION OF ANNUAL PROBABILITY OF FAILURE

Establish Annual Maximum Seastate Data

Determine Coefficients to Define Base Shear

Determine Ultimate Capacity of Platform

Determine Probablllty of Failure Versus bullbbull

-bull DETERMINE UPDATED PROBABILITY OF FAILURE

Pf - P [S gt R]

Pf I P [~ gt Rb I b] f 8 (b) db

Pr - J P [S gt Rb I b] f bull 8 (b) db

bull TOTAL PROBABILITIES OF FAILURE

Total Pf - 1 bull Il [1 middotPf] - l Pf bull

Total Pf - 1 bull Il [1middot Pf] - l Pfmiddot

Total Pr = 1 - n [1middot PfbullJ = l Pfshy

I I I

EXAMPLE APPLICATION bull PHASE I

PLATFORM ST 151 K

bull GULF OF MEXICO

bull WATER DEPTH = 137 FT

bull ANNUAL MAXIMUM HS MEDIAN 175 ft COV 031

bull ASSUMED STORM DURATION 3 hoursmiddot

bull ASSUMED NUMBER OF WAVES 800

bull MEDIAN HS FOR STORM HOURS 1 2 AND 3

bull MEDIAN CURRENT VALUE bull

bull DETERMINE ULTIMATE CAPACITY FOR THREE DIRECTIONS

bull DETERMINE ANNUAL FAILURE PROBABILITIES

Broadsidemiddot Diagonal End-on

0037 0025 0007

0071 0044 0018

0022 0016 0004

DETERMINE TOTAL ANNUAL PROBABILITY OF FAILURE

Pf - 007

Pf - 013

Pr - 004

bull

EXAMPLE USE OF BIAS FACTORS - PHASE II middot

bull DETERMINE PREDICTIVE PROBABILITY OF FAILURE

Case - 1

Pf _ S S Pf (b b2) fa1a2 (b b2) db db2

g bull -1 (b1 R I S) - 1

92 - (b2 R2 I S) - 1

Case -2

Pf s ~ [bJ f11 (b) db

Pf [bJ p [g lt 0)

g - bull 1 (b1 R1 I S) - 1

1

bull1) l

I

BENEFITS OF ANDREW JIP bull PHASE II

lt

bull IMPROVED LOAD AND RESISTANCE RECIPE

bull IMPROVED CALIBRATION PROCEDURE

bull MULTIPLE BIAS FACTORS

bull PROCEDURE TO INCLUDE EFFECT OF OTHER PLATFORMS

bull PROCEDURE TO USE MULTIPLE BIAS FACTORS

  • Structure Bookmarks
    • ANDREW PHASE II JOINT1 INDUSTRY PROJECT
    • PROGRESS MEETING NUMBER 1
    • ANDREW PHASE II JOINT INDUSTRY PROJECT PROGRESS MEETING NUMBER 1 NOVEMBER 16 AGENDA
    • ANDREW PHASE II UST OF PARTICIPANTS
    • ANDREW PHASE I OBJECTIVES AND RESULTS
    • ANDREW PHASE II PROJECT OBJECTIVES
    • SUBSEQUENT PROJECTS
    • ANDREW PHASE II PROJECT TEAM
    • ANDREW PHASE II UPDATE MODELING middotAND ANALYSIS RECIPE
    • -ANDREW PHASE II CONDUCTOR RESISTANCE
    • ANDREW PHASE Il WA VE FORCE PROFILE
    • ANDREW PHASE II WA VE FORCE PROFILEmiddot
    • ANDREW PHASE II BRACE MODELING
    • ANDREW PHASE II FOUNDATIONS
    • ANDREW PHASE II middot middot JOINT MODELING
    • ANDREW PHASE Il JOINT MODELING OPTIONS DISCUSSION
    • ANDREW PHASE II JOINT MODEJJNG OPTIONS ILLUSTRATIONS
    • ANDREW PHASE II JOINT MODELING PHASE II METHOD
    • ANDREW PHASE II ANDREW STORM CONDmONS UPDATED
    • Platform Selection bullbullbull1111111111111
    • BasisCriteria for Selection bullbullbull1111111111111
    • CALIBRATION
    • WHERE TO INTRODUCE BIAS FACTOR (B
    • PHASE I -KEY CONCLUSIONS
    • bull CALIBRATION ISSUES PHASE II
    • PHASE II WORK SCOPE (CALIBRATION)
    • SOFTWARE
    • SENSITIVITY STUDIES POSTERIOR DISTRIBUTION OF B
    • WEIGHTING APPROACH bull
    • INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON
    • CATEGORIZATION OF ADDITIONAL PLATFORMS
    • UPDATING OF CALIBRATION RESULTS
    • APPLICATIONS OF BIAS FACTORS
    • EXAMPLE USE OF BIAS FACTORS
    • EXAMPLE APPLICATION bull PHASE I PLATFORM ST 151 K
    • EXAMPLE USE OF BIAS FACTORS -PHASE II middot
    • BENEFITS OF ANDREW JIP bull PHASE II
Page 38:  · . ' ANDREW PHASE II PROJECT OBJECTIVES . ' • ·UPDATE MODELING AND ANALYSIS RECIPE· JOINTS ' ' ·BRACES WAVE PROFILE FOUNDATIONS . ' ' CONDUCTORS· • ANALVZE PLATFORMS USING

ANDREW PHASE II

JOINT MODELING PHASE II METHOD

e

I I I

ANDREW PHASE II

ANDREW STORM CONDmONS UPDATED

bull -PHASE I USED OCEANWEATHER HINDCAST-(Nov 1992)

bull middotmiddotPHASE II WILL USE THE REVISED HINDCAST (Nov 1994)

bull lt c lt I lt

2-2

l---4-~~~+-~~-+-~~~1--~~-+-~~--l--~T--+-~~--i24

f-tJVtMfiz~ 1Cfl (r STvb(

Plollol bullbull 5middotNOYmiddot9214bull1J Irbullbull Illbull [ANDREVl~NDREVCUSTbullMAXll 5-NOY-1992 10bull54 - 9208

HEIGHT lml

I~ MAXI

I L 1~middotamp SZ~- ~A I 13c

I J) ( r I X~0iWJ~0gtJ gt L gt gt I 1c I 120

-- N

~ deg

I I I I I I e C I ~=-41 122 -94 -92 -90 -88 -86 -84 -82 -80

-

Figure 7 contours of maximum hindcast significant wave height

_

~

~tl ~

NOltNttIJ 17lt ltNS S iJJ)y

Maximum Significant Wave Height - Hurricane Andrew 1992

30 ~

26 ~ I ~-~~~- == -shy -~ 35-_

24deg1--shy --------~- middot-middot-------middotmiddotmiddot----middot------middot-middot~middot- - - ~-----1---middot

22middot ~~~~~~~-~ -94 -92 -90 -88 -84deg-86 -82 -somiddot

c

-

Platform Selection bullbullbull1111111111111

0 BasisCriteria for Selection middot

0 Review of Phase I Platforms middot 0 Alternative Platforms

0 PMB Recommendations

0 Participants Comments

gt

L

l I

BasisCriteria for Selection bullbullbull1111111111111

0 Benefit for Calibration - Classification (UnexpectedExpected)

- Failure Mode (StructuresFoundations)

- Damaged StructuresKnown response

- Better Representation of Full ~opulation

0 Availability of Data - Existing Models

- Inspection Data

- Damage Data

PREDICTED BEHAVIOR

-

OBSERVED

BEHAVIOR

-SURVIVED DAMAGED FAILED

SURVIVED

ST134W WD90A

-MC311 -

MC397 -

--

-

-

ST151K ST130Q

-

-

-

DAMAGED - -

-

-

--

-

-

ST52 SS139 ST177B -

ST161A

-_

-

FAILED

-

-

-

ST151H ST130A ST72

Operator - Platfonns - Platfonns Survived Damagedamp

- Falled -

bull

-

-

Jle eBDIGIPllDbl -Amocoshy 30 1 Chevron -

143 12 Exxon 72 Mobll 22 1 Phllllps 7 bull

-Shell 39 Unocal 25 1

Total Partlclnts Platfonns 338 15

Other Operators bull- 310 13

-

Total Platfonns - -

648 28

fbml

Platforms Considered 6 7

tie of Total Platfonns 1 25

OVERVIEW OF PLATFORMS SURVIVEDDAMAGEDFAILED

-

shy

v

Platforms Evaluated in Phase I

-

-Platfonn Water Year Number Analysis Avallabillty of Availability Platform

- Name Depth rl of Model DetaDed rl ClassUlcatlon

Installation legs Inspection Soll Report Based on Information Phase I Calibration - -ft

- -- Suaill flalfnrm ~BSi -STISIK 137 1963 8 CAP Yes Yes Un eel Sarvlval

-STIJOO 170 1964 4 CAPmiddotshy Yes ST134data u ST134W 137 1981 4 CAP Yes Yesmiddot E

-

edSarvlval -

middotsurvival

WD90A 184 - 1964 8 CAP Yes E -

- Sarvlvill

MC3ll 343 1978 8 KARMA

MC397 468 1991 4 KARMA -

bull ~reglmmiddot- -

Yes Sure Survival

bull Yes - Sare Sanlval-

-T23STS2 63 1969 4 CAP

T2SSS139 62 1969 4 CAP -

Yes

Yes

Yes Ex

Yes Une

bull SurvivalLikely to DamalI

middot bull SanlvalLlkely to DamalI

ST161A 118 1964 8 USFOS

Eail11a pound1al(bullllm Clss ST177B 142 --1965 8 CAP

STISIH 137 1964 - 8 CAP C

STIJOA 140 19S8 8 CAP - -

Yes

Yes

ollamed

Yes

- Yes E

-

at about 1 mUe

Yes

ST134data

bull SafvlvalLlkelv to Dam- bull- -

E middotrahre

Emeried Failure -

Likely FaUare

T21ST72 61 1969 4 CAP Collamed No Likely FaUure

Platforms Identified for Evaluating Their Likelihood of Falling Under UnexpeCted Survival Category

CllEllATOa

AllB

llLOCll

Sl1lvcnJllE

NAME

YEAR

INSTAUJCD

WATER

IJBPJB

middotmiddotDISTANCE

ftOMSllOllE

(MIUIS)

Ill

(M)

AMOCO BllW

STllI

WDI075

A

bull D

4

n uo 17J

44

32

bull 20

10

bull CllEVllOH STtllO c I 1bull 21 11

STl130 D 0 161 21 11 STllJ4 I 0 137 D - 11

STIUI G I 137 32 11

STIUI I 0 121 32 11

STIUI

J 0 140 32 11

STIUI It 4 137 32 11

STllSI PROl)I 0 137 32 11

STllSI PROD-2 0 137 32 11

STll7 A 0

140 35 11

WDl117 c 65 214 34 10

DXON STshy E 51 14 bull STl165 A t3 bull 34 WD0073

WD0073

WD0073

A _ c

4

62

4

IA IA

17J

22

22

22

bullbullbull WD003-shy E 65 161 Z7

A 0 IN 31 MOBD Bl12I

ssn A bull

50

50

31

7

bull 7

MIJllPllY SS114 B 7 50 7 7

PlllUIPS SSl14t A 55 33 7

SAMEDAN ST17J A 4 117 7 bull 7

SJIELL WD0103 A 65 223 27 WDllOJ bull 65 221 27 WDll04 c 65 221 27 WDOI04 D 67 2 27

UNOCAL

SSl20I

SSl20I

bull SSl20I

SS0209

SS0209

SS0209

SS0209

SS 0253

SS 0253

E

I

PtJllE

A

bull D

G A

c

0

4

4

7J

61

0

0

69

103

97

97

9S

100

95

100

165

175

34

34

34

34

34

34

34

54

54

bullbullbullbullbullbullbullbullbull

Platforms Recommended for Phase II

llalr y_ w- Aot Av- _ I 1 Doplla Model nm bull bull _ bull I I H11 s18-1 Irht lbwl~ bullbullbull(-illtdlel OM lllMtD

II ---

-STISIIC 137 1963 I CAP Yeo Yeo I Yeo

STtshy -170 1964 4 CAP Yeo ST134dlU I Yeo

STl34W 137 Yeo Yeo -1981 4 CAP D Yeo

WD90A 114 1964 I CAP Yeo - D

MC311 30 1971 I KARMA Yeo -sre-

MC397 middot 461 1991 4 ICARMA Yeo Sireshy - - r- -

__ T13ST5l 63 1969 4 CAP Yeo Yeobull middot- ~n- D Yeo _ Tl5SSl39 62 1969 4 CAP Yeo Yeo I Yeo

_ STIT711 142 1965 I CAP Yeo I Yeo middot-1shySTl61A Ill 1964 I USFOS Yeo Yeo - D bulloshy

- r-

STISIH 137 1964 I CAP y D y -_ STl31A 140 1951 I CAP Yeo STl34dlU I Yeo

T21Sll 61 1969 4 CAP No rar-middot I Yeo

~-rmiddot-middot SS209A 95 1967 16 CAP Yeo D -SSU4R

11 UI A 41 -in D SS 2741AIM CAP bull

_ _ -shy - -shy- -shy

CALIBRATION

bull PHASEIAPPROACH

bull CALIBRATION ISSUES

bull PHASE II WORK SCOPE

bull SOFlWARE AVAILABILITY

bull EFFECT OF ADDITIONAL PLATFORMS

bull WEIGHTING APPROACH

bull EXAMPLE APPLICATION middot

(l=Ofce Mode] Wind

~

fcurrent l -middot

~SW ~leldl4

ultimate resistance force I first

design load Illgt

i _

deflection

-

bull

4

Historical Hurrlcane Dariiage and Survival of Platforms

Components Required for Calibration (Petrauskas 1992)

PRIOR Distribution

PLATFORM EXPERIENCE

BAYESIAN Survivals and Failures

UPDATING

Waves Winds and Currents bull

POSTERIOR Distribution

- A_middotJA ~ Survival Data Failure Data

Result Result

Load Resistance

middotfl

-

Bayesian Updaing middotGeneral Description (Petrauskas 1992)

L Imiddot

WHERE TO INTRODUCE BIAS FACTOR (B

On Resistance (Rmue - B (RPREDICreD I

middotOn Loads (Smue middot - B (SPRE01cTEo

On Safety Margin (RSmue - B (RSPREDICTED

CALIBRATION METHODOLOGY

gt

r

FIELD OBSERVATIONS

Observed Behavior

survivals Damages

Failures

r

bull

CAPACITY ANALYSIS

Predicted Load and Resistance Predicted Failure Modes

Survival Damage Failure cases

RELIABILITY ANALYSIS

- Probability of Failure ~

Ukellhood Failure

BAYESIAN UPDATING

Prior Distnbution of B Combined Likelihood Function

Posterior Dlstnbution of B

BIAS FACTOR

Mean Value cov bull

Represents Modeling Uncertainties

Prior Distribution of B Likelihood Functionmiddot Observing Success case STISIK and Posterior Distribution of B

2 00

uo

160

140

120

e 100 L $ uoIshy

0 60 ~ ~ 040

020

0 00

0 00 0 50

_ I 00

b

I 50 2 00

-shy lk [bl STISIK)

--shy Pnor-f(b)

--shy Pos1enor r(bl ISIK)

Prior or B Mean bull 10 COV bull 030

Posterlor of Bbull Meanbull 136COV 0 16

2 50

Likelihood Function and Posterior Distribution Example Platform STISIK

- -

~

- Prior and Posterior Distributions or B Failure cases (ST177B STlSlH ST130A T21) Damage cases (T23 T25 ST161A) Success cases (STlSlK ST130Q ST134W WD90A MC311 MC397)

400

350

300

2 so 0-

I

bull bull

bull bull gt

- bullbull ~ bull bull I bull

_ bull bull bullbull I bull bull bullII V -

bull bull

I bull ~ bull

bullbull bullI bull bullI bull bull bull

middot -~

bull__ ~ middotI middotmiddot

L 2 00

e ~ I SO

-I 00

- 0 so

000 gt

0 00 020 0 40 060 0 80 I 00 120 140 160 110 200

b -

-

shy

Prior-r

- - - - Posterior fLlll failure cases

bull bull bull bull bull bull bull Posterior flall damage cases

- bull bull - bull middot Postenor flall success cases

Posterior flall success fulure damage cases

-

Posterior of B Mean= 119 COV 0101

Posterior Distribution or Blas Factor (8) bull Different Categories and All Cases Combined

shy

l

PHASE I - KEY CONCLUSIONS

bull GLOBAL CORRECTION FACTOR

bull 15 TO 20 CONSERVATISM MEAN)

bull UNEXPECTED SURVIVALS HAVE GREATEST IMPACT

bull DAMAGED CASES MORE INFORMATIVE

bull CALIBRATION ISSUES PHASE II

bull

EFFECT OF IMPROVED CAPACITY ANALYSIS RECIPE

bull APPLICABILITY TO MULTIPLE FAILURE MODES

bull NO FOUNDATION DAMAGE CASE INCLUDED

bull IMPROVED REPRESENTATION OF FULL ANDREW DISTRIBUTION

-bull NUMBER OF PLATFORMS CONSIDERED

I 1 I

I t ) bull ~

PHASE II WORK SCOPE (CALIBRATION)

bull

bull ESTABLISH MULTIPLE BIAS FACTORS

Use Nine Base Case Platforms

middot - Capacity Analysis for Revised recipe

Foundation Blas Factor from Foundation JIP

bull DEVELOP WEIGHTING APPROACH

bull UPDATE MULTIPLE BIAS FACTORS

bullEffect of Platforms Not In the Nine Base Cases

bull EXAMPLE APPLICATION OF BIAS FACTORS I

bullbull

PHASE I PHASE II

CAPACITYbull ANALYSISRu bull

---------------------1----middot

Jc1 - iJbullraquoJ dbull

FttlS frt[f~l ratdhJJ

-

LIKELIHOOD FUNCTION

DEVELOPMENT

lk Jb Jfall1N) bull Jb)

111 Ill Isurvtvoq bull 1 - bl

-middot -------------------shy ~---------------------middot-middot

r111ir8 Cb1 111 Cbl BAYESIAN

UPDATING

~cove

R1R2 bull middot---------------------~----

Jblbull P(g cOJ

1 bull JbRI (MBSI I bull 1

~middot

llcJllJfalluro) bull P(gcOJ Ill 111 b21fallun) bull P ((g1 CO) U 1112 c OD

middot--------------------shy ----------------------shy bull

r111111r8 Cbl 111 Ill i-1nro1

deg ~1b2)laquottfrr (b1b21-lnfo) re Ill bull 1-11middot b2) bull lodegamp

middot

i1 covamp i2 cov~i

-

shy

--

Likelihood Function vs b All Success Case STISI K STl30Q STl34W WD90A MC311 MC397

I 0000

0 9000

OIOOO

0 7000

~ 0amp000

ta osoOO

~ 0 4000

0 3000

0 2000

0 1000

bull 0 0000

bull

I

middot

middot bull I

Ibull

-sibull ~ -middot-middot-middot-middot -middot-

----7----+-----~ I

Ibull bull bull I

rbull

--- Jk ISIK[blsuccess)

- bull - bull - lk 130Q[blsuccess) bull

lk 134W[blsuccess)

Jk-WD90A(blsuccc~I

- - - lk MC397[blsucccss)

--ltgt-- lk middotMC31 l[blsucccss)

--- Jk [bl 6 out of6succcsHascs)

05 I 5 2 250

b

Likelihood Functionsmiddot Success Cases - Phbullsc L

shy

bullbull bull bull bull bull bull -------------------~

bull bull bullbull bullbull bullbull bullbull bullbull bullbull bull

bull bullbull

bullbullbullbull

bullbull

bullbullbullbullbullbullbull

bullbullbullbullbullbull

bullbullbullbullbullbullbull

z 0

t z ) LL

c 0 0c i ~ i

J

~ sect ~--------------~ e l5

I

I

SOFTWARE

bull PF PROGRAM

bull CALREL

bull PF +CALREL

bull RELACS

bull

gt i

l

SENSITIVITY STUDIES

POSTERIOR DISTRIBUTION OF B

bull EFFECT OF CHANGE IN DISTRIBUTION OF PRIOR OF B

bull EFFECT OF REMOVAL OF ONE PLATFORM FROM EACH CATEGORY

bull EFFECT OF REMOVAL OF THREE DAMAGE CASES

bull EFFECT OF ADDITIONAL 13 SIMILAR PLATFORMS

-I

Prior and Posterior Distributions of B Variation In COV of PriorofB fromOl to04 -- -

4500

-4000

-

- -

3500 --middot - bull bull - bull bull Prior-r ICOV =02

3000

- - - - Prior-r ICOV =03 -

bull -~ e 2soo -

bull bull bull bull bull bull bull Pnor-r ICOV =o 4t frJ ~ 2000 Posterior rlCOV =0 2

bull bull Postenor rlCOV =0 3 bull bull 1500 bull middot bull - shy -middot- Posterior rlCOV =04

middot~ I 000 - bull

-middotmiddot ~ ~middotmiddot M -~

- middot 0 500

wbullbullbull _ I middot~ middotmiddotmiddotmiddotmiddot middotmiddotmiddotmiddotmiddot - -~ ~ __ middot_ _ bull -shy0 000

0 000 0200 0 400 0 600 0 800 IOOO 1200 1400 1600 1800 2000 -

- - b

-

=shy

Posterior Distribution of Bias Factor (B) bull Effect of Variation In COV of Prior of B

bull

bullPrior and Posterior Distributions of B Evaluating Effect of Removal of One Platform from Each Category from the Base Case of 13 Platforms

400

~ ~ 350

j IA i

3 00 bull I bull I ---Prior-r I bull l

250 1 I 1

I bull bull ---Posterior r lall 13 casesshy- ~E 2 00 bull bull bull bull bull bull bull Posterior r without STISIK

- ~ 1 bull - bull I - bull bull - bull bull Posterior r Iwithout T2I I 50 I

J-tt~i - bull - bull - Posterior r Iwithout T2S

tOO [-

I I ~ ~ I 0 50

0 00

___ r J i

l~T

~--

0 00 0 20 0 40 0 60 080 I 00 120 140 160 110 2 00

b

zshy

Posterior Distribution of Bias Factor (8) bull Effect of Removal of middotPlatform from Any Category

bullPrior and Posterior Distributions or B Evaluating Effect or Removal or 3 Damage Cases from the Base Case or 13 Platrorms

4

35

3

25

e L 2 ~ bulle shy

I 15

bull I

05

0

0 02 04 06 I 2 I 4 16 IS 2

I I

I

I

I Ibull

J -tmiddot L--

bullI

~

I

fl

II ~

I I v I

middot bull bull

bull L

~ ll- I-~lt

OS I

b

I

bull

1middot bullbullbullbull

~ __

bull

- - - - Prior-r

- - - - Postenor flall failure cases

Postenor flall success cases

--- Posterior flall 13 cases

---Posterior fI success failure cases only

IPosterior or B Mean =t11 COV =0121

-

_I

Posterir Distribution or Bias Factor (B) bull Effect or Removal or 3-Damage Cases

-

bullPrior and Posterior Distributions or B Evaluating Effect or Additional 13 Similar Observations as

Used in Andrew JIP The POsterior or B from Andrew-JIP Considered as New Prior or B

- 500

450

400

350

bull bull

bull bull bull bull J bull

bull bull f ~ I

I ~ Ibull bull

bullbull bull

l bull I -bullbull bull bull

I

I bull bull

bull I I~~bull bull bull

bull ~

I bull bull

~ ~_ ~~~

- - New Prior r bull

Posterior 114 new failure cases ~ 300- bull bull bull bull bull bull bull Posterior f 13 new damage casesl 150 ~

- bull bull - bull bull Posterior 116 new success cases i 2 00

Postenor 1113 new success failure -150 damage cases

I 00

0 50 -

0 00 0 00 0 20 0 40 0 60 0 80 I 00 120 I 40 160 110 2 00

Posterior ol B Mean= 120 COV= 0071 b

shy

shy

_rshy

Posterior Distribution of Bias Factor (B) bull Effect of Additional 13-Similar Observations

1

1 I

WEIGHTING APPROACH bull

bull ASSESSMENT OF PLATFORMS DATA

Select Platforms Most Mappable to the Analyzed Platforms

Unlikely SurvivalDamage Cases Identification

Based on Cut-Off Criteria Agemiddot Hs etc

bull INFORMATION REQUIRED FOR COMPARISON

bull CATEGORIZATION OF ADDITIONALmiddot PLATFORMS

bull UPDATING OF CALIBRATION RESULTS

--

OVERVIEW OF PLATFORMS SURVIVEDDAMAGED -~

Operator -

- Platforms

Survived CS)

Platforms

Damaged CD)

Platforms Used In

Phase IJIP -

JI~ ~Bill~IRBDll -

Amoco -

Chevron -

Exxon

Mob II

Phllllps

Shell

Unocal -

30

143

72 22

7

39 25shy -

1

12

bull

1 bull

1

1(S)+1 (D)

3 (S) +3(D)

2 (S)

bull shy

bull

bull

bull

Total Partlclnants Platforms 338 - 15 10C6S+4D) -

Other Operators

310

13

3 (D)

Total Platforms 648 28 13C6S+7D) -

Summary of Andrew JIP bull Phase I Platforms

Plairshy w v- N-ber Lea-le 1 shy car Hind_ o-w N-bull Deplh fl fl IJpe la bull n Yeadon Jadtet lbelledtf liefalit -- _ -OD

IL H - -shy _ICllshy IL IL - - rQClto ~ _

K STUIK 137 1963 8 365 608S4 Cellardeck

-middot STl-shy 170 1964 4 -~ K 39 62269 Cellardeck - --ST134W 137 1911 4 43 608S4 Smvlnlshy- _ - -

WD90A 184 1964 8 K 267Sbull34 83 SOS31 SUboftllar deck SmvMI

MC311 343 1978 8 x S7 62 - MC397 468 1991 4 x shy S33bull6SO 65911 -shySmeSmvlvol -- gt -

r-- T23ST52 63 1969 4 ton- shyKshy 52594 S03 -T25SSl39 62 1969 4 Kshy SOS94 S063 - middot ton-shy__ -

bullbull a ton--shyST161A 118 1964 8 K - 347S 51495 Cellarclecti

r~

ST1771 142 1965 8 K 39 60167 CeDordeck failme

STISIH 137 1964 8 K 395 608S4 CeDardeck Failme

K ST130A 140 1958 a SS 6095 I lblv Pai1uno - T21ST72 61 1969 4 Kshy S4S94 4974 I lhlv Pai1uno -

- -

JE

INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON

bull WATER DEPTH

bull bull YEAR OF INSTALLATIONDESIGN

bull PLATFORM ORIENTATION

bull PLATFORM LATITUDE AND LONGITUDE

bull KEY SKETCHES OF PLATFORM STRUCTURAL FRAMING

Number of Legs Plies Leg Batter Number of Conductors bull Deck Elevations (Including sub-cellar deck If any) Jacket Framing Elevations Brace type In Vertical Frames Sizes of Primary Members (leg pile vertical frames braces primary horizontals) Joint Details (can gap etc) LegPiie Annulus Grouting Status

bull middotPOST ANDREW INSPECTION REPORT Significant Wave Height Hs (ft)

bull INFORMATION ON DAMAGES PRIOR TO ANDREW

bull RESULTS FROM IN-HOUSE STUDIES IF AVAILABLE

bull ~RIGINAL DESIGN CRITERIA INFORMATION IF AVAILABLE

bull SOIL SHEAR STRENGTH PROFILE IF AVAILABLE

Platforms Identified for Evaluating Therr Likelihood of Falling Under Unexpected Survival Category

OPlRATOR

middotADA

BLOClt

llmucnJllE

NAME

YEAll

INSTAUJD

WA1ER

DBPTll

DISTANCE

JlllOM SHOllE

(MILU)

Bl

(M)

AMOCO 111113

STl11

WDI075

A

bull D

St

17

120

1n

44

20

10 I

CllllVllQfl STlllO c 61 110 21 11 STIUO D Q 1110 21 11

STl134 p Q 137 29 11

ST0151 G 61 137 11

ST0151 I 3 121 11

ST0151 J Q 140 11

STl151 It 137 11

ST0151 PJl001 Q UT 11

ST0151 PROD-2 Q UT 11

ST017 A 3 140 35 11

WDIUT c 65 214 34 10

llXXON STI055 II 51 6T 14 I

STl15 A t3 34

WDI073

WD0073

Amiddotshy Q

1111 1111

22

22

I

I

WDI073 c 1n 22 I

- WDOCl93-shy II

65 1110 2T A 3 194 31 MOBU 111121

SS IOl7

A B

501

50

50

31

T

I

T 6

MtlllPllY SS114 B T 50 T T

PlllllIPS SS0149 A 5 55 33 T

SAMllDAN STln A 107 T T

SBllLL WDl103 A 65 223 2T WDl103 B 65 221 2T WD0104 c 65 221 2T WD0104 D 6T 20 2T

UNOCAL SS020I

SS 020I SSl20I

SS 0209

SS0209

II p

PURll

A

B

3

n 61

103

97

97

95

100

34

34

34

34 34

bullbull bullbullbull

- SS0209 D u 95 34 I

- SS 1209 SSl253

SS0253

G A

c

6T

2

100

165

1T5

34

54

54

bullbull I

I I

I I

SOUTH MARCH

~LAND

N

SOUTH PaTOI ENE ISWltD AREA

MMS Defined Limits ror Offshore Structures Requiring Inspections Following Andrew

t---Jbulln3 T25bull I bullTtl

bull1$1Htl1Abull bull1SIK1779bull

~

middot-DaTA AREA~T

CRANO rsLE

AREA bull1300

bullll4W

=shy

bullPath of Hurricane Andrew with Platforms Used In Calibration

Likelihood Function vs b All Success Case STISI K STIJOQ ST134W WD90A MCJll MCJ97

I 0000

0 9000

0 8000

07000

-Bo6000

tiiii 0 5000 -

04000

0 3000

0 2000

01000

0 0000

0 05 25

ri1rt middot-middot-middot-middot-middot-

I 5 2

b

---11 middot151K(blsu~ccsI

- bull - bull - fk 130Q(bfU~lCSS)

lie bull134W(blsuccc]

lk-WD90A[blsultCS]

- - - lk middotMC397(bfUtlCS]

--o-- lk-MCJll(bbuctcsJ

--- ll (bl 6 out of 6 Mlttcs case]

_

( Likelihood Functions middotSuccess Cases - p - 1 c I

CATEGORIZATION OF ADDITIONAL PLATFORMS

I I I

bull CATEGORIES

Unexpected Survivals

Expected Survivals

bull APPROACH

Identify Corresponding Platform(s) Evaluated In the JIP

Determine Max Wave Height at Platform Location

Evaluate Whether Wave Hit the Deck

middot - Evaluate Likelihood of R and S to Differ from the Base Cases

Review Joint Information

Assess Special Features

bull IDENTIFY NUMBER OF PLATFORMS IN EACH CATEGORY

I

UPDATING OF CALIBRATION RESULTS

bull INCLUDE EFFECT OF ADDITIONAL PLATFORMS ON bullBbull

Consider Increased 1Uncertalntles In R and S

Develop bullNeW Ukellhood Functions for Different Categories

middotLet bullxbull Number of Ukely Unexpected Survival Cases and y Number of Ukely Expected Survival Cases

Determine Combined Likelihood Function lk (b Iadditional platforms) middot

bull REVISED LIKELIHOOD FUNCTION

lk(b I all cases) = lk(b I 9 platforms) middot x lk(b Ix platforms) x lk(b I y platforms)

bull ESTABLISH REVISED BIAS FACTOR B

f8 (b) u f8 (b) lk (b I all cases)

II

t t ~

shy

APPLICATIONS OF BIAS FACTORS

ii APPLICABLE TO PARTICULAR LOAD AND RESISTANCE RECIPE

bull USE BY THE API COMMITTEES AND MMS

Criteria Revisions

Recommended Design Guidelines Revisions bull

bullbull REASSESSMENT STUDIES AND DECISION MAKING

middot

bull RISK ASSESSMENT STUDIES BY THE OPERATORS

Individual Platforms

Fleet of Platforms

middot IMR Program Implementation

bullbull NOT TO BE USED IN DESIGN

I I fl J

EXAMPLE USE OF BIAS FACTORS

bull PHASE I EXAMPLE APPLICATION

bull EVALUATION OF ANNUAL PROBABILITY OF FAILURE

Establish Annual Maximum Seastate Data

Determine Coefficients to Define Base Shear

Determine Ultimate Capacity of Platform

Determine Probablllty of Failure Versus bullbbull

-bull DETERMINE UPDATED PROBABILITY OF FAILURE

Pf - P [S gt R]

Pf I P [~ gt Rb I b] f 8 (b) db

Pr - J P [S gt Rb I b] f bull 8 (b) db

bull TOTAL PROBABILITIES OF FAILURE

Total Pf - 1 bull Il [1 middotPf] - l Pf bull

Total Pf - 1 bull Il [1middot Pf] - l Pfmiddot

Total Pr = 1 - n [1middot PfbullJ = l Pfshy

I I I

EXAMPLE APPLICATION bull PHASE I

PLATFORM ST 151 K

bull GULF OF MEXICO

bull WATER DEPTH = 137 FT

bull ANNUAL MAXIMUM HS MEDIAN 175 ft COV 031

bull ASSUMED STORM DURATION 3 hoursmiddot

bull ASSUMED NUMBER OF WAVES 800

bull MEDIAN HS FOR STORM HOURS 1 2 AND 3

bull MEDIAN CURRENT VALUE bull

bull DETERMINE ULTIMATE CAPACITY FOR THREE DIRECTIONS

bull DETERMINE ANNUAL FAILURE PROBABILITIES

Broadsidemiddot Diagonal End-on

0037 0025 0007

0071 0044 0018

0022 0016 0004

DETERMINE TOTAL ANNUAL PROBABILITY OF FAILURE

Pf - 007

Pf - 013

Pr - 004

bull

EXAMPLE USE OF BIAS FACTORS - PHASE II middot

bull DETERMINE PREDICTIVE PROBABILITY OF FAILURE

Case - 1

Pf _ S S Pf (b b2) fa1a2 (b b2) db db2

g bull -1 (b1 R I S) - 1

92 - (b2 R2 I S) - 1

Case -2

Pf s ~ [bJ f11 (b) db

Pf [bJ p [g lt 0)

g - bull 1 (b1 R1 I S) - 1

1

bull1) l

I

BENEFITS OF ANDREW JIP bull PHASE II

lt

bull IMPROVED LOAD AND RESISTANCE RECIPE

bull IMPROVED CALIBRATION PROCEDURE

bull MULTIPLE BIAS FACTORS

bull PROCEDURE TO INCLUDE EFFECT OF OTHER PLATFORMS

bull PROCEDURE TO USE MULTIPLE BIAS FACTORS

  • Structure Bookmarks
    • ANDREW PHASE II JOINT1 INDUSTRY PROJECT
    • PROGRESS MEETING NUMBER 1
    • ANDREW PHASE II JOINT INDUSTRY PROJECT PROGRESS MEETING NUMBER 1 NOVEMBER 16 AGENDA
    • ANDREW PHASE II UST OF PARTICIPANTS
    • ANDREW PHASE I OBJECTIVES AND RESULTS
    • ANDREW PHASE II PROJECT OBJECTIVES
    • SUBSEQUENT PROJECTS
    • ANDREW PHASE II PROJECT TEAM
    • ANDREW PHASE II UPDATE MODELING middotAND ANALYSIS RECIPE
    • -ANDREW PHASE II CONDUCTOR RESISTANCE
    • ANDREW PHASE Il WA VE FORCE PROFILE
    • ANDREW PHASE II WA VE FORCE PROFILEmiddot
    • ANDREW PHASE II BRACE MODELING
    • ANDREW PHASE II FOUNDATIONS
    • ANDREW PHASE II middot middot JOINT MODELING
    • ANDREW PHASE Il JOINT MODELING OPTIONS DISCUSSION
    • ANDREW PHASE II JOINT MODEJJNG OPTIONS ILLUSTRATIONS
    • ANDREW PHASE II JOINT MODELING PHASE II METHOD
    • ANDREW PHASE II ANDREW STORM CONDmONS UPDATED
    • Platform Selection bullbullbull1111111111111
    • BasisCriteria for Selection bullbullbull1111111111111
    • CALIBRATION
    • WHERE TO INTRODUCE BIAS FACTOR (B
    • PHASE I -KEY CONCLUSIONS
    • bull CALIBRATION ISSUES PHASE II
    • PHASE II WORK SCOPE (CALIBRATION)
    • SOFTWARE
    • SENSITIVITY STUDIES POSTERIOR DISTRIBUTION OF B
    • WEIGHTING APPROACH bull
    • INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON
    • CATEGORIZATION OF ADDITIONAL PLATFORMS
    • UPDATING OF CALIBRATION RESULTS
    • APPLICATIONS OF BIAS FACTORS
    • EXAMPLE USE OF BIAS FACTORS
    • EXAMPLE APPLICATION bull PHASE I PLATFORM ST 151 K
    • EXAMPLE USE OF BIAS FACTORS -PHASE II middot
    • BENEFITS OF ANDREW JIP bull PHASE II
Page 39:  · . ' ANDREW PHASE II PROJECT OBJECTIVES . ' • ·UPDATE MODELING AND ANALYSIS RECIPE· JOINTS ' ' ·BRACES WAVE PROFILE FOUNDATIONS . ' ' CONDUCTORS· • ANALVZE PLATFORMS USING

I I I

ANDREW PHASE II

ANDREW STORM CONDmONS UPDATED

bull -PHASE I USED OCEANWEATHER HINDCAST-(Nov 1992)

bull middotmiddotPHASE II WILL USE THE REVISED HINDCAST (Nov 1994)

bull lt c lt I lt

2-2

l---4-~~~+-~~-+-~~~1--~~-+-~~--l--~T--+-~~--i24

f-tJVtMfiz~ 1Cfl (r STvb(

Plollol bullbull 5middotNOYmiddot9214bull1J Irbullbull Illbull [ANDREVl~NDREVCUSTbullMAXll 5-NOY-1992 10bull54 - 9208

HEIGHT lml

I~ MAXI

I L 1~middotamp SZ~- ~A I 13c

I J) ( r I X~0iWJ~0gtJ gt L gt gt I 1c I 120

-- N

~ deg

I I I I I I e C I ~=-41 122 -94 -92 -90 -88 -86 -84 -82 -80

-

Figure 7 contours of maximum hindcast significant wave height

_

~

~tl ~

NOltNttIJ 17lt ltNS S iJJ)y

Maximum Significant Wave Height - Hurricane Andrew 1992

30 ~

26 ~ I ~-~~~- == -shy -~ 35-_

24deg1--shy --------~- middot-middot-------middotmiddotmiddot----middot------middot-middot~middot- - - ~-----1---middot

22middot ~~~~~~~-~ -94 -92 -90 -88 -84deg-86 -82 -somiddot

c

-

Platform Selection bullbullbull1111111111111

0 BasisCriteria for Selection middot

0 Review of Phase I Platforms middot 0 Alternative Platforms

0 PMB Recommendations

0 Participants Comments

gt

L

l I

BasisCriteria for Selection bullbullbull1111111111111

0 Benefit for Calibration - Classification (UnexpectedExpected)

- Failure Mode (StructuresFoundations)

- Damaged StructuresKnown response

- Better Representation of Full ~opulation

0 Availability of Data - Existing Models

- Inspection Data

- Damage Data

PREDICTED BEHAVIOR

-

OBSERVED

BEHAVIOR

-SURVIVED DAMAGED FAILED

SURVIVED

ST134W WD90A

-MC311 -

MC397 -

--

-

-

ST151K ST130Q

-

-

-

DAMAGED - -

-

-

--

-

-

ST52 SS139 ST177B -

ST161A

-_

-

FAILED

-

-

-

ST151H ST130A ST72

Operator - Platfonns - Platfonns Survived Damagedamp

- Falled -

bull

-

-

Jle eBDIGIPllDbl -Amocoshy 30 1 Chevron -

143 12 Exxon 72 Mobll 22 1 Phllllps 7 bull

-Shell 39 Unocal 25 1

Total Partlclnts Platfonns 338 15

Other Operators bull- 310 13

-

Total Platfonns - -

648 28

fbml

Platforms Considered 6 7

tie of Total Platfonns 1 25

OVERVIEW OF PLATFORMS SURVIVEDDAMAGEDFAILED

-

shy

v

Platforms Evaluated in Phase I

-

-Platfonn Water Year Number Analysis Avallabillty of Availability Platform

- Name Depth rl of Model DetaDed rl ClassUlcatlon

Installation legs Inspection Soll Report Based on Information Phase I Calibration - -ft

- -- Suaill flalfnrm ~BSi -STISIK 137 1963 8 CAP Yes Yes Un eel Sarvlval

-STIJOO 170 1964 4 CAPmiddotshy Yes ST134data u ST134W 137 1981 4 CAP Yes Yesmiddot E

-

edSarvlval -

middotsurvival

WD90A 184 - 1964 8 CAP Yes E -

- Sarvlvill

MC3ll 343 1978 8 KARMA

MC397 468 1991 4 KARMA -

bull ~reglmmiddot- -

Yes Sure Survival

bull Yes - Sare Sanlval-

-T23STS2 63 1969 4 CAP

T2SSS139 62 1969 4 CAP -

Yes

Yes

Yes Ex

Yes Une

bull SurvivalLikely to DamalI

middot bull SanlvalLlkely to DamalI

ST161A 118 1964 8 USFOS

Eail11a pound1al(bullllm Clss ST177B 142 --1965 8 CAP

STISIH 137 1964 - 8 CAP C

STIJOA 140 19S8 8 CAP - -

Yes

Yes

ollamed

Yes

- Yes E

-

at about 1 mUe

Yes

ST134data

bull SafvlvalLlkelv to Dam- bull- -

E middotrahre

Emeried Failure -

Likely FaUare

T21ST72 61 1969 4 CAP Collamed No Likely FaUure

Platforms Identified for Evaluating Their Likelihood of Falling Under UnexpeCted Survival Category

CllEllATOa

AllB

llLOCll

Sl1lvcnJllE

NAME

YEAR

INSTAUJCD

WATER

IJBPJB

middotmiddotDISTANCE

ftOMSllOllE

(MIUIS)

Ill

(M)

AMOCO BllW

STllI

WDI075

A

bull D

4

n uo 17J

44

32

bull 20

10

bull CllEVllOH STtllO c I 1bull 21 11

STl130 D 0 161 21 11 STllJ4 I 0 137 D - 11

STIUI G I 137 32 11

STIUI I 0 121 32 11

STIUI

J 0 140 32 11

STIUI It 4 137 32 11

STllSI PROl)I 0 137 32 11

STllSI PROD-2 0 137 32 11

STll7 A 0

140 35 11

WDl117 c 65 214 34 10

DXON STshy E 51 14 bull STl165 A t3 bull 34 WD0073

WD0073

WD0073

A _ c

4

62

4

IA IA

17J

22

22

22

bullbullbull WD003-shy E 65 161 Z7

A 0 IN 31 MOBD Bl12I

ssn A bull

50

50

31

7

bull 7

MIJllPllY SS114 B 7 50 7 7

PlllUIPS SSl14t A 55 33 7

SAMEDAN ST17J A 4 117 7 bull 7

SJIELL WD0103 A 65 223 27 WDllOJ bull 65 221 27 WDll04 c 65 221 27 WDOI04 D 67 2 27

UNOCAL

SSl20I

SSl20I

bull SSl20I

SS0209

SS0209

SS0209

SS0209

SS 0253

SS 0253

E

I

PtJllE

A

bull D

G A

c

0

4

4

7J

61

0

0

69

103

97

97

9S

100

95

100

165

175

34

34

34

34

34

34

34

54

54

bullbullbullbullbullbullbullbullbull

Platforms Recommended for Phase II

llalr y_ w- Aot Av- _ I 1 Doplla Model nm bull bull _ bull I I H11 s18-1 Irht lbwl~ bullbullbull(-illtdlel OM lllMtD

II ---

-STISIIC 137 1963 I CAP Yeo Yeo I Yeo

STtshy -170 1964 4 CAP Yeo ST134dlU I Yeo

STl34W 137 Yeo Yeo -1981 4 CAP D Yeo

WD90A 114 1964 I CAP Yeo - D

MC311 30 1971 I KARMA Yeo -sre-

MC397 middot 461 1991 4 ICARMA Yeo Sireshy - - r- -

__ T13ST5l 63 1969 4 CAP Yeo Yeobull middot- ~n- D Yeo _ Tl5SSl39 62 1969 4 CAP Yeo Yeo I Yeo

_ STIT711 142 1965 I CAP Yeo I Yeo middot-1shySTl61A Ill 1964 I USFOS Yeo Yeo - D bulloshy

- r-

STISIH 137 1964 I CAP y D y -_ STl31A 140 1951 I CAP Yeo STl34dlU I Yeo

T21Sll 61 1969 4 CAP No rar-middot I Yeo

~-rmiddot-middot SS209A 95 1967 16 CAP Yeo D -SSU4R

11 UI A 41 -in D SS 2741AIM CAP bull

_ _ -shy - -shy- -shy

CALIBRATION

bull PHASEIAPPROACH

bull CALIBRATION ISSUES

bull PHASE II WORK SCOPE

bull SOFlWARE AVAILABILITY

bull EFFECT OF ADDITIONAL PLATFORMS

bull WEIGHTING APPROACH

bull EXAMPLE APPLICATION middot

(l=Ofce Mode] Wind

~

fcurrent l -middot

~SW ~leldl4

ultimate resistance force I first

design load Illgt

i _

deflection

-

bull

4

Historical Hurrlcane Dariiage and Survival of Platforms

Components Required for Calibration (Petrauskas 1992)

PRIOR Distribution

PLATFORM EXPERIENCE

BAYESIAN Survivals and Failures

UPDATING

Waves Winds and Currents bull

POSTERIOR Distribution

- A_middotJA ~ Survival Data Failure Data

Result Result

Load Resistance

middotfl

-

Bayesian Updaing middotGeneral Description (Petrauskas 1992)

L Imiddot

WHERE TO INTRODUCE BIAS FACTOR (B

On Resistance (Rmue - B (RPREDICreD I

middotOn Loads (Smue middot - B (SPRE01cTEo

On Safety Margin (RSmue - B (RSPREDICTED

CALIBRATION METHODOLOGY

gt

r

FIELD OBSERVATIONS

Observed Behavior

survivals Damages

Failures

r

bull

CAPACITY ANALYSIS

Predicted Load and Resistance Predicted Failure Modes

Survival Damage Failure cases

RELIABILITY ANALYSIS

- Probability of Failure ~

Ukellhood Failure

BAYESIAN UPDATING

Prior Distnbution of B Combined Likelihood Function

Posterior Dlstnbution of B

BIAS FACTOR

Mean Value cov bull

Represents Modeling Uncertainties

Prior Distribution of B Likelihood Functionmiddot Observing Success case STISIK and Posterior Distribution of B

2 00

uo

160

140

120

e 100 L $ uoIshy

0 60 ~ ~ 040

020

0 00

0 00 0 50

_ I 00

b

I 50 2 00

-shy lk [bl STISIK)

--shy Pnor-f(b)

--shy Pos1enor r(bl ISIK)

Prior or B Mean bull 10 COV bull 030

Posterlor of Bbull Meanbull 136COV 0 16

2 50

Likelihood Function and Posterior Distribution Example Platform STISIK

- -

~

- Prior and Posterior Distributions or B Failure cases (ST177B STlSlH ST130A T21) Damage cases (T23 T25 ST161A) Success cases (STlSlK ST130Q ST134W WD90A MC311 MC397)

400

350

300

2 so 0-

I

bull bull

bull bull gt

- bullbull ~ bull bull I bull

_ bull bull bullbull I bull bull bullII V -

bull bull

I bull ~ bull

bullbull bullI bull bullI bull bull bull

middot -~

bull__ ~ middotI middotmiddot

L 2 00

e ~ I SO

-I 00

- 0 so

000 gt

0 00 020 0 40 060 0 80 I 00 120 140 160 110 200

b -

-

shy

Prior-r

- - - - Posterior fLlll failure cases

bull bull bull bull bull bull bull Posterior flall damage cases

- bull bull - bull middot Postenor flall success cases

Posterior flall success fulure damage cases

-

Posterior of B Mean= 119 COV 0101

Posterior Distribution or Blas Factor (8) bull Different Categories and All Cases Combined

shy

l

PHASE I - KEY CONCLUSIONS

bull GLOBAL CORRECTION FACTOR

bull 15 TO 20 CONSERVATISM MEAN)

bull UNEXPECTED SURVIVALS HAVE GREATEST IMPACT

bull DAMAGED CASES MORE INFORMATIVE

bull CALIBRATION ISSUES PHASE II

bull

EFFECT OF IMPROVED CAPACITY ANALYSIS RECIPE

bull APPLICABILITY TO MULTIPLE FAILURE MODES

bull NO FOUNDATION DAMAGE CASE INCLUDED

bull IMPROVED REPRESENTATION OF FULL ANDREW DISTRIBUTION

-bull NUMBER OF PLATFORMS CONSIDERED

I 1 I

I t ) bull ~

PHASE II WORK SCOPE (CALIBRATION)

bull

bull ESTABLISH MULTIPLE BIAS FACTORS

Use Nine Base Case Platforms

middot - Capacity Analysis for Revised recipe

Foundation Blas Factor from Foundation JIP

bull DEVELOP WEIGHTING APPROACH

bull UPDATE MULTIPLE BIAS FACTORS

bullEffect of Platforms Not In the Nine Base Cases

bull EXAMPLE APPLICATION OF BIAS FACTORS I

bullbull

PHASE I PHASE II

CAPACITYbull ANALYSISRu bull

---------------------1----middot

Jc1 - iJbullraquoJ dbull

FttlS frt[f~l ratdhJJ

-

LIKELIHOOD FUNCTION

DEVELOPMENT

lk Jb Jfall1N) bull Jb)

111 Ill Isurvtvoq bull 1 - bl

-middot -------------------shy ~---------------------middot-middot

r111ir8 Cb1 111 Cbl BAYESIAN

UPDATING

~cove

R1R2 bull middot---------------------~----

Jblbull P(g cOJ

1 bull JbRI (MBSI I bull 1

~middot

llcJllJfalluro) bull P(gcOJ Ill 111 b21fallun) bull P ((g1 CO) U 1112 c OD

middot--------------------shy ----------------------shy bull

r111111r8 Cbl 111 Ill i-1nro1

deg ~1b2)laquottfrr (b1b21-lnfo) re Ill bull 1-11middot b2) bull lodegamp

middot

i1 covamp i2 cov~i

-

shy

--

Likelihood Function vs b All Success Case STISI K STl30Q STl34W WD90A MC311 MC397

I 0000

0 9000

OIOOO

0 7000

~ 0amp000

ta osoOO

~ 0 4000

0 3000

0 2000

0 1000

bull 0 0000

bull

I

middot

middot bull I

Ibull

-sibull ~ -middot-middot-middot-middot -middot-

----7----+-----~ I

Ibull bull bull I

rbull

--- Jk ISIK[blsuccess)

- bull - bull - lk 130Q[blsuccess) bull

lk 134W[blsuccess)

Jk-WD90A(blsuccc~I

- - - lk MC397[blsucccss)

--ltgt-- lk middotMC31 l[blsucccss)

--- Jk [bl 6 out of6succcsHascs)

05 I 5 2 250

b

Likelihood Functionsmiddot Success Cases - Phbullsc L

shy

bullbull bull bull bull bull bull -------------------~

bull bull bullbull bullbull bullbull bullbull bullbull bullbull bull

bull bullbull

bullbullbullbull

bullbull

bullbullbullbullbullbullbull

bullbullbullbullbullbull

bullbullbullbullbullbullbull

z 0

t z ) LL

c 0 0c i ~ i

J

~ sect ~--------------~ e l5

I

I

SOFTWARE

bull PF PROGRAM

bull CALREL

bull PF +CALREL

bull RELACS

bull

gt i

l

SENSITIVITY STUDIES

POSTERIOR DISTRIBUTION OF B

bull EFFECT OF CHANGE IN DISTRIBUTION OF PRIOR OF B

bull EFFECT OF REMOVAL OF ONE PLATFORM FROM EACH CATEGORY

bull EFFECT OF REMOVAL OF THREE DAMAGE CASES

bull EFFECT OF ADDITIONAL 13 SIMILAR PLATFORMS

-I

Prior and Posterior Distributions of B Variation In COV of PriorofB fromOl to04 -- -

4500

-4000

-

- -

3500 --middot - bull bull - bull bull Prior-r ICOV =02

3000

- - - - Prior-r ICOV =03 -

bull -~ e 2soo -

bull bull bull bull bull bull bull Pnor-r ICOV =o 4t frJ ~ 2000 Posterior rlCOV =0 2

bull bull Postenor rlCOV =0 3 bull bull 1500 bull middot bull - shy -middot- Posterior rlCOV =04

middot~ I 000 - bull

-middotmiddot ~ ~middotmiddot M -~

- middot 0 500

wbullbullbull _ I middot~ middotmiddotmiddotmiddotmiddot middotmiddotmiddotmiddotmiddot - -~ ~ __ middot_ _ bull -shy0 000

0 000 0200 0 400 0 600 0 800 IOOO 1200 1400 1600 1800 2000 -

- - b

-

=shy

Posterior Distribution of Bias Factor (B) bull Effect of Variation In COV of Prior of B

bull

bullPrior and Posterior Distributions of B Evaluating Effect of Removal of One Platform from Each Category from the Base Case of 13 Platforms

400

~ ~ 350

j IA i

3 00 bull I bull I ---Prior-r I bull l

250 1 I 1

I bull bull ---Posterior r lall 13 casesshy- ~E 2 00 bull bull bull bull bull bull bull Posterior r without STISIK

- ~ 1 bull - bull I - bull bull - bull bull Posterior r Iwithout T2I I 50 I

J-tt~i - bull - bull - Posterior r Iwithout T2S

tOO [-

I I ~ ~ I 0 50

0 00

___ r J i

l~T

~--

0 00 0 20 0 40 0 60 080 I 00 120 140 160 110 2 00

b

zshy

Posterior Distribution of Bias Factor (8) bull Effect of Removal of middotPlatform from Any Category

bullPrior and Posterior Distributions or B Evaluating Effect or Removal or 3 Damage Cases from the Base Case or 13 Platrorms

4

35

3

25

e L 2 ~ bulle shy

I 15

bull I

05

0

0 02 04 06 I 2 I 4 16 IS 2

I I

I

I

I Ibull

J -tmiddot L--

bullI

~

I

fl

II ~

I I v I

middot bull bull

bull L

~ ll- I-~lt

OS I

b

I

bull

1middot bullbullbullbull

~ __

bull

- - - - Prior-r

- - - - Postenor flall failure cases

Postenor flall success cases

--- Posterior flall 13 cases

---Posterior fI success failure cases only

IPosterior or B Mean =t11 COV =0121

-

_I

Posterir Distribution or Bias Factor (B) bull Effect or Removal or 3-Damage Cases

-

bullPrior and Posterior Distributions or B Evaluating Effect or Additional 13 Similar Observations as

Used in Andrew JIP The POsterior or B from Andrew-JIP Considered as New Prior or B

- 500

450

400

350

bull bull

bull bull bull bull J bull

bull bull f ~ I

I ~ Ibull bull

bullbull bull

l bull I -bullbull bull bull

I

I bull bull

bull I I~~bull bull bull

bull ~

I bull bull

~ ~_ ~~~

- - New Prior r bull

Posterior 114 new failure cases ~ 300- bull bull bull bull bull bull bull Posterior f 13 new damage casesl 150 ~

- bull bull - bull bull Posterior 116 new success cases i 2 00

Postenor 1113 new success failure -150 damage cases

I 00

0 50 -

0 00 0 00 0 20 0 40 0 60 0 80 I 00 120 I 40 160 110 2 00

Posterior ol B Mean= 120 COV= 0071 b

shy

shy

_rshy

Posterior Distribution of Bias Factor (B) bull Effect of Additional 13-Similar Observations

1

1 I

WEIGHTING APPROACH bull

bull ASSESSMENT OF PLATFORMS DATA

Select Platforms Most Mappable to the Analyzed Platforms

Unlikely SurvivalDamage Cases Identification

Based on Cut-Off Criteria Agemiddot Hs etc

bull INFORMATION REQUIRED FOR COMPARISON

bull CATEGORIZATION OF ADDITIONALmiddot PLATFORMS

bull UPDATING OF CALIBRATION RESULTS

--

OVERVIEW OF PLATFORMS SURVIVEDDAMAGED -~

Operator -

- Platforms

Survived CS)

Platforms

Damaged CD)

Platforms Used In

Phase IJIP -

JI~ ~Bill~IRBDll -

Amoco -

Chevron -

Exxon

Mob II

Phllllps

Shell

Unocal -

30

143

72 22

7

39 25shy -

1

12

bull

1 bull

1

1(S)+1 (D)

3 (S) +3(D)

2 (S)

bull shy

bull

bull

bull

Total Partlclnants Platforms 338 - 15 10C6S+4D) -

Other Operators

310

13

3 (D)

Total Platforms 648 28 13C6S+7D) -

Summary of Andrew JIP bull Phase I Platforms

Plairshy w v- N-ber Lea-le 1 shy car Hind_ o-w N-bull Deplh fl fl IJpe la bull n Yeadon Jadtet lbelledtf liefalit -- _ -OD

IL H - -shy _ICllshy IL IL - - rQClto ~ _

K STUIK 137 1963 8 365 608S4 Cellardeck

-middot STl-shy 170 1964 4 -~ K 39 62269 Cellardeck - --ST134W 137 1911 4 43 608S4 Smvlnlshy- _ - -

WD90A 184 1964 8 K 267Sbull34 83 SOS31 SUboftllar deck SmvMI

MC311 343 1978 8 x S7 62 - MC397 468 1991 4 x shy S33bull6SO 65911 -shySmeSmvlvol -- gt -

r-- T23ST52 63 1969 4 ton- shyKshy 52594 S03 -T25SSl39 62 1969 4 Kshy SOS94 S063 - middot ton-shy__ -

bullbull a ton--shyST161A 118 1964 8 K - 347S 51495 Cellarclecti

r~

ST1771 142 1965 8 K 39 60167 CeDordeck failme

STISIH 137 1964 8 K 395 608S4 CeDardeck Failme

K ST130A 140 1958 a SS 6095 I lblv Pai1uno - T21ST72 61 1969 4 Kshy S4S94 4974 I lhlv Pai1uno -

- -

JE

INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON

bull WATER DEPTH

bull bull YEAR OF INSTALLATIONDESIGN

bull PLATFORM ORIENTATION

bull PLATFORM LATITUDE AND LONGITUDE

bull KEY SKETCHES OF PLATFORM STRUCTURAL FRAMING

Number of Legs Plies Leg Batter Number of Conductors bull Deck Elevations (Including sub-cellar deck If any) Jacket Framing Elevations Brace type In Vertical Frames Sizes of Primary Members (leg pile vertical frames braces primary horizontals) Joint Details (can gap etc) LegPiie Annulus Grouting Status

bull middotPOST ANDREW INSPECTION REPORT Significant Wave Height Hs (ft)

bull INFORMATION ON DAMAGES PRIOR TO ANDREW

bull RESULTS FROM IN-HOUSE STUDIES IF AVAILABLE

bull ~RIGINAL DESIGN CRITERIA INFORMATION IF AVAILABLE

bull SOIL SHEAR STRENGTH PROFILE IF AVAILABLE

Platforms Identified for Evaluating Therr Likelihood of Falling Under Unexpected Survival Category

OPlRATOR

middotADA

BLOClt

llmucnJllE

NAME

YEAll

INSTAUJD

WA1ER

DBPTll

DISTANCE

JlllOM SHOllE

(MILU)

Bl

(M)

AMOCO 111113

STl11

WDI075

A

bull D

St

17

120

1n

44

20

10 I

CllllVllQfl STlllO c 61 110 21 11 STIUO D Q 1110 21 11

STl134 p Q 137 29 11

ST0151 G 61 137 11

ST0151 I 3 121 11

ST0151 J Q 140 11

STl151 It 137 11

ST0151 PJl001 Q UT 11

ST0151 PROD-2 Q UT 11

ST017 A 3 140 35 11

WDIUT c 65 214 34 10

llXXON STI055 II 51 6T 14 I

STl15 A t3 34

WDI073

WD0073

Amiddotshy Q

1111 1111

22

22

I

I

WDI073 c 1n 22 I

- WDOCl93-shy II

65 1110 2T A 3 194 31 MOBU 111121

SS IOl7

A B

501

50

50

31

T

I

T 6

MtlllPllY SS114 B T 50 T T

PlllllIPS SS0149 A 5 55 33 T

SAMllDAN STln A 107 T T

SBllLL WDl103 A 65 223 2T WDl103 B 65 221 2T WD0104 c 65 221 2T WD0104 D 6T 20 2T

UNOCAL SS020I

SS 020I SSl20I

SS 0209

SS0209

II p

PURll

A

B

3

n 61

103

97

97

95

100

34

34

34

34 34

bullbull bullbullbull

- SS0209 D u 95 34 I

- SS 1209 SSl253

SS0253

G A

c

6T

2

100

165

1T5

34

54

54

bullbull I

I I

I I

SOUTH MARCH

~LAND

N

SOUTH PaTOI ENE ISWltD AREA

MMS Defined Limits ror Offshore Structures Requiring Inspections Following Andrew

t---Jbulln3 T25bull I bullTtl

bull1$1Htl1Abull bull1SIK1779bull

~

middot-DaTA AREA~T

CRANO rsLE

AREA bull1300

bullll4W

=shy

bullPath of Hurricane Andrew with Platforms Used In Calibration

Likelihood Function vs b All Success Case STISI K STIJOQ ST134W WD90A MCJll MCJ97

I 0000

0 9000

0 8000

07000

-Bo6000

tiiii 0 5000 -

04000

0 3000

0 2000

01000

0 0000

0 05 25

ri1rt middot-middot-middot-middot-middot-

I 5 2

b

---11 middot151K(blsu~ccsI

- bull - bull - fk 130Q(bfU~lCSS)

lie bull134W(blsuccc]

lk-WD90A[blsultCS]

- - - lk middotMC397(bfUtlCS]

--o-- lk-MCJll(bbuctcsJ

--- ll (bl 6 out of 6 Mlttcs case]

_

( Likelihood Functions middotSuccess Cases - p - 1 c I

CATEGORIZATION OF ADDITIONAL PLATFORMS

I I I

bull CATEGORIES

Unexpected Survivals

Expected Survivals

bull APPROACH

Identify Corresponding Platform(s) Evaluated In the JIP

Determine Max Wave Height at Platform Location

Evaluate Whether Wave Hit the Deck

middot - Evaluate Likelihood of R and S to Differ from the Base Cases

Review Joint Information

Assess Special Features

bull IDENTIFY NUMBER OF PLATFORMS IN EACH CATEGORY

I

UPDATING OF CALIBRATION RESULTS

bull INCLUDE EFFECT OF ADDITIONAL PLATFORMS ON bullBbull

Consider Increased 1Uncertalntles In R and S

Develop bullNeW Ukellhood Functions for Different Categories

middotLet bullxbull Number of Ukely Unexpected Survival Cases and y Number of Ukely Expected Survival Cases

Determine Combined Likelihood Function lk (b Iadditional platforms) middot

bull REVISED LIKELIHOOD FUNCTION

lk(b I all cases) = lk(b I 9 platforms) middot x lk(b Ix platforms) x lk(b I y platforms)

bull ESTABLISH REVISED BIAS FACTOR B

f8 (b) u f8 (b) lk (b I all cases)

II

t t ~

shy

APPLICATIONS OF BIAS FACTORS

ii APPLICABLE TO PARTICULAR LOAD AND RESISTANCE RECIPE

bull USE BY THE API COMMITTEES AND MMS

Criteria Revisions

Recommended Design Guidelines Revisions bull

bullbull REASSESSMENT STUDIES AND DECISION MAKING

middot

bull RISK ASSESSMENT STUDIES BY THE OPERATORS

Individual Platforms

Fleet of Platforms

middot IMR Program Implementation

bullbull NOT TO BE USED IN DESIGN

I I fl J

EXAMPLE USE OF BIAS FACTORS

bull PHASE I EXAMPLE APPLICATION

bull EVALUATION OF ANNUAL PROBABILITY OF FAILURE

Establish Annual Maximum Seastate Data

Determine Coefficients to Define Base Shear

Determine Ultimate Capacity of Platform

Determine Probablllty of Failure Versus bullbbull

-bull DETERMINE UPDATED PROBABILITY OF FAILURE

Pf - P [S gt R]

Pf I P [~ gt Rb I b] f 8 (b) db

Pr - J P [S gt Rb I b] f bull 8 (b) db

bull TOTAL PROBABILITIES OF FAILURE

Total Pf - 1 bull Il [1 middotPf] - l Pf bull

Total Pf - 1 bull Il [1middot Pf] - l Pfmiddot

Total Pr = 1 - n [1middot PfbullJ = l Pfshy

I I I

EXAMPLE APPLICATION bull PHASE I

PLATFORM ST 151 K

bull GULF OF MEXICO

bull WATER DEPTH = 137 FT

bull ANNUAL MAXIMUM HS MEDIAN 175 ft COV 031

bull ASSUMED STORM DURATION 3 hoursmiddot

bull ASSUMED NUMBER OF WAVES 800

bull MEDIAN HS FOR STORM HOURS 1 2 AND 3

bull MEDIAN CURRENT VALUE bull

bull DETERMINE ULTIMATE CAPACITY FOR THREE DIRECTIONS

bull DETERMINE ANNUAL FAILURE PROBABILITIES

Broadsidemiddot Diagonal End-on

0037 0025 0007

0071 0044 0018

0022 0016 0004

DETERMINE TOTAL ANNUAL PROBABILITY OF FAILURE

Pf - 007

Pf - 013

Pr - 004

bull

EXAMPLE USE OF BIAS FACTORS - PHASE II middot

bull DETERMINE PREDICTIVE PROBABILITY OF FAILURE

Case - 1

Pf _ S S Pf (b b2) fa1a2 (b b2) db db2

g bull -1 (b1 R I S) - 1

92 - (b2 R2 I S) - 1

Case -2

Pf s ~ [bJ f11 (b) db

Pf [bJ p [g lt 0)

g - bull 1 (b1 R1 I S) - 1

1

bull1) l

I

BENEFITS OF ANDREW JIP bull PHASE II

lt

bull IMPROVED LOAD AND RESISTANCE RECIPE

bull IMPROVED CALIBRATION PROCEDURE

bull MULTIPLE BIAS FACTORS

bull PROCEDURE TO INCLUDE EFFECT OF OTHER PLATFORMS

bull PROCEDURE TO USE MULTIPLE BIAS FACTORS

  • Structure Bookmarks
    • ANDREW PHASE II JOINT1 INDUSTRY PROJECT
    • PROGRESS MEETING NUMBER 1
    • ANDREW PHASE II JOINT INDUSTRY PROJECT PROGRESS MEETING NUMBER 1 NOVEMBER 16 AGENDA
    • ANDREW PHASE II UST OF PARTICIPANTS
    • ANDREW PHASE I OBJECTIVES AND RESULTS
    • ANDREW PHASE II PROJECT OBJECTIVES
    • SUBSEQUENT PROJECTS
    • ANDREW PHASE II PROJECT TEAM
    • ANDREW PHASE II UPDATE MODELING middotAND ANALYSIS RECIPE
    • -ANDREW PHASE II CONDUCTOR RESISTANCE
    • ANDREW PHASE Il WA VE FORCE PROFILE
    • ANDREW PHASE II WA VE FORCE PROFILEmiddot
    • ANDREW PHASE II BRACE MODELING
    • ANDREW PHASE II FOUNDATIONS
    • ANDREW PHASE II middot middot JOINT MODELING
    • ANDREW PHASE Il JOINT MODELING OPTIONS DISCUSSION
    • ANDREW PHASE II JOINT MODEJJNG OPTIONS ILLUSTRATIONS
    • ANDREW PHASE II JOINT MODELING PHASE II METHOD
    • ANDREW PHASE II ANDREW STORM CONDmONS UPDATED
    • Platform Selection bullbullbull1111111111111
    • BasisCriteria for Selection bullbullbull1111111111111
    • CALIBRATION
    • WHERE TO INTRODUCE BIAS FACTOR (B
    • PHASE I -KEY CONCLUSIONS
    • bull CALIBRATION ISSUES PHASE II
    • PHASE II WORK SCOPE (CALIBRATION)
    • SOFTWARE
    • SENSITIVITY STUDIES POSTERIOR DISTRIBUTION OF B
    • WEIGHTING APPROACH bull
    • INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON
    • CATEGORIZATION OF ADDITIONAL PLATFORMS
    • UPDATING OF CALIBRATION RESULTS
    • APPLICATIONS OF BIAS FACTORS
    • EXAMPLE USE OF BIAS FACTORS
    • EXAMPLE APPLICATION bull PHASE I PLATFORM ST 151 K
    • EXAMPLE USE OF BIAS FACTORS -PHASE II middot
    • BENEFITS OF ANDREW JIP bull PHASE II
Page 40:  · . ' ANDREW PHASE II PROJECT OBJECTIVES . ' • ·UPDATE MODELING AND ANALYSIS RECIPE· JOINTS ' ' ·BRACES WAVE PROFILE FOUNDATIONS . ' ' CONDUCTORS· • ANALVZE PLATFORMS USING

bull lt c lt I lt

2-2

l---4-~~~+-~~-+-~~~1--~~-+-~~--l--~T--+-~~--i24

f-tJVtMfiz~ 1Cfl (r STvb(

Plollol bullbull 5middotNOYmiddot9214bull1J Irbullbull Illbull [ANDREVl~NDREVCUSTbullMAXll 5-NOY-1992 10bull54 - 9208

HEIGHT lml

I~ MAXI

I L 1~middotamp SZ~- ~A I 13c

I J) ( r I X~0iWJ~0gtJ gt L gt gt I 1c I 120

-- N

~ deg

I I I I I I e C I ~=-41 122 -94 -92 -90 -88 -86 -84 -82 -80

-

Figure 7 contours of maximum hindcast significant wave height

_

~

~tl ~

NOltNttIJ 17lt ltNS S iJJ)y

Maximum Significant Wave Height - Hurricane Andrew 1992

30 ~

26 ~ I ~-~~~- == -shy -~ 35-_

24deg1--shy --------~- middot-middot-------middotmiddotmiddot----middot------middot-middot~middot- - - ~-----1---middot

22middot ~~~~~~~-~ -94 -92 -90 -88 -84deg-86 -82 -somiddot

c

-

Platform Selection bullbullbull1111111111111

0 BasisCriteria for Selection middot

0 Review of Phase I Platforms middot 0 Alternative Platforms

0 PMB Recommendations

0 Participants Comments

gt

L

l I

BasisCriteria for Selection bullbullbull1111111111111

0 Benefit for Calibration - Classification (UnexpectedExpected)

- Failure Mode (StructuresFoundations)

- Damaged StructuresKnown response

- Better Representation of Full ~opulation

0 Availability of Data - Existing Models

- Inspection Data

- Damage Data

PREDICTED BEHAVIOR

-

OBSERVED

BEHAVIOR

-SURVIVED DAMAGED FAILED

SURVIVED

ST134W WD90A

-MC311 -

MC397 -

--

-

-

ST151K ST130Q

-

-

-

DAMAGED - -

-

-

--

-

-

ST52 SS139 ST177B -

ST161A

-_

-

FAILED

-

-

-

ST151H ST130A ST72

Operator - Platfonns - Platfonns Survived Damagedamp

- Falled -

bull

-

-

Jle eBDIGIPllDbl -Amocoshy 30 1 Chevron -

143 12 Exxon 72 Mobll 22 1 Phllllps 7 bull

-Shell 39 Unocal 25 1

Total Partlclnts Platfonns 338 15

Other Operators bull- 310 13

-

Total Platfonns - -

648 28

fbml

Platforms Considered 6 7

tie of Total Platfonns 1 25

OVERVIEW OF PLATFORMS SURVIVEDDAMAGEDFAILED

-

shy

v

Platforms Evaluated in Phase I

-

-Platfonn Water Year Number Analysis Avallabillty of Availability Platform

- Name Depth rl of Model DetaDed rl ClassUlcatlon

Installation legs Inspection Soll Report Based on Information Phase I Calibration - -ft

- -- Suaill flalfnrm ~BSi -STISIK 137 1963 8 CAP Yes Yes Un eel Sarvlval

-STIJOO 170 1964 4 CAPmiddotshy Yes ST134data u ST134W 137 1981 4 CAP Yes Yesmiddot E

-

edSarvlval -

middotsurvival

WD90A 184 - 1964 8 CAP Yes E -

- Sarvlvill

MC3ll 343 1978 8 KARMA

MC397 468 1991 4 KARMA -

bull ~reglmmiddot- -

Yes Sure Survival

bull Yes - Sare Sanlval-

-T23STS2 63 1969 4 CAP

T2SSS139 62 1969 4 CAP -

Yes

Yes

Yes Ex

Yes Une

bull SurvivalLikely to DamalI

middot bull SanlvalLlkely to DamalI

ST161A 118 1964 8 USFOS

Eail11a pound1al(bullllm Clss ST177B 142 --1965 8 CAP

STISIH 137 1964 - 8 CAP C

STIJOA 140 19S8 8 CAP - -

Yes

Yes

ollamed

Yes

- Yes E

-

at about 1 mUe

Yes

ST134data

bull SafvlvalLlkelv to Dam- bull- -

E middotrahre

Emeried Failure -

Likely FaUare

T21ST72 61 1969 4 CAP Collamed No Likely FaUure

Platforms Identified for Evaluating Their Likelihood of Falling Under UnexpeCted Survival Category

CllEllATOa

AllB

llLOCll

Sl1lvcnJllE

NAME

YEAR

INSTAUJCD

WATER

IJBPJB

middotmiddotDISTANCE

ftOMSllOllE

(MIUIS)

Ill

(M)

AMOCO BllW

STllI

WDI075

A

bull D

4

n uo 17J

44

32

bull 20

10

bull CllEVllOH STtllO c I 1bull 21 11

STl130 D 0 161 21 11 STllJ4 I 0 137 D - 11

STIUI G I 137 32 11

STIUI I 0 121 32 11

STIUI

J 0 140 32 11

STIUI It 4 137 32 11

STllSI PROl)I 0 137 32 11

STllSI PROD-2 0 137 32 11

STll7 A 0

140 35 11

WDl117 c 65 214 34 10

DXON STshy E 51 14 bull STl165 A t3 bull 34 WD0073

WD0073

WD0073

A _ c

4

62

4

IA IA

17J

22

22

22

bullbullbull WD003-shy E 65 161 Z7

A 0 IN 31 MOBD Bl12I

ssn A bull

50

50

31

7

bull 7

MIJllPllY SS114 B 7 50 7 7

PlllUIPS SSl14t A 55 33 7

SAMEDAN ST17J A 4 117 7 bull 7

SJIELL WD0103 A 65 223 27 WDllOJ bull 65 221 27 WDll04 c 65 221 27 WDOI04 D 67 2 27

UNOCAL

SSl20I

SSl20I

bull SSl20I

SS0209

SS0209

SS0209

SS0209

SS 0253

SS 0253

E

I

PtJllE

A

bull D

G A

c

0

4

4

7J

61

0

0

69

103

97

97

9S

100

95

100

165

175

34

34

34

34

34

34

34

54

54

bullbullbullbullbullbullbullbullbull

Platforms Recommended for Phase II

llalr y_ w- Aot Av- _ I 1 Doplla Model nm bull bull _ bull I I H11 s18-1 Irht lbwl~ bullbullbull(-illtdlel OM lllMtD

II ---

-STISIIC 137 1963 I CAP Yeo Yeo I Yeo

STtshy -170 1964 4 CAP Yeo ST134dlU I Yeo

STl34W 137 Yeo Yeo -1981 4 CAP D Yeo

WD90A 114 1964 I CAP Yeo - D

MC311 30 1971 I KARMA Yeo -sre-

MC397 middot 461 1991 4 ICARMA Yeo Sireshy - - r- -

__ T13ST5l 63 1969 4 CAP Yeo Yeobull middot- ~n- D Yeo _ Tl5SSl39 62 1969 4 CAP Yeo Yeo I Yeo

_ STIT711 142 1965 I CAP Yeo I Yeo middot-1shySTl61A Ill 1964 I USFOS Yeo Yeo - D bulloshy

- r-

STISIH 137 1964 I CAP y D y -_ STl31A 140 1951 I CAP Yeo STl34dlU I Yeo

T21Sll 61 1969 4 CAP No rar-middot I Yeo

~-rmiddot-middot SS209A 95 1967 16 CAP Yeo D -SSU4R

11 UI A 41 -in D SS 2741AIM CAP bull

_ _ -shy - -shy- -shy

CALIBRATION

bull PHASEIAPPROACH

bull CALIBRATION ISSUES

bull PHASE II WORK SCOPE

bull SOFlWARE AVAILABILITY

bull EFFECT OF ADDITIONAL PLATFORMS

bull WEIGHTING APPROACH

bull EXAMPLE APPLICATION middot

(l=Ofce Mode] Wind

~

fcurrent l -middot

~SW ~leldl4

ultimate resistance force I first

design load Illgt

i _

deflection

-

bull

4

Historical Hurrlcane Dariiage and Survival of Platforms

Components Required for Calibration (Petrauskas 1992)

PRIOR Distribution

PLATFORM EXPERIENCE

BAYESIAN Survivals and Failures

UPDATING

Waves Winds and Currents bull

POSTERIOR Distribution

- A_middotJA ~ Survival Data Failure Data

Result Result

Load Resistance

middotfl

-

Bayesian Updaing middotGeneral Description (Petrauskas 1992)

L Imiddot

WHERE TO INTRODUCE BIAS FACTOR (B

On Resistance (Rmue - B (RPREDICreD I

middotOn Loads (Smue middot - B (SPRE01cTEo

On Safety Margin (RSmue - B (RSPREDICTED

CALIBRATION METHODOLOGY

gt

r

FIELD OBSERVATIONS

Observed Behavior

survivals Damages

Failures

r

bull

CAPACITY ANALYSIS

Predicted Load and Resistance Predicted Failure Modes

Survival Damage Failure cases

RELIABILITY ANALYSIS

- Probability of Failure ~

Ukellhood Failure

BAYESIAN UPDATING

Prior Distnbution of B Combined Likelihood Function

Posterior Dlstnbution of B

BIAS FACTOR

Mean Value cov bull

Represents Modeling Uncertainties

Prior Distribution of B Likelihood Functionmiddot Observing Success case STISIK and Posterior Distribution of B

2 00

uo

160

140

120

e 100 L $ uoIshy

0 60 ~ ~ 040

020

0 00

0 00 0 50

_ I 00

b

I 50 2 00

-shy lk [bl STISIK)

--shy Pnor-f(b)

--shy Pos1enor r(bl ISIK)

Prior or B Mean bull 10 COV bull 030

Posterlor of Bbull Meanbull 136COV 0 16

2 50

Likelihood Function and Posterior Distribution Example Platform STISIK

- -

~

- Prior and Posterior Distributions or B Failure cases (ST177B STlSlH ST130A T21) Damage cases (T23 T25 ST161A) Success cases (STlSlK ST130Q ST134W WD90A MC311 MC397)

400

350

300

2 so 0-

I

bull bull

bull bull gt

- bullbull ~ bull bull I bull

_ bull bull bullbull I bull bull bullII V -

bull bull

I bull ~ bull

bullbull bullI bull bullI bull bull bull

middot -~

bull__ ~ middotI middotmiddot

L 2 00

e ~ I SO

-I 00

- 0 so

000 gt

0 00 020 0 40 060 0 80 I 00 120 140 160 110 200

b -

-

shy

Prior-r

- - - - Posterior fLlll failure cases

bull bull bull bull bull bull bull Posterior flall damage cases

- bull bull - bull middot Postenor flall success cases

Posterior flall success fulure damage cases

-

Posterior of B Mean= 119 COV 0101

Posterior Distribution or Blas Factor (8) bull Different Categories and All Cases Combined

shy

l

PHASE I - KEY CONCLUSIONS

bull GLOBAL CORRECTION FACTOR

bull 15 TO 20 CONSERVATISM MEAN)

bull UNEXPECTED SURVIVALS HAVE GREATEST IMPACT

bull DAMAGED CASES MORE INFORMATIVE

bull CALIBRATION ISSUES PHASE II

bull

EFFECT OF IMPROVED CAPACITY ANALYSIS RECIPE

bull APPLICABILITY TO MULTIPLE FAILURE MODES

bull NO FOUNDATION DAMAGE CASE INCLUDED

bull IMPROVED REPRESENTATION OF FULL ANDREW DISTRIBUTION

-bull NUMBER OF PLATFORMS CONSIDERED

I 1 I

I t ) bull ~

PHASE II WORK SCOPE (CALIBRATION)

bull

bull ESTABLISH MULTIPLE BIAS FACTORS

Use Nine Base Case Platforms

middot - Capacity Analysis for Revised recipe

Foundation Blas Factor from Foundation JIP

bull DEVELOP WEIGHTING APPROACH

bull UPDATE MULTIPLE BIAS FACTORS

bullEffect of Platforms Not In the Nine Base Cases

bull EXAMPLE APPLICATION OF BIAS FACTORS I

bullbull

PHASE I PHASE II

CAPACITYbull ANALYSISRu bull

---------------------1----middot

Jc1 - iJbullraquoJ dbull

FttlS frt[f~l ratdhJJ

-

LIKELIHOOD FUNCTION

DEVELOPMENT

lk Jb Jfall1N) bull Jb)

111 Ill Isurvtvoq bull 1 - bl

-middot -------------------shy ~---------------------middot-middot

r111ir8 Cb1 111 Cbl BAYESIAN

UPDATING

~cove

R1R2 bull middot---------------------~----

Jblbull P(g cOJ

1 bull JbRI (MBSI I bull 1

~middot

llcJllJfalluro) bull P(gcOJ Ill 111 b21fallun) bull P ((g1 CO) U 1112 c OD

middot--------------------shy ----------------------shy bull

r111111r8 Cbl 111 Ill i-1nro1

deg ~1b2)laquottfrr (b1b21-lnfo) re Ill bull 1-11middot b2) bull lodegamp

middot

i1 covamp i2 cov~i

-

shy

--

Likelihood Function vs b All Success Case STISI K STl30Q STl34W WD90A MC311 MC397

I 0000

0 9000

OIOOO

0 7000

~ 0amp000

ta osoOO

~ 0 4000

0 3000

0 2000

0 1000

bull 0 0000

bull

I

middot

middot bull I

Ibull

-sibull ~ -middot-middot-middot-middot -middot-

----7----+-----~ I

Ibull bull bull I

rbull

--- Jk ISIK[blsuccess)

- bull - bull - lk 130Q[blsuccess) bull

lk 134W[blsuccess)

Jk-WD90A(blsuccc~I

- - - lk MC397[blsucccss)

--ltgt-- lk middotMC31 l[blsucccss)

--- Jk [bl 6 out of6succcsHascs)

05 I 5 2 250

b

Likelihood Functionsmiddot Success Cases - Phbullsc L

shy

bullbull bull bull bull bull bull -------------------~

bull bull bullbull bullbull bullbull bullbull bullbull bullbull bull

bull bullbull

bullbullbullbull

bullbull

bullbullbullbullbullbullbull

bullbullbullbullbullbull

bullbullbullbullbullbullbull

z 0

t z ) LL

c 0 0c i ~ i

J

~ sect ~--------------~ e l5

I

I

SOFTWARE

bull PF PROGRAM

bull CALREL

bull PF +CALREL

bull RELACS

bull

gt i

l

SENSITIVITY STUDIES

POSTERIOR DISTRIBUTION OF B

bull EFFECT OF CHANGE IN DISTRIBUTION OF PRIOR OF B

bull EFFECT OF REMOVAL OF ONE PLATFORM FROM EACH CATEGORY

bull EFFECT OF REMOVAL OF THREE DAMAGE CASES

bull EFFECT OF ADDITIONAL 13 SIMILAR PLATFORMS

-I

Prior and Posterior Distributions of B Variation In COV of PriorofB fromOl to04 -- -

4500

-4000

-

- -

3500 --middot - bull bull - bull bull Prior-r ICOV =02

3000

- - - - Prior-r ICOV =03 -

bull -~ e 2soo -

bull bull bull bull bull bull bull Pnor-r ICOV =o 4t frJ ~ 2000 Posterior rlCOV =0 2

bull bull Postenor rlCOV =0 3 bull bull 1500 bull middot bull - shy -middot- Posterior rlCOV =04

middot~ I 000 - bull

-middotmiddot ~ ~middotmiddot M -~

- middot 0 500

wbullbullbull _ I middot~ middotmiddotmiddotmiddotmiddot middotmiddotmiddotmiddotmiddot - -~ ~ __ middot_ _ bull -shy0 000

0 000 0200 0 400 0 600 0 800 IOOO 1200 1400 1600 1800 2000 -

- - b

-

=shy

Posterior Distribution of Bias Factor (B) bull Effect of Variation In COV of Prior of B

bull

bullPrior and Posterior Distributions of B Evaluating Effect of Removal of One Platform from Each Category from the Base Case of 13 Platforms

400

~ ~ 350

j IA i

3 00 bull I bull I ---Prior-r I bull l

250 1 I 1

I bull bull ---Posterior r lall 13 casesshy- ~E 2 00 bull bull bull bull bull bull bull Posterior r without STISIK

- ~ 1 bull - bull I - bull bull - bull bull Posterior r Iwithout T2I I 50 I

J-tt~i - bull - bull - Posterior r Iwithout T2S

tOO [-

I I ~ ~ I 0 50

0 00

___ r J i

l~T

~--

0 00 0 20 0 40 0 60 080 I 00 120 140 160 110 2 00

b

zshy

Posterior Distribution of Bias Factor (8) bull Effect of Removal of middotPlatform from Any Category

bullPrior and Posterior Distributions or B Evaluating Effect or Removal or 3 Damage Cases from the Base Case or 13 Platrorms

4

35

3

25

e L 2 ~ bulle shy

I 15

bull I

05

0

0 02 04 06 I 2 I 4 16 IS 2

I I

I

I

I Ibull

J -tmiddot L--

bullI

~

I

fl

II ~

I I v I

middot bull bull

bull L

~ ll- I-~lt

OS I

b

I

bull

1middot bullbullbullbull

~ __

bull

- - - - Prior-r

- - - - Postenor flall failure cases

Postenor flall success cases

--- Posterior flall 13 cases

---Posterior fI success failure cases only

IPosterior or B Mean =t11 COV =0121

-

_I

Posterir Distribution or Bias Factor (B) bull Effect or Removal or 3-Damage Cases

-

bullPrior and Posterior Distributions or B Evaluating Effect or Additional 13 Similar Observations as

Used in Andrew JIP The POsterior or B from Andrew-JIP Considered as New Prior or B

- 500

450

400

350

bull bull

bull bull bull bull J bull

bull bull f ~ I

I ~ Ibull bull

bullbull bull

l bull I -bullbull bull bull

I

I bull bull

bull I I~~bull bull bull

bull ~

I bull bull

~ ~_ ~~~

- - New Prior r bull

Posterior 114 new failure cases ~ 300- bull bull bull bull bull bull bull Posterior f 13 new damage casesl 150 ~

- bull bull - bull bull Posterior 116 new success cases i 2 00

Postenor 1113 new success failure -150 damage cases

I 00

0 50 -

0 00 0 00 0 20 0 40 0 60 0 80 I 00 120 I 40 160 110 2 00

Posterior ol B Mean= 120 COV= 0071 b

shy

shy

_rshy

Posterior Distribution of Bias Factor (B) bull Effect of Additional 13-Similar Observations

1

1 I

WEIGHTING APPROACH bull

bull ASSESSMENT OF PLATFORMS DATA

Select Platforms Most Mappable to the Analyzed Platforms

Unlikely SurvivalDamage Cases Identification

Based on Cut-Off Criteria Agemiddot Hs etc

bull INFORMATION REQUIRED FOR COMPARISON

bull CATEGORIZATION OF ADDITIONALmiddot PLATFORMS

bull UPDATING OF CALIBRATION RESULTS

--

OVERVIEW OF PLATFORMS SURVIVEDDAMAGED -~

Operator -

- Platforms

Survived CS)

Platforms

Damaged CD)

Platforms Used In

Phase IJIP -

JI~ ~Bill~IRBDll -

Amoco -

Chevron -

Exxon

Mob II

Phllllps

Shell

Unocal -

30

143

72 22

7

39 25shy -

1

12

bull

1 bull

1

1(S)+1 (D)

3 (S) +3(D)

2 (S)

bull shy

bull

bull

bull

Total Partlclnants Platforms 338 - 15 10C6S+4D) -

Other Operators

310

13

3 (D)

Total Platforms 648 28 13C6S+7D) -

Summary of Andrew JIP bull Phase I Platforms

Plairshy w v- N-ber Lea-le 1 shy car Hind_ o-w N-bull Deplh fl fl IJpe la bull n Yeadon Jadtet lbelledtf liefalit -- _ -OD

IL H - -shy _ICllshy IL IL - - rQClto ~ _

K STUIK 137 1963 8 365 608S4 Cellardeck

-middot STl-shy 170 1964 4 -~ K 39 62269 Cellardeck - --ST134W 137 1911 4 43 608S4 Smvlnlshy- _ - -

WD90A 184 1964 8 K 267Sbull34 83 SOS31 SUboftllar deck SmvMI

MC311 343 1978 8 x S7 62 - MC397 468 1991 4 x shy S33bull6SO 65911 -shySmeSmvlvol -- gt -

r-- T23ST52 63 1969 4 ton- shyKshy 52594 S03 -T25SSl39 62 1969 4 Kshy SOS94 S063 - middot ton-shy__ -

bullbull a ton--shyST161A 118 1964 8 K - 347S 51495 Cellarclecti

r~

ST1771 142 1965 8 K 39 60167 CeDordeck failme

STISIH 137 1964 8 K 395 608S4 CeDardeck Failme

K ST130A 140 1958 a SS 6095 I lblv Pai1uno - T21ST72 61 1969 4 Kshy S4S94 4974 I lhlv Pai1uno -

- -

JE

INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON

bull WATER DEPTH

bull bull YEAR OF INSTALLATIONDESIGN

bull PLATFORM ORIENTATION

bull PLATFORM LATITUDE AND LONGITUDE

bull KEY SKETCHES OF PLATFORM STRUCTURAL FRAMING

Number of Legs Plies Leg Batter Number of Conductors bull Deck Elevations (Including sub-cellar deck If any) Jacket Framing Elevations Brace type In Vertical Frames Sizes of Primary Members (leg pile vertical frames braces primary horizontals) Joint Details (can gap etc) LegPiie Annulus Grouting Status

bull middotPOST ANDREW INSPECTION REPORT Significant Wave Height Hs (ft)

bull INFORMATION ON DAMAGES PRIOR TO ANDREW

bull RESULTS FROM IN-HOUSE STUDIES IF AVAILABLE

bull ~RIGINAL DESIGN CRITERIA INFORMATION IF AVAILABLE

bull SOIL SHEAR STRENGTH PROFILE IF AVAILABLE

Platforms Identified for Evaluating Therr Likelihood of Falling Under Unexpected Survival Category

OPlRATOR

middotADA

BLOClt

llmucnJllE

NAME

YEAll

INSTAUJD

WA1ER

DBPTll

DISTANCE

JlllOM SHOllE

(MILU)

Bl

(M)

AMOCO 111113

STl11

WDI075

A

bull D

St

17

120

1n

44

20

10 I

CllllVllQfl STlllO c 61 110 21 11 STIUO D Q 1110 21 11

STl134 p Q 137 29 11

ST0151 G 61 137 11

ST0151 I 3 121 11

ST0151 J Q 140 11

STl151 It 137 11

ST0151 PJl001 Q UT 11

ST0151 PROD-2 Q UT 11

ST017 A 3 140 35 11

WDIUT c 65 214 34 10

llXXON STI055 II 51 6T 14 I

STl15 A t3 34

WDI073

WD0073

Amiddotshy Q

1111 1111

22

22

I

I

WDI073 c 1n 22 I

- WDOCl93-shy II

65 1110 2T A 3 194 31 MOBU 111121

SS IOl7

A B

501

50

50

31

T

I

T 6

MtlllPllY SS114 B T 50 T T

PlllllIPS SS0149 A 5 55 33 T

SAMllDAN STln A 107 T T

SBllLL WDl103 A 65 223 2T WDl103 B 65 221 2T WD0104 c 65 221 2T WD0104 D 6T 20 2T

UNOCAL SS020I

SS 020I SSl20I

SS 0209

SS0209

II p

PURll

A

B

3

n 61

103

97

97

95

100

34

34

34

34 34

bullbull bullbullbull

- SS0209 D u 95 34 I

- SS 1209 SSl253

SS0253

G A

c

6T

2

100

165

1T5

34

54

54

bullbull I

I I

I I

SOUTH MARCH

~LAND

N

SOUTH PaTOI ENE ISWltD AREA

MMS Defined Limits ror Offshore Structures Requiring Inspections Following Andrew

t---Jbulln3 T25bull I bullTtl

bull1$1Htl1Abull bull1SIK1779bull

~

middot-DaTA AREA~T

CRANO rsLE

AREA bull1300

bullll4W

=shy

bullPath of Hurricane Andrew with Platforms Used In Calibration

Likelihood Function vs b All Success Case STISI K STIJOQ ST134W WD90A MCJll MCJ97

I 0000

0 9000

0 8000

07000

-Bo6000

tiiii 0 5000 -

04000

0 3000

0 2000

01000

0 0000

0 05 25

ri1rt middot-middot-middot-middot-middot-

I 5 2

b

---11 middot151K(blsu~ccsI

- bull - bull - fk 130Q(bfU~lCSS)

lie bull134W(blsuccc]

lk-WD90A[blsultCS]

- - - lk middotMC397(bfUtlCS]

--o-- lk-MCJll(bbuctcsJ

--- ll (bl 6 out of 6 Mlttcs case]

_

( Likelihood Functions middotSuccess Cases - p - 1 c I

CATEGORIZATION OF ADDITIONAL PLATFORMS

I I I

bull CATEGORIES

Unexpected Survivals

Expected Survivals

bull APPROACH

Identify Corresponding Platform(s) Evaluated In the JIP

Determine Max Wave Height at Platform Location

Evaluate Whether Wave Hit the Deck

middot - Evaluate Likelihood of R and S to Differ from the Base Cases

Review Joint Information

Assess Special Features

bull IDENTIFY NUMBER OF PLATFORMS IN EACH CATEGORY

I

UPDATING OF CALIBRATION RESULTS

bull INCLUDE EFFECT OF ADDITIONAL PLATFORMS ON bullBbull

Consider Increased 1Uncertalntles In R and S

Develop bullNeW Ukellhood Functions for Different Categories

middotLet bullxbull Number of Ukely Unexpected Survival Cases and y Number of Ukely Expected Survival Cases

Determine Combined Likelihood Function lk (b Iadditional platforms) middot

bull REVISED LIKELIHOOD FUNCTION

lk(b I all cases) = lk(b I 9 platforms) middot x lk(b Ix platforms) x lk(b I y platforms)

bull ESTABLISH REVISED BIAS FACTOR B

f8 (b) u f8 (b) lk (b I all cases)

II

t t ~

shy

APPLICATIONS OF BIAS FACTORS

ii APPLICABLE TO PARTICULAR LOAD AND RESISTANCE RECIPE

bull USE BY THE API COMMITTEES AND MMS

Criteria Revisions

Recommended Design Guidelines Revisions bull

bullbull REASSESSMENT STUDIES AND DECISION MAKING

middot

bull RISK ASSESSMENT STUDIES BY THE OPERATORS

Individual Platforms

Fleet of Platforms

middot IMR Program Implementation

bullbull NOT TO BE USED IN DESIGN

I I fl J

EXAMPLE USE OF BIAS FACTORS

bull PHASE I EXAMPLE APPLICATION

bull EVALUATION OF ANNUAL PROBABILITY OF FAILURE

Establish Annual Maximum Seastate Data

Determine Coefficients to Define Base Shear

Determine Ultimate Capacity of Platform

Determine Probablllty of Failure Versus bullbbull

-bull DETERMINE UPDATED PROBABILITY OF FAILURE

Pf - P [S gt R]

Pf I P [~ gt Rb I b] f 8 (b) db

Pr - J P [S gt Rb I b] f bull 8 (b) db

bull TOTAL PROBABILITIES OF FAILURE

Total Pf - 1 bull Il [1 middotPf] - l Pf bull

Total Pf - 1 bull Il [1middot Pf] - l Pfmiddot

Total Pr = 1 - n [1middot PfbullJ = l Pfshy

I I I

EXAMPLE APPLICATION bull PHASE I

PLATFORM ST 151 K

bull GULF OF MEXICO

bull WATER DEPTH = 137 FT

bull ANNUAL MAXIMUM HS MEDIAN 175 ft COV 031

bull ASSUMED STORM DURATION 3 hoursmiddot

bull ASSUMED NUMBER OF WAVES 800

bull MEDIAN HS FOR STORM HOURS 1 2 AND 3

bull MEDIAN CURRENT VALUE bull

bull DETERMINE ULTIMATE CAPACITY FOR THREE DIRECTIONS

bull DETERMINE ANNUAL FAILURE PROBABILITIES

Broadsidemiddot Diagonal End-on

0037 0025 0007

0071 0044 0018

0022 0016 0004

DETERMINE TOTAL ANNUAL PROBABILITY OF FAILURE

Pf - 007

Pf - 013

Pr - 004

bull

EXAMPLE USE OF BIAS FACTORS - PHASE II middot

bull DETERMINE PREDICTIVE PROBABILITY OF FAILURE

Case - 1

Pf _ S S Pf (b b2) fa1a2 (b b2) db db2

g bull -1 (b1 R I S) - 1

92 - (b2 R2 I S) - 1

Case -2

Pf s ~ [bJ f11 (b) db

Pf [bJ p [g lt 0)

g - bull 1 (b1 R1 I S) - 1

1

bull1) l

I

BENEFITS OF ANDREW JIP bull PHASE II

lt

bull IMPROVED LOAD AND RESISTANCE RECIPE

bull IMPROVED CALIBRATION PROCEDURE

bull MULTIPLE BIAS FACTORS

bull PROCEDURE TO INCLUDE EFFECT OF OTHER PLATFORMS

bull PROCEDURE TO USE MULTIPLE BIAS FACTORS

  • Structure Bookmarks
    • ANDREW PHASE II JOINT1 INDUSTRY PROJECT
    • PROGRESS MEETING NUMBER 1
    • ANDREW PHASE II JOINT INDUSTRY PROJECT PROGRESS MEETING NUMBER 1 NOVEMBER 16 AGENDA
    • ANDREW PHASE II UST OF PARTICIPANTS
    • ANDREW PHASE I OBJECTIVES AND RESULTS
    • ANDREW PHASE II PROJECT OBJECTIVES
    • SUBSEQUENT PROJECTS
    • ANDREW PHASE II PROJECT TEAM
    • ANDREW PHASE II UPDATE MODELING middotAND ANALYSIS RECIPE
    • -ANDREW PHASE II CONDUCTOR RESISTANCE
    • ANDREW PHASE Il WA VE FORCE PROFILE
    • ANDREW PHASE II WA VE FORCE PROFILEmiddot
    • ANDREW PHASE II BRACE MODELING
    • ANDREW PHASE II FOUNDATIONS
    • ANDREW PHASE II middot middot JOINT MODELING
    • ANDREW PHASE Il JOINT MODELING OPTIONS DISCUSSION
    • ANDREW PHASE II JOINT MODEJJNG OPTIONS ILLUSTRATIONS
    • ANDREW PHASE II JOINT MODELING PHASE II METHOD
    • ANDREW PHASE II ANDREW STORM CONDmONS UPDATED
    • Platform Selection bullbullbull1111111111111
    • BasisCriteria for Selection bullbullbull1111111111111
    • CALIBRATION
    • WHERE TO INTRODUCE BIAS FACTOR (B
    • PHASE I -KEY CONCLUSIONS
    • bull CALIBRATION ISSUES PHASE II
    • PHASE II WORK SCOPE (CALIBRATION)
    • SOFTWARE
    • SENSITIVITY STUDIES POSTERIOR DISTRIBUTION OF B
    • WEIGHTING APPROACH bull
    • INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON
    • CATEGORIZATION OF ADDITIONAL PLATFORMS
    • UPDATING OF CALIBRATION RESULTS
    • APPLICATIONS OF BIAS FACTORS
    • EXAMPLE USE OF BIAS FACTORS
    • EXAMPLE APPLICATION bull PHASE I PLATFORM ST 151 K
    • EXAMPLE USE OF BIAS FACTORS -PHASE II middot
    • BENEFITS OF ANDREW JIP bull PHASE II
Page 41:  · . ' ANDREW PHASE II PROJECT OBJECTIVES . ' • ·UPDATE MODELING AND ANALYSIS RECIPE· JOINTS ' ' ·BRACES WAVE PROFILE FOUNDATIONS . ' ' CONDUCTORS· • ANALVZE PLATFORMS USING

_

~

~tl ~

NOltNttIJ 17lt ltNS S iJJ)y

Maximum Significant Wave Height - Hurricane Andrew 1992

30 ~

26 ~ I ~-~~~- == -shy -~ 35-_

24deg1--shy --------~- middot-middot-------middotmiddotmiddot----middot------middot-middot~middot- - - ~-----1---middot

22middot ~~~~~~~-~ -94 -92 -90 -88 -84deg-86 -82 -somiddot

c

-

Platform Selection bullbullbull1111111111111

0 BasisCriteria for Selection middot

0 Review of Phase I Platforms middot 0 Alternative Platforms

0 PMB Recommendations

0 Participants Comments

gt

L

l I

BasisCriteria for Selection bullbullbull1111111111111

0 Benefit for Calibration - Classification (UnexpectedExpected)

- Failure Mode (StructuresFoundations)

- Damaged StructuresKnown response

- Better Representation of Full ~opulation

0 Availability of Data - Existing Models

- Inspection Data

- Damage Data

PREDICTED BEHAVIOR

-

OBSERVED

BEHAVIOR

-SURVIVED DAMAGED FAILED

SURVIVED

ST134W WD90A

-MC311 -

MC397 -

--

-

-

ST151K ST130Q

-

-

-

DAMAGED - -

-

-

--

-

-

ST52 SS139 ST177B -

ST161A

-_

-

FAILED

-

-

-

ST151H ST130A ST72

Operator - Platfonns - Platfonns Survived Damagedamp

- Falled -

bull

-

-

Jle eBDIGIPllDbl -Amocoshy 30 1 Chevron -

143 12 Exxon 72 Mobll 22 1 Phllllps 7 bull

-Shell 39 Unocal 25 1

Total Partlclnts Platfonns 338 15

Other Operators bull- 310 13

-

Total Platfonns - -

648 28

fbml

Platforms Considered 6 7

tie of Total Platfonns 1 25

OVERVIEW OF PLATFORMS SURVIVEDDAMAGEDFAILED

-

shy

v

Platforms Evaluated in Phase I

-

-Platfonn Water Year Number Analysis Avallabillty of Availability Platform

- Name Depth rl of Model DetaDed rl ClassUlcatlon

Installation legs Inspection Soll Report Based on Information Phase I Calibration - -ft

- -- Suaill flalfnrm ~BSi -STISIK 137 1963 8 CAP Yes Yes Un eel Sarvlval

-STIJOO 170 1964 4 CAPmiddotshy Yes ST134data u ST134W 137 1981 4 CAP Yes Yesmiddot E

-

edSarvlval -

middotsurvival

WD90A 184 - 1964 8 CAP Yes E -

- Sarvlvill

MC3ll 343 1978 8 KARMA

MC397 468 1991 4 KARMA -

bull ~reglmmiddot- -

Yes Sure Survival

bull Yes - Sare Sanlval-

-T23STS2 63 1969 4 CAP

T2SSS139 62 1969 4 CAP -

Yes

Yes

Yes Ex

Yes Une

bull SurvivalLikely to DamalI

middot bull SanlvalLlkely to DamalI

ST161A 118 1964 8 USFOS

Eail11a pound1al(bullllm Clss ST177B 142 --1965 8 CAP

STISIH 137 1964 - 8 CAP C

STIJOA 140 19S8 8 CAP - -

Yes

Yes

ollamed

Yes

- Yes E

-

at about 1 mUe

Yes

ST134data

bull SafvlvalLlkelv to Dam- bull- -

E middotrahre

Emeried Failure -

Likely FaUare

T21ST72 61 1969 4 CAP Collamed No Likely FaUure

Platforms Identified for Evaluating Their Likelihood of Falling Under UnexpeCted Survival Category

CllEllATOa

AllB

llLOCll

Sl1lvcnJllE

NAME

YEAR

INSTAUJCD

WATER

IJBPJB

middotmiddotDISTANCE

ftOMSllOllE

(MIUIS)

Ill

(M)

AMOCO BllW

STllI

WDI075

A

bull D

4

n uo 17J

44

32

bull 20

10

bull CllEVllOH STtllO c I 1bull 21 11

STl130 D 0 161 21 11 STllJ4 I 0 137 D - 11

STIUI G I 137 32 11

STIUI I 0 121 32 11

STIUI

J 0 140 32 11

STIUI It 4 137 32 11

STllSI PROl)I 0 137 32 11

STllSI PROD-2 0 137 32 11

STll7 A 0

140 35 11

WDl117 c 65 214 34 10

DXON STshy E 51 14 bull STl165 A t3 bull 34 WD0073

WD0073

WD0073

A _ c

4

62

4

IA IA

17J

22

22

22

bullbullbull WD003-shy E 65 161 Z7

A 0 IN 31 MOBD Bl12I

ssn A bull

50

50

31

7

bull 7

MIJllPllY SS114 B 7 50 7 7

PlllUIPS SSl14t A 55 33 7

SAMEDAN ST17J A 4 117 7 bull 7

SJIELL WD0103 A 65 223 27 WDllOJ bull 65 221 27 WDll04 c 65 221 27 WDOI04 D 67 2 27

UNOCAL

SSl20I

SSl20I

bull SSl20I

SS0209

SS0209

SS0209

SS0209

SS 0253

SS 0253

E

I

PtJllE

A

bull D

G A

c

0

4

4

7J

61

0

0

69

103

97

97

9S

100

95

100

165

175

34

34

34

34

34

34

34

54

54

bullbullbullbullbullbullbullbullbull

Platforms Recommended for Phase II

llalr y_ w- Aot Av- _ I 1 Doplla Model nm bull bull _ bull I I H11 s18-1 Irht lbwl~ bullbullbull(-illtdlel OM lllMtD

II ---

-STISIIC 137 1963 I CAP Yeo Yeo I Yeo

STtshy -170 1964 4 CAP Yeo ST134dlU I Yeo

STl34W 137 Yeo Yeo -1981 4 CAP D Yeo

WD90A 114 1964 I CAP Yeo - D

MC311 30 1971 I KARMA Yeo -sre-

MC397 middot 461 1991 4 ICARMA Yeo Sireshy - - r- -

__ T13ST5l 63 1969 4 CAP Yeo Yeobull middot- ~n- D Yeo _ Tl5SSl39 62 1969 4 CAP Yeo Yeo I Yeo

_ STIT711 142 1965 I CAP Yeo I Yeo middot-1shySTl61A Ill 1964 I USFOS Yeo Yeo - D bulloshy

- r-

STISIH 137 1964 I CAP y D y -_ STl31A 140 1951 I CAP Yeo STl34dlU I Yeo

T21Sll 61 1969 4 CAP No rar-middot I Yeo

~-rmiddot-middot SS209A 95 1967 16 CAP Yeo D -SSU4R

11 UI A 41 -in D SS 2741AIM CAP bull

_ _ -shy - -shy- -shy

CALIBRATION

bull PHASEIAPPROACH

bull CALIBRATION ISSUES

bull PHASE II WORK SCOPE

bull SOFlWARE AVAILABILITY

bull EFFECT OF ADDITIONAL PLATFORMS

bull WEIGHTING APPROACH

bull EXAMPLE APPLICATION middot

(l=Ofce Mode] Wind

~

fcurrent l -middot

~SW ~leldl4

ultimate resistance force I first

design load Illgt

i _

deflection

-

bull

4

Historical Hurrlcane Dariiage and Survival of Platforms

Components Required for Calibration (Petrauskas 1992)

PRIOR Distribution

PLATFORM EXPERIENCE

BAYESIAN Survivals and Failures

UPDATING

Waves Winds and Currents bull

POSTERIOR Distribution

- A_middotJA ~ Survival Data Failure Data

Result Result

Load Resistance

middotfl

-

Bayesian Updaing middotGeneral Description (Petrauskas 1992)

L Imiddot

WHERE TO INTRODUCE BIAS FACTOR (B

On Resistance (Rmue - B (RPREDICreD I

middotOn Loads (Smue middot - B (SPRE01cTEo

On Safety Margin (RSmue - B (RSPREDICTED

CALIBRATION METHODOLOGY

gt

r

FIELD OBSERVATIONS

Observed Behavior

survivals Damages

Failures

r

bull

CAPACITY ANALYSIS

Predicted Load and Resistance Predicted Failure Modes

Survival Damage Failure cases

RELIABILITY ANALYSIS

- Probability of Failure ~

Ukellhood Failure

BAYESIAN UPDATING

Prior Distnbution of B Combined Likelihood Function

Posterior Dlstnbution of B

BIAS FACTOR

Mean Value cov bull

Represents Modeling Uncertainties

Prior Distribution of B Likelihood Functionmiddot Observing Success case STISIK and Posterior Distribution of B

2 00

uo

160

140

120

e 100 L $ uoIshy

0 60 ~ ~ 040

020

0 00

0 00 0 50

_ I 00

b

I 50 2 00

-shy lk [bl STISIK)

--shy Pnor-f(b)

--shy Pos1enor r(bl ISIK)

Prior or B Mean bull 10 COV bull 030

Posterlor of Bbull Meanbull 136COV 0 16

2 50

Likelihood Function and Posterior Distribution Example Platform STISIK

- -

~

- Prior and Posterior Distributions or B Failure cases (ST177B STlSlH ST130A T21) Damage cases (T23 T25 ST161A) Success cases (STlSlK ST130Q ST134W WD90A MC311 MC397)

400

350

300

2 so 0-

I

bull bull

bull bull gt

- bullbull ~ bull bull I bull

_ bull bull bullbull I bull bull bullII V -

bull bull

I bull ~ bull

bullbull bullI bull bullI bull bull bull

middot -~

bull__ ~ middotI middotmiddot

L 2 00

e ~ I SO

-I 00

- 0 so

000 gt

0 00 020 0 40 060 0 80 I 00 120 140 160 110 200

b -

-

shy

Prior-r

- - - - Posterior fLlll failure cases

bull bull bull bull bull bull bull Posterior flall damage cases

- bull bull - bull middot Postenor flall success cases

Posterior flall success fulure damage cases

-

Posterior of B Mean= 119 COV 0101

Posterior Distribution or Blas Factor (8) bull Different Categories and All Cases Combined

shy

l

PHASE I - KEY CONCLUSIONS

bull GLOBAL CORRECTION FACTOR

bull 15 TO 20 CONSERVATISM MEAN)

bull UNEXPECTED SURVIVALS HAVE GREATEST IMPACT

bull DAMAGED CASES MORE INFORMATIVE

bull CALIBRATION ISSUES PHASE II

bull

EFFECT OF IMPROVED CAPACITY ANALYSIS RECIPE

bull APPLICABILITY TO MULTIPLE FAILURE MODES

bull NO FOUNDATION DAMAGE CASE INCLUDED

bull IMPROVED REPRESENTATION OF FULL ANDREW DISTRIBUTION

-bull NUMBER OF PLATFORMS CONSIDERED

I 1 I

I t ) bull ~

PHASE II WORK SCOPE (CALIBRATION)

bull

bull ESTABLISH MULTIPLE BIAS FACTORS

Use Nine Base Case Platforms

middot - Capacity Analysis for Revised recipe

Foundation Blas Factor from Foundation JIP

bull DEVELOP WEIGHTING APPROACH

bull UPDATE MULTIPLE BIAS FACTORS

bullEffect of Platforms Not In the Nine Base Cases

bull EXAMPLE APPLICATION OF BIAS FACTORS I

bullbull

PHASE I PHASE II

CAPACITYbull ANALYSISRu bull

---------------------1----middot

Jc1 - iJbullraquoJ dbull

FttlS frt[f~l ratdhJJ

-

LIKELIHOOD FUNCTION

DEVELOPMENT

lk Jb Jfall1N) bull Jb)

111 Ill Isurvtvoq bull 1 - bl

-middot -------------------shy ~---------------------middot-middot

r111ir8 Cb1 111 Cbl BAYESIAN

UPDATING

~cove

R1R2 bull middot---------------------~----

Jblbull P(g cOJ

1 bull JbRI (MBSI I bull 1

~middot

llcJllJfalluro) bull P(gcOJ Ill 111 b21fallun) bull P ((g1 CO) U 1112 c OD

middot--------------------shy ----------------------shy bull

r111111r8 Cbl 111 Ill i-1nro1

deg ~1b2)laquottfrr (b1b21-lnfo) re Ill bull 1-11middot b2) bull lodegamp

middot

i1 covamp i2 cov~i

-

shy

--

Likelihood Function vs b All Success Case STISI K STl30Q STl34W WD90A MC311 MC397

I 0000

0 9000

OIOOO

0 7000

~ 0amp000

ta osoOO

~ 0 4000

0 3000

0 2000

0 1000

bull 0 0000

bull

I

middot

middot bull I

Ibull

-sibull ~ -middot-middot-middot-middot -middot-

----7----+-----~ I

Ibull bull bull I

rbull

--- Jk ISIK[blsuccess)

- bull - bull - lk 130Q[blsuccess) bull

lk 134W[blsuccess)

Jk-WD90A(blsuccc~I

- - - lk MC397[blsucccss)

--ltgt-- lk middotMC31 l[blsucccss)

--- Jk [bl 6 out of6succcsHascs)

05 I 5 2 250

b

Likelihood Functionsmiddot Success Cases - Phbullsc L

shy

bullbull bull bull bull bull bull -------------------~

bull bull bullbull bullbull bullbull bullbull bullbull bullbull bull

bull bullbull

bullbullbullbull

bullbull

bullbullbullbullbullbullbull

bullbullbullbullbullbull

bullbullbullbullbullbullbull

z 0

t z ) LL

c 0 0c i ~ i

J

~ sect ~--------------~ e l5

I

I

SOFTWARE

bull PF PROGRAM

bull CALREL

bull PF +CALREL

bull RELACS

bull

gt i

l

SENSITIVITY STUDIES

POSTERIOR DISTRIBUTION OF B

bull EFFECT OF CHANGE IN DISTRIBUTION OF PRIOR OF B

bull EFFECT OF REMOVAL OF ONE PLATFORM FROM EACH CATEGORY

bull EFFECT OF REMOVAL OF THREE DAMAGE CASES

bull EFFECT OF ADDITIONAL 13 SIMILAR PLATFORMS

-I

Prior and Posterior Distributions of B Variation In COV of PriorofB fromOl to04 -- -

4500

-4000

-

- -

3500 --middot - bull bull - bull bull Prior-r ICOV =02

3000

- - - - Prior-r ICOV =03 -

bull -~ e 2soo -

bull bull bull bull bull bull bull Pnor-r ICOV =o 4t frJ ~ 2000 Posterior rlCOV =0 2

bull bull Postenor rlCOV =0 3 bull bull 1500 bull middot bull - shy -middot- Posterior rlCOV =04

middot~ I 000 - bull

-middotmiddot ~ ~middotmiddot M -~

- middot 0 500

wbullbullbull _ I middot~ middotmiddotmiddotmiddotmiddot middotmiddotmiddotmiddotmiddot - -~ ~ __ middot_ _ bull -shy0 000

0 000 0200 0 400 0 600 0 800 IOOO 1200 1400 1600 1800 2000 -

- - b

-

=shy

Posterior Distribution of Bias Factor (B) bull Effect of Variation In COV of Prior of B

bull

bullPrior and Posterior Distributions of B Evaluating Effect of Removal of One Platform from Each Category from the Base Case of 13 Platforms

400

~ ~ 350

j IA i

3 00 bull I bull I ---Prior-r I bull l

250 1 I 1

I bull bull ---Posterior r lall 13 casesshy- ~E 2 00 bull bull bull bull bull bull bull Posterior r without STISIK

- ~ 1 bull - bull I - bull bull - bull bull Posterior r Iwithout T2I I 50 I

J-tt~i - bull - bull - Posterior r Iwithout T2S

tOO [-

I I ~ ~ I 0 50

0 00

___ r J i

l~T

~--

0 00 0 20 0 40 0 60 080 I 00 120 140 160 110 2 00

b

zshy

Posterior Distribution of Bias Factor (8) bull Effect of Removal of middotPlatform from Any Category

bullPrior and Posterior Distributions or B Evaluating Effect or Removal or 3 Damage Cases from the Base Case or 13 Platrorms

4

35

3

25

e L 2 ~ bulle shy

I 15

bull I

05

0

0 02 04 06 I 2 I 4 16 IS 2

I I

I

I

I Ibull

J -tmiddot L--

bullI

~

I

fl

II ~

I I v I

middot bull bull

bull L

~ ll- I-~lt

OS I

b

I

bull

1middot bullbullbullbull

~ __

bull

- - - - Prior-r

- - - - Postenor flall failure cases

Postenor flall success cases

--- Posterior flall 13 cases

---Posterior fI success failure cases only

IPosterior or B Mean =t11 COV =0121

-

_I

Posterir Distribution or Bias Factor (B) bull Effect or Removal or 3-Damage Cases

-

bullPrior and Posterior Distributions or B Evaluating Effect or Additional 13 Similar Observations as

Used in Andrew JIP The POsterior or B from Andrew-JIP Considered as New Prior or B

- 500

450

400

350

bull bull

bull bull bull bull J bull

bull bull f ~ I

I ~ Ibull bull

bullbull bull

l bull I -bullbull bull bull

I

I bull bull

bull I I~~bull bull bull

bull ~

I bull bull

~ ~_ ~~~

- - New Prior r bull

Posterior 114 new failure cases ~ 300- bull bull bull bull bull bull bull Posterior f 13 new damage casesl 150 ~

- bull bull - bull bull Posterior 116 new success cases i 2 00

Postenor 1113 new success failure -150 damage cases

I 00

0 50 -

0 00 0 00 0 20 0 40 0 60 0 80 I 00 120 I 40 160 110 2 00

Posterior ol B Mean= 120 COV= 0071 b

shy

shy

_rshy

Posterior Distribution of Bias Factor (B) bull Effect of Additional 13-Similar Observations

1

1 I

WEIGHTING APPROACH bull

bull ASSESSMENT OF PLATFORMS DATA

Select Platforms Most Mappable to the Analyzed Platforms

Unlikely SurvivalDamage Cases Identification

Based on Cut-Off Criteria Agemiddot Hs etc

bull INFORMATION REQUIRED FOR COMPARISON

bull CATEGORIZATION OF ADDITIONALmiddot PLATFORMS

bull UPDATING OF CALIBRATION RESULTS

--

OVERVIEW OF PLATFORMS SURVIVEDDAMAGED -~

Operator -

- Platforms

Survived CS)

Platforms

Damaged CD)

Platforms Used In

Phase IJIP -

JI~ ~Bill~IRBDll -

Amoco -

Chevron -

Exxon

Mob II

Phllllps

Shell

Unocal -

30

143

72 22

7

39 25shy -

1

12

bull

1 bull

1

1(S)+1 (D)

3 (S) +3(D)

2 (S)

bull shy

bull

bull

bull

Total Partlclnants Platforms 338 - 15 10C6S+4D) -

Other Operators

310

13

3 (D)

Total Platforms 648 28 13C6S+7D) -

Summary of Andrew JIP bull Phase I Platforms

Plairshy w v- N-ber Lea-le 1 shy car Hind_ o-w N-bull Deplh fl fl IJpe la bull n Yeadon Jadtet lbelledtf liefalit -- _ -OD

IL H - -shy _ICllshy IL IL - - rQClto ~ _

K STUIK 137 1963 8 365 608S4 Cellardeck

-middot STl-shy 170 1964 4 -~ K 39 62269 Cellardeck - --ST134W 137 1911 4 43 608S4 Smvlnlshy- _ - -

WD90A 184 1964 8 K 267Sbull34 83 SOS31 SUboftllar deck SmvMI

MC311 343 1978 8 x S7 62 - MC397 468 1991 4 x shy S33bull6SO 65911 -shySmeSmvlvol -- gt -

r-- T23ST52 63 1969 4 ton- shyKshy 52594 S03 -T25SSl39 62 1969 4 Kshy SOS94 S063 - middot ton-shy__ -

bullbull a ton--shyST161A 118 1964 8 K - 347S 51495 Cellarclecti

r~

ST1771 142 1965 8 K 39 60167 CeDordeck failme

STISIH 137 1964 8 K 395 608S4 CeDardeck Failme

K ST130A 140 1958 a SS 6095 I lblv Pai1uno - T21ST72 61 1969 4 Kshy S4S94 4974 I lhlv Pai1uno -

- -

JE

INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON

bull WATER DEPTH

bull bull YEAR OF INSTALLATIONDESIGN

bull PLATFORM ORIENTATION

bull PLATFORM LATITUDE AND LONGITUDE

bull KEY SKETCHES OF PLATFORM STRUCTURAL FRAMING

Number of Legs Plies Leg Batter Number of Conductors bull Deck Elevations (Including sub-cellar deck If any) Jacket Framing Elevations Brace type In Vertical Frames Sizes of Primary Members (leg pile vertical frames braces primary horizontals) Joint Details (can gap etc) LegPiie Annulus Grouting Status

bull middotPOST ANDREW INSPECTION REPORT Significant Wave Height Hs (ft)

bull INFORMATION ON DAMAGES PRIOR TO ANDREW

bull RESULTS FROM IN-HOUSE STUDIES IF AVAILABLE

bull ~RIGINAL DESIGN CRITERIA INFORMATION IF AVAILABLE

bull SOIL SHEAR STRENGTH PROFILE IF AVAILABLE

Platforms Identified for Evaluating Therr Likelihood of Falling Under Unexpected Survival Category

OPlRATOR

middotADA

BLOClt

llmucnJllE

NAME

YEAll

INSTAUJD

WA1ER

DBPTll

DISTANCE

JlllOM SHOllE

(MILU)

Bl

(M)

AMOCO 111113

STl11

WDI075

A

bull D

St

17

120

1n

44

20

10 I

CllllVllQfl STlllO c 61 110 21 11 STIUO D Q 1110 21 11

STl134 p Q 137 29 11

ST0151 G 61 137 11

ST0151 I 3 121 11

ST0151 J Q 140 11

STl151 It 137 11

ST0151 PJl001 Q UT 11

ST0151 PROD-2 Q UT 11

ST017 A 3 140 35 11

WDIUT c 65 214 34 10

llXXON STI055 II 51 6T 14 I

STl15 A t3 34

WDI073

WD0073

Amiddotshy Q

1111 1111

22

22

I

I

WDI073 c 1n 22 I

- WDOCl93-shy II

65 1110 2T A 3 194 31 MOBU 111121

SS IOl7

A B

501

50

50

31

T

I

T 6

MtlllPllY SS114 B T 50 T T

PlllllIPS SS0149 A 5 55 33 T

SAMllDAN STln A 107 T T

SBllLL WDl103 A 65 223 2T WDl103 B 65 221 2T WD0104 c 65 221 2T WD0104 D 6T 20 2T

UNOCAL SS020I

SS 020I SSl20I

SS 0209

SS0209

II p

PURll

A

B

3

n 61

103

97

97

95

100

34

34

34

34 34

bullbull bullbullbull

- SS0209 D u 95 34 I

- SS 1209 SSl253

SS0253

G A

c

6T

2

100

165

1T5

34

54

54

bullbull I

I I

I I

SOUTH MARCH

~LAND

N

SOUTH PaTOI ENE ISWltD AREA

MMS Defined Limits ror Offshore Structures Requiring Inspections Following Andrew

t---Jbulln3 T25bull I bullTtl

bull1$1Htl1Abull bull1SIK1779bull

~

middot-DaTA AREA~T

CRANO rsLE

AREA bull1300

bullll4W

=shy

bullPath of Hurricane Andrew with Platforms Used In Calibration

Likelihood Function vs b All Success Case STISI K STIJOQ ST134W WD90A MCJll MCJ97

I 0000

0 9000

0 8000

07000

-Bo6000

tiiii 0 5000 -

04000

0 3000

0 2000

01000

0 0000

0 05 25

ri1rt middot-middot-middot-middot-middot-

I 5 2

b

---11 middot151K(blsu~ccsI

- bull - bull - fk 130Q(bfU~lCSS)

lie bull134W(blsuccc]

lk-WD90A[blsultCS]

- - - lk middotMC397(bfUtlCS]

--o-- lk-MCJll(bbuctcsJ

--- ll (bl 6 out of 6 Mlttcs case]

_

( Likelihood Functions middotSuccess Cases - p - 1 c I

CATEGORIZATION OF ADDITIONAL PLATFORMS

I I I

bull CATEGORIES

Unexpected Survivals

Expected Survivals

bull APPROACH

Identify Corresponding Platform(s) Evaluated In the JIP

Determine Max Wave Height at Platform Location

Evaluate Whether Wave Hit the Deck

middot - Evaluate Likelihood of R and S to Differ from the Base Cases

Review Joint Information

Assess Special Features

bull IDENTIFY NUMBER OF PLATFORMS IN EACH CATEGORY

I

UPDATING OF CALIBRATION RESULTS

bull INCLUDE EFFECT OF ADDITIONAL PLATFORMS ON bullBbull

Consider Increased 1Uncertalntles In R and S

Develop bullNeW Ukellhood Functions for Different Categories

middotLet bullxbull Number of Ukely Unexpected Survival Cases and y Number of Ukely Expected Survival Cases

Determine Combined Likelihood Function lk (b Iadditional platforms) middot

bull REVISED LIKELIHOOD FUNCTION

lk(b I all cases) = lk(b I 9 platforms) middot x lk(b Ix platforms) x lk(b I y platforms)

bull ESTABLISH REVISED BIAS FACTOR B

f8 (b) u f8 (b) lk (b I all cases)

II

t t ~

shy

APPLICATIONS OF BIAS FACTORS

ii APPLICABLE TO PARTICULAR LOAD AND RESISTANCE RECIPE

bull USE BY THE API COMMITTEES AND MMS

Criteria Revisions

Recommended Design Guidelines Revisions bull

bullbull REASSESSMENT STUDIES AND DECISION MAKING

middot

bull RISK ASSESSMENT STUDIES BY THE OPERATORS

Individual Platforms

Fleet of Platforms

middot IMR Program Implementation

bullbull NOT TO BE USED IN DESIGN

I I fl J

EXAMPLE USE OF BIAS FACTORS

bull PHASE I EXAMPLE APPLICATION

bull EVALUATION OF ANNUAL PROBABILITY OF FAILURE

Establish Annual Maximum Seastate Data

Determine Coefficients to Define Base Shear

Determine Ultimate Capacity of Platform

Determine Probablllty of Failure Versus bullbbull

-bull DETERMINE UPDATED PROBABILITY OF FAILURE

Pf - P [S gt R]

Pf I P [~ gt Rb I b] f 8 (b) db

Pr - J P [S gt Rb I b] f bull 8 (b) db

bull TOTAL PROBABILITIES OF FAILURE

Total Pf - 1 bull Il [1 middotPf] - l Pf bull

Total Pf - 1 bull Il [1middot Pf] - l Pfmiddot

Total Pr = 1 - n [1middot PfbullJ = l Pfshy

I I I

EXAMPLE APPLICATION bull PHASE I

PLATFORM ST 151 K

bull GULF OF MEXICO

bull WATER DEPTH = 137 FT

bull ANNUAL MAXIMUM HS MEDIAN 175 ft COV 031

bull ASSUMED STORM DURATION 3 hoursmiddot

bull ASSUMED NUMBER OF WAVES 800

bull MEDIAN HS FOR STORM HOURS 1 2 AND 3

bull MEDIAN CURRENT VALUE bull

bull DETERMINE ULTIMATE CAPACITY FOR THREE DIRECTIONS

bull DETERMINE ANNUAL FAILURE PROBABILITIES

Broadsidemiddot Diagonal End-on

0037 0025 0007

0071 0044 0018

0022 0016 0004

DETERMINE TOTAL ANNUAL PROBABILITY OF FAILURE

Pf - 007

Pf - 013

Pr - 004

bull

EXAMPLE USE OF BIAS FACTORS - PHASE II middot

bull DETERMINE PREDICTIVE PROBABILITY OF FAILURE

Case - 1

Pf _ S S Pf (b b2) fa1a2 (b b2) db db2

g bull -1 (b1 R I S) - 1

92 - (b2 R2 I S) - 1

Case -2

Pf s ~ [bJ f11 (b) db

Pf [bJ p [g lt 0)

g - bull 1 (b1 R1 I S) - 1

1

bull1) l

I

BENEFITS OF ANDREW JIP bull PHASE II

lt

bull IMPROVED LOAD AND RESISTANCE RECIPE

bull IMPROVED CALIBRATION PROCEDURE

bull MULTIPLE BIAS FACTORS

bull PROCEDURE TO INCLUDE EFFECT OF OTHER PLATFORMS

bull PROCEDURE TO USE MULTIPLE BIAS FACTORS

  • Structure Bookmarks
    • ANDREW PHASE II JOINT1 INDUSTRY PROJECT
    • PROGRESS MEETING NUMBER 1
    • ANDREW PHASE II JOINT INDUSTRY PROJECT PROGRESS MEETING NUMBER 1 NOVEMBER 16 AGENDA
    • ANDREW PHASE II UST OF PARTICIPANTS
    • ANDREW PHASE I OBJECTIVES AND RESULTS
    • ANDREW PHASE II PROJECT OBJECTIVES
    • SUBSEQUENT PROJECTS
    • ANDREW PHASE II PROJECT TEAM
    • ANDREW PHASE II UPDATE MODELING middotAND ANALYSIS RECIPE
    • -ANDREW PHASE II CONDUCTOR RESISTANCE
    • ANDREW PHASE Il WA VE FORCE PROFILE
    • ANDREW PHASE II WA VE FORCE PROFILEmiddot
    • ANDREW PHASE II BRACE MODELING
    • ANDREW PHASE II FOUNDATIONS
    • ANDREW PHASE II middot middot JOINT MODELING
    • ANDREW PHASE Il JOINT MODELING OPTIONS DISCUSSION
    • ANDREW PHASE II JOINT MODEJJNG OPTIONS ILLUSTRATIONS
    • ANDREW PHASE II JOINT MODELING PHASE II METHOD
    • ANDREW PHASE II ANDREW STORM CONDmONS UPDATED
    • Platform Selection bullbullbull1111111111111
    • BasisCriteria for Selection bullbullbull1111111111111
    • CALIBRATION
    • WHERE TO INTRODUCE BIAS FACTOR (B
    • PHASE I -KEY CONCLUSIONS
    • bull CALIBRATION ISSUES PHASE II
    • PHASE II WORK SCOPE (CALIBRATION)
    • SOFTWARE
    • SENSITIVITY STUDIES POSTERIOR DISTRIBUTION OF B
    • WEIGHTING APPROACH bull
    • INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON
    • CATEGORIZATION OF ADDITIONAL PLATFORMS
    • UPDATING OF CALIBRATION RESULTS
    • APPLICATIONS OF BIAS FACTORS
    • EXAMPLE USE OF BIAS FACTORS
    • EXAMPLE APPLICATION bull PHASE I PLATFORM ST 151 K
    • EXAMPLE USE OF BIAS FACTORS -PHASE II middot
    • BENEFITS OF ANDREW JIP bull PHASE II
Page 42:  · . ' ANDREW PHASE II PROJECT OBJECTIVES . ' • ·UPDATE MODELING AND ANALYSIS RECIPE· JOINTS ' ' ·BRACES WAVE PROFILE FOUNDATIONS . ' ' CONDUCTORS· • ANALVZE PLATFORMS USING

-

Platform Selection bullbullbull1111111111111

0 BasisCriteria for Selection middot

0 Review of Phase I Platforms middot 0 Alternative Platforms

0 PMB Recommendations

0 Participants Comments

gt

L

l I

BasisCriteria for Selection bullbullbull1111111111111

0 Benefit for Calibration - Classification (UnexpectedExpected)

- Failure Mode (StructuresFoundations)

- Damaged StructuresKnown response

- Better Representation of Full ~opulation

0 Availability of Data - Existing Models

- Inspection Data

- Damage Data

PREDICTED BEHAVIOR

-

OBSERVED

BEHAVIOR

-SURVIVED DAMAGED FAILED

SURVIVED

ST134W WD90A

-MC311 -

MC397 -

--

-

-

ST151K ST130Q

-

-

-

DAMAGED - -

-

-

--

-

-

ST52 SS139 ST177B -

ST161A

-_

-

FAILED

-

-

-

ST151H ST130A ST72

Operator - Platfonns - Platfonns Survived Damagedamp

- Falled -

bull

-

-

Jle eBDIGIPllDbl -Amocoshy 30 1 Chevron -

143 12 Exxon 72 Mobll 22 1 Phllllps 7 bull

-Shell 39 Unocal 25 1

Total Partlclnts Platfonns 338 15

Other Operators bull- 310 13

-

Total Platfonns - -

648 28

fbml

Platforms Considered 6 7

tie of Total Platfonns 1 25

OVERVIEW OF PLATFORMS SURVIVEDDAMAGEDFAILED

-

shy

v

Platforms Evaluated in Phase I

-

-Platfonn Water Year Number Analysis Avallabillty of Availability Platform

- Name Depth rl of Model DetaDed rl ClassUlcatlon

Installation legs Inspection Soll Report Based on Information Phase I Calibration - -ft

- -- Suaill flalfnrm ~BSi -STISIK 137 1963 8 CAP Yes Yes Un eel Sarvlval

-STIJOO 170 1964 4 CAPmiddotshy Yes ST134data u ST134W 137 1981 4 CAP Yes Yesmiddot E

-

edSarvlval -

middotsurvival

WD90A 184 - 1964 8 CAP Yes E -

- Sarvlvill

MC3ll 343 1978 8 KARMA

MC397 468 1991 4 KARMA -

bull ~reglmmiddot- -

Yes Sure Survival

bull Yes - Sare Sanlval-

-T23STS2 63 1969 4 CAP

T2SSS139 62 1969 4 CAP -

Yes

Yes

Yes Ex

Yes Une

bull SurvivalLikely to DamalI

middot bull SanlvalLlkely to DamalI

ST161A 118 1964 8 USFOS

Eail11a pound1al(bullllm Clss ST177B 142 --1965 8 CAP

STISIH 137 1964 - 8 CAP C

STIJOA 140 19S8 8 CAP - -

Yes

Yes

ollamed

Yes

- Yes E

-

at about 1 mUe

Yes

ST134data

bull SafvlvalLlkelv to Dam- bull- -

E middotrahre

Emeried Failure -

Likely FaUare

T21ST72 61 1969 4 CAP Collamed No Likely FaUure

Platforms Identified for Evaluating Their Likelihood of Falling Under UnexpeCted Survival Category

CllEllATOa

AllB

llLOCll

Sl1lvcnJllE

NAME

YEAR

INSTAUJCD

WATER

IJBPJB

middotmiddotDISTANCE

ftOMSllOllE

(MIUIS)

Ill

(M)

AMOCO BllW

STllI

WDI075

A

bull D

4

n uo 17J

44

32

bull 20

10

bull CllEVllOH STtllO c I 1bull 21 11

STl130 D 0 161 21 11 STllJ4 I 0 137 D - 11

STIUI G I 137 32 11

STIUI I 0 121 32 11

STIUI

J 0 140 32 11

STIUI It 4 137 32 11

STllSI PROl)I 0 137 32 11

STllSI PROD-2 0 137 32 11

STll7 A 0

140 35 11

WDl117 c 65 214 34 10

DXON STshy E 51 14 bull STl165 A t3 bull 34 WD0073

WD0073

WD0073

A _ c

4

62

4

IA IA

17J

22

22

22

bullbullbull WD003-shy E 65 161 Z7

A 0 IN 31 MOBD Bl12I

ssn A bull

50

50

31

7

bull 7

MIJllPllY SS114 B 7 50 7 7

PlllUIPS SSl14t A 55 33 7

SAMEDAN ST17J A 4 117 7 bull 7

SJIELL WD0103 A 65 223 27 WDllOJ bull 65 221 27 WDll04 c 65 221 27 WDOI04 D 67 2 27

UNOCAL

SSl20I

SSl20I

bull SSl20I

SS0209

SS0209

SS0209

SS0209

SS 0253

SS 0253

E

I

PtJllE

A

bull D

G A

c

0

4

4

7J

61

0

0

69

103

97

97

9S

100

95

100

165

175

34

34

34

34

34

34

34

54

54

bullbullbullbullbullbullbullbullbull

Platforms Recommended for Phase II

llalr y_ w- Aot Av- _ I 1 Doplla Model nm bull bull _ bull I I H11 s18-1 Irht lbwl~ bullbullbull(-illtdlel OM lllMtD

II ---

-STISIIC 137 1963 I CAP Yeo Yeo I Yeo

STtshy -170 1964 4 CAP Yeo ST134dlU I Yeo

STl34W 137 Yeo Yeo -1981 4 CAP D Yeo

WD90A 114 1964 I CAP Yeo - D

MC311 30 1971 I KARMA Yeo -sre-

MC397 middot 461 1991 4 ICARMA Yeo Sireshy - - r- -

__ T13ST5l 63 1969 4 CAP Yeo Yeobull middot- ~n- D Yeo _ Tl5SSl39 62 1969 4 CAP Yeo Yeo I Yeo

_ STIT711 142 1965 I CAP Yeo I Yeo middot-1shySTl61A Ill 1964 I USFOS Yeo Yeo - D bulloshy

- r-

STISIH 137 1964 I CAP y D y -_ STl31A 140 1951 I CAP Yeo STl34dlU I Yeo

T21Sll 61 1969 4 CAP No rar-middot I Yeo

~-rmiddot-middot SS209A 95 1967 16 CAP Yeo D -SSU4R

11 UI A 41 -in D SS 2741AIM CAP bull

_ _ -shy - -shy- -shy

CALIBRATION

bull PHASEIAPPROACH

bull CALIBRATION ISSUES

bull PHASE II WORK SCOPE

bull SOFlWARE AVAILABILITY

bull EFFECT OF ADDITIONAL PLATFORMS

bull WEIGHTING APPROACH

bull EXAMPLE APPLICATION middot

(l=Ofce Mode] Wind

~

fcurrent l -middot

~SW ~leldl4

ultimate resistance force I first

design load Illgt

i _

deflection

-

bull

4

Historical Hurrlcane Dariiage and Survival of Platforms

Components Required for Calibration (Petrauskas 1992)

PRIOR Distribution

PLATFORM EXPERIENCE

BAYESIAN Survivals and Failures

UPDATING

Waves Winds and Currents bull

POSTERIOR Distribution

- A_middotJA ~ Survival Data Failure Data

Result Result

Load Resistance

middotfl

-

Bayesian Updaing middotGeneral Description (Petrauskas 1992)

L Imiddot

WHERE TO INTRODUCE BIAS FACTOR (B

On Resistance (Rmue - B (RPREDICreD I

middotOn Loads (Smue middot - B (SPRE01cTEo

On Safety Margin (RSmue - B (RSPREDICTED

CALIBRATION METHODOLOGY

gt

r

FIELD OBSERVATIONS

Observed Behavior

survivals Damages

Failures

r

bull

CAPACITY ANALYSIS

Predicted Load and Resistance Predicted Failure Modes

Survival Damage Failure cases

RELIABILITY ANALYSIS

- Probability of Failure ~

Ukellhood Failure

BAYESIAN UPDATING

Prior Distnbution of B Combined Likelihood Function

Posterior Dlstnbution of B

BIAS FACTOR

Mean Value cov bull

Represents Modeling Uncertainties

Prior Distribution of B Likelihood Functionmiddot Observing Success case STISIK and Posterior Distribution of B

2 00

uo

160

140

120

e 100 L $ uoIshy

0 60 ~ ~ 040

020

0 00

0 00 0 50

_ I 00

b

I 50 2 00

-shy lk [bl STISIK)

--shy Pnor-f(b)

--shy Pos1enor r(bl ISIK)

Prior or B Mean bull 10 COV bull 030

Posterlor of Bbull Meanbull 136COV 0 16

2 50

Likelihood Function and Posterior Distribution Example Platform STISIK

- -

~

- Prior and Posterior Distributions or B Failure cases (ST177B STlSlH ST130A T21) Damage cases (T23 T25 ST161A) Success cases (STlSlK ST130Q ST134W WD90A MC311 MC397)

400

350

300

2 so 0-

I

bull bull

bull bull gt

- bullbull ~ bull bull I bull

_ bull bull bullbull I bull bull bullII V -

bull bull

I bull ~ bull

bullbull bullI bull bullI bull bull bull

middot -~

bull__ ~ middotI middotmiddot

L 2 00

e ~ I SO

-I 00

- 0 so

000 gt

0 00 020 0 40 060 0 80 I 00 120 140 160 110 200

b -

-

shy

Prior-r

- - - - Posterior fLlll failure cases

bull bull bull bull bull bull bull Posterior flall damage cases

- bull bull - bull middot Postenor flall success cases

Posterior flall success fulure damage cases

-

Posterior of B Mean= 119 COV 0101

Posterior Distribution or Blas Factor (8) bull Different Categories and All Cases Combined

shy

l

PHASE I - KEY CONCLUSIONS

bull GLOBAL CORRECTION FACTOR

bull 15 TO 20 CONSERVATISM MEAN)

bull UNEXPECTED SURVIVALS HAVE GREATEST IMPACT

bull DAMAGED CASES MORE INFORMATIVE

bull CALIBRATION ISSUES PHASE II

bull

EFFECT OF IMPROVED CAPACITY ANALYSIS RECIPE

bull APPLICABILITY TO MULTIPLE FAILURE MODES

bull NO FOUNDATION DAMAGE CASE INCLUDED

bull IMPROVED REPRESENTATION OF FULL ANDREW DISTRIBUTION

-bull NUMBER OF PLATFORMS CONSIDERED

I 1 I

I t ) bull ~

PHASE II WORK SCOPE (CALIBRATION)

bull

bull ESTABLISH MULTIPLE BIAS FACTORS

Use Nine Base Case Platforms

middot - Capacity Analysis for Revised recipe

Foundation Blas Factor from Foundation JIP

bull DEVELOP WEIGHTING APPROACH

bull UPDATE MULTIPLE BIAS FACTORS

bullEffect of Platforms Not In the Nine Base Cases

bull EXAMPLE APPLICATION OF BIAS FACTORS I

bullbull

PHASE I PHASE II

CAPACITYbull ANALYSISRu bull

---------------------1----middot

Jc1 - iJbullraquoJ dbull

FttlS frt[f~l ratdhJJ

-

LIKELIHOOD FUNCTION

DEVELOPMENT

lk Jb Jfall1N) bull Jb)

111 Ill Isurvtvoq bull 1 - bl

-middot -------------------shy ~---------------------middot-middot

r111ir8 Cb1 111 Cbl BAYESIAN

UPDATING

~cove

R1R2 bull middot---------------------~----

Jblbull P(g cOJ

1 bull JbRI (MBSI I bull 1

~middot

llcJllJfalluro) bull P(gcOJ Ill 111 b21fallun) bull P ((g1 CO) U 1112 c OD

middot--------------------shy ----------------------shy bull

r111111r8 Cbl 111 Ill i-1nro1

deg ~1b2)laquottfrr (b1b21-lnfo) re Ill bull 1-11middot b2) bull lodegamp

middot

i1 covamp i2 cov~i

-

shy

--

Likelihood Function vs b All Success Case STISI K STl30Q STl34W WD90A MC311 MC397

I 0000

0 9000

OIOOO

0 7000

~ 0amp000

ta osoOO

~ 0 4000

0 3000

0 2000

0 1000

bull 0 0000

bull

I

middot

middot bull I

Ibull

-sibull ~ -middot-middot-middot-middot -middot-

----7----+-----~ I

Ibull bull bull I

rbull

--- Jk ISIK[blsuccess)

- bull - bull - lk 130Q[blsuccess) bull

lk 134W[blsuccess)

Jk-WD90A(blsuccc~I

- - - lk MC397[blsucccss)

--ltgt-- lk middotMC31 l[blsucccss)

--- Jk [bl 6 out of6succcsHascs)

05 I 5 2 250

b

Likelihood Functionsmiddot Success Cases - Phbullsc L

shy

bullbull bull bull bull bull bull -------------------~

bull bull bullbull bullbull bullbull bullbull bullbull bullbull bull

bull bullbull

bullbullbullbull

bullbull

bullbullbullbullbullbullbull

bullbullbullbullbullbull

bullbullbullbullbullbullbull

z 0

t z ) LL

c 0 0c i ~ i

J

~ sect ~--------------~ e l5

I

I

SOFTWARE

bull PF PROGRAM

bull CALREL

bull PF +CALREL

bull RELACS

bull

gt i

l

SENSITIVITY STUDIES

POSTERIOR DISTRIBUTION OF B

bull EFFECT OF CHANGE IN DISTRIBUTION OF PRIOR OF B

bull EFFECT OF REMOVAL OF ONE PLATFORM FROM EACH CATEGORY

bull EFFECT OF REMOVAL OF THREE DAMAGE CASES

bull EFFECT OF ADDITIONAL 13 SIMILAR PLATFORMS

-I

Prior and Posterior Distributions of B Variation In COV of PriorofB fromOl to04 -- -

4500

-4000

-

- -

3500 --middot - bull bull - bull bull Prior-r ICOV =02

3000

- - - - Prior-r ICOV =03 -

bull -~ e 2soo -

bull bull bull bull bull bull bull Pnor-r ICOV =o 4t frJ ~ 2000 Posterior rlCOV =0 2

bull bull Postenor rlCOV =0 3 bull bull 1500 bull middot bull - shy -middot- Posterior rlCOV =04

middot~ I 000 - bull

-middotmiddot ~ ~middotmiddot M -~

- middot 0 500

wbullbullbull _ I middot~ middotmiddotmiddotmiddotmiddot middotmiddotmiddotmiddotmiddot - -~ ~ __ middot_ _ bull -shy0 000

0 000 0200 0 400 0 600 0 800 IOOO 1200 1400 1600 1800 2000 -

- - b

-

=shy

Posterior Distribution of Bias Factor (B) bull Effect of Variation In COV of Prior of B

bull

bullPrior and Posterior Distributions of B Evaluating Effect of Removal of One Platform from Each Category from the Base Case of 13 Platforms

400

~ ~ 350

j IA i

3 00 bull I bull I ---Prior-r I bull l

250 1 I 1

I bull bull ---Posterior r lall 13 casesshy- ~E 2 00 bull bull bull bull bull bull bull Posterior r without STISIK

- ~ 1 bull - bull I - bull bull - bull bull Posterior r Iwithout T2I I 50 I

J-tt~i - bull - bull - Posterior r Iwithout T2S

tOO [-

I I ~ ~ I 0 50

0 00

___ r J i

l~T

~--

0 00 0 20 0 40 0 60 080 I 00 120 140 160 110 2 00

b

zshy

Posterior Distribution of Bias Factor (8) bull Effect of Removal of middotPlatform from Any Category

bullPrior and Posterior Distributions or B Evaluating Effect or Removal or 3 Damage Cases from the Base Case or 13 Platrorms

4

35

3

25

e L 2 ~ bulle shy

I 15

bull I

05

0

0 02 04 06 I 2 I 4 16 IS 2

I I

I

I

I Ibull

J -tmiddot L--

bullI

~

I

fl

II ~

I I v I

middot bull bull

bull L

~ ll- I-~lt

OS I

b

I

bull

1middot bullbullbullbull

~ __

bull

- - - - Prior-r

- - - - Postenor flall failure cases

Postenor flall success cases

--- Posterior flall 13 cases

---Posterior fI success failure cases only

IPosterior or B Mean =t11 COV =0121

-

_I

Posterir Distribution or Bias Factor (B) bull Effect or Removal or 3-Damage Cases

-

bullPrior and Posterior Distributions or B Evaluating Effect or Additional 13 Similar Observations as

Used in Andrew JIP The POsterior or B from Andrew-JIP Considered as New Prior or B

- 500

450

400

350

bull bull

bull bull bull bull J bull

bull bull f ~ I

I ~ Ibull bull

bullbull bull

l bull I -bullbull bull bull

I

I bull bull

bull I I~~bull bull bull

bull ~

I bull bull

~ ~_ ~~~

- - New Prior r bull

Posterior 114 new failure cases ~ 300- bull bull bull bull bull bull bull Posterior f 13 new damage casesl 150 ~

- bull bull - bull bull Posterior 116 new success cases i 2 00

Postenor 1113 new success failure -150 damage cases

I 00

0 50 -

0 00 0 00 0 20 0 40 0 60 0 80 I 00 120 I 40 160 110 2 00

Posterior ol B Mean= 120 COV= 0071 b

shy

shy

_rshy

Posterior Distribution of Bias Factor (B) bull Effect of Additional 13-Similar Observations

1

1 I

WEIGHTING APPROACH bull

bull ASSESSMENT OF PLATFORMS DATA

Select Platforms Most Mappable to the Analyzed Platforms

Unlikely SurvivalDamage Cases Identification

Based on Cut-Off Criteria Agemiddot Hs etc

bull INFORMATION REQUIRED FOR COMPARISON

bull CATEGORIZATION OF ADDITIONALmiddot PLATFORMS

bull UPDATING OF CALIBRATION RESULTS

--

OVERVIEW OF PLATFORMS SURVIVEDDAMAGED -~

Operator -

- Platforms

Survived CS)

Platforms

Damaged CD)

Platforms Used In

Phase IJIP -

JI~ ~Bill~IRBDll -

Amoco -

Chevron -

Exxon

Mob II

Phllllps

Shell

Unocal -

30

143

72 22

7

39 25shy -

1

12

bull

1 bull

1

1(S)+1 (D)

3 (S) +3(D)

2 (S)

bull shy

bull

bull

bull

Total Partlclnants Platforms 338 - 15 10C6S+4D) -

Other Operators

310

13

3 (D)

Total Platforms 648 28 13C6S+7D) -

Summary of Andrew JIP bull Phase I Platforms

Plairshy w v- N-ber Lea-le 1 shy car Hind_ o-w N-bull Deplh fl fl IJpe la bull n Yeadon Jadtet lbelledtf liefalit -- _ -OD

IL H - -shy _ICllshy IL IL - - rQClto ~ _

K STUIK 137 1963 8 365 608S4 Cellardeck

-middot STl-shy 170 1964 4 -~ K 39 62269 Cellardeck - --ST134W 137 1911 4 43 608S4 Smvlnlshy- _ - -

WD90A 184 1964 8 K 267Sbull34 83 SOS31 SUboftllar deck SmvMI

MC311 343 1978 8 x S7 62 - MC397 468 1991 4 x shy S33bull6SO 65911 -shySmeSmvlvol -- gt -

r-- T23ST52 63 1969 4 ton- shyKshy 52594 S03 -T25SSl39 62 1969 4 Kshy SOS94 S063 - middot ton-shy__ -

bullbull a ton--shyST161A 118 1964 8 K - 347S 51495 Cellarclecti

r~

ST1771 142 1965 8 K 39 60167 CeDordeck failme

STISIH 137 1964 8 K 395 608S4 CeDardeck Failme

K ST130A 140 1958 a SS 6095 I lblv Pai1uno - T21ST72 61 1969 4 Kshy S4S94 4974 I lhlv Pai1uno -

- -

JE

INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON

bull WATER DEPTH

bull bull YEAR OF INSTALLATIONDESIGN

bull PLATFORM ORIENTATION

bull PLATFORM LATITUDE AND LONGITUDE

bull KEY SKETCHES OF PLATFORM STRUCTURAL FRAMING

Number of Legs Plies Leg Batter Number of Conductors bull Deck Elevations (Including sub-cellar deck If any) Jacket Framing Elevations Brace type In Vertical Frames Sizes of Primary Members (leg pile vertical frames braces primary horizontals) Joint Details (can gap etc) LegPiie Annulus Grouting Status

bull middotPOST ANDREW INSPECTION REPORT Significant Wave Height Hs (ft)

bull INFORMATION ON DAMAGES PRIOR TO ANDREW

bull RESULTS FROM IN-HOUSE STUDIES IF AVAILABLE

bull ~RIGINAL DESIGN CRITERIA INFORMATION IF AVAILABLE

bull SOIL SHEAR STRENGTH PROFILE IF AVAILABLE

Platforms Identified for Evaluating Therr Likelihood of Falling Under Unexpected Survival Category

OPlRATOR

middotADA

BLOClt

llmucnJllE

NAME

YEAll

INSTAUJD

WA1ER

DBPTll

DISTANCE

JlllOM SHOllE

(MILU)

Bl

(M)

AMOCO 111113

STl11

WDI075

A

bull D

St

17

120

1n

44

20

10 I

CllllVllQfl STlllO c 61 110 21 11 STIUO D Q 1110 21 11

STl134 p Q 137 29 11

ST0151 G 61 137 11

ST0151 I 3 121 11

ST0151 J Q 140 11

STl151 It 137 11

ST0151 PJl001 Q UT 11

ST0151 PROD-2 Q UT 11

ST017 A 3 140 35 11

WDIUT c 65 214 34 10

llXXON STI055 II 51 6T 14 I

STl15 A t3 34

WDI073

WD0073

Amiddotshy Q

1111 1111

22

22

I

I

WDI073 c 1n 22 I

- WDOCl93-shy II

65 1110 2T A 3 194 31 MOBU 111121

SS IOl7

A B

501

50

50

31

T

I

T 6

MtlllPllY SS114 B T 50 T T

PlllllIPS SS0149 A 5 55 33 T

SAMllDAN STln A 107 T T

SBllLL WDl103 A 65 223 2T WDl103 B 65 221 2T WD0104 c 65 221 2T WD0104 D 6T 20 2T

UNOCAL SS020I

SS 020I SSl20I

SS 0209

SS0209

II p

PURll

A

B

3

n 61

103

97

97

95

100

34

34

34

34 34

bullbull bullbullbull

- SS0209 D u 95 34 I

- SS 1209 SSl253

SS0253

G A

c

6T

2

100

165

1T5

34

54

54

bullbull I

I I

I I

SOUTH MARCH

~LAND

N

SOUTH PaTOI ENE ISWltD AREA

MMS Defined Limits ror Offshore Structures Requiring Inspections Following Andrew

t---Jbulln3 T25bull I bullTtl

bull1$1Htl1Abull bull1SIK1779bull

~

middot-DaTA AREA~T

CRANO rsLE

AREA bull1300

bullll4W

=shy

bullPath of Hurricane Andrew with Platforms Used In Calibration

Likelihood Function vs b All Success Case STISI K STIJOQ ST134W WD90A MCJll MCJ97

I 0000

0 9000

0 8000

07000

-Bo6000

tiiii 0 5000 -

04000

0 3000

0 2000

01000

0 0000

0 05 25

ri1rt middot-middot-middot-middot-middot-

I 5 2

b

---11 middot151K(blsu~ccsI

- bull - bull - fk 130Q(bfU~lCSS)

lie bull134W(blsuccc]

lk-WD90A[blsultCS]

- - - lk middotMC397(bfUtlCS]

--o-- lk-MCJll(bbuctcsJ

--- ll (bl 6 out of 6 Mlttcs case]

_

( Likelihood Functions middotSuccess Cases - p - 1 c I

CATEGORIZATION OF ADDITIONAL PLATFORMS

I I I

bull CATEGORIES

Unexpected Survivals

Expected Survivals

bull APPROACH

Identify Corresponding Platform(s) Evaluated In the JIP

Determine Max Wave Height at Platform Location

Evaluate Whether Wave Hit the Deck

middot - Evaluate Likelihood of R and S to Differ from the Base Cases

Review Joint Information

Assess Special Features

bull IDENTIFY NUMBER OF PLATFORMS IN EACH CATEGORY

I

UPDATING OF CALIBRATION RESULTS

bull INCLUDE EFFECT OF ADDITIONAL PLATFORMS ON bullBbull

Consider Increased 1Uncertalntles In R and S

Develop bullNeW Ukellhood Functions for Different Categories

middotLet bullxbull Number of Ukely Unexpected Survival Cases and y Number of Ukely Expected Survival Cases

Determine Combined Likelihood Function lk (b Iadditional platforms) middot

bull REVISED LIKELIHOOD FUNCTION

lk(b I all cases) = lk(b I 9 platforms) middot x lk(b Ix platforms) x lk(b I y platforms)

bull ESTABLISH REVISED BIAS FACTOR B

f8 (b) u f8 (b) lk (b I all cases)

II

t t ~

shy

APPLICATIONS OF BIAS FACTORS

ii APPLICABLE TO PARTICULAR LOAD AND RESISTANCE RECIPE

bull USE BY THE API COMMITTEES AND MMS

Criteria Revisions

Recommended Design Guidelines Revisions bull

bullbull REASSESSMENT STUDIES AND DECISION MAKING

middot

bull RISK ASSESSMENT STUDIES BY THE OPERATORS

Individual Platforms

Fleet of Platforms

middot IMR Program Implementation

bullbull NOT TO BE USED IN DESIGN

I I fl J

EXAMPLE USE OF BIAS FACTORS

bull PHASE I EXAMPLE APPLICATION

bull EVALUATION OF ANNUAL PROBABILITY OF FAILURE

Establish Annual Maximum Seastate Data

Determine Coefficients to Define Base Shear

Determine Ultimate Capacity of Platform

Determine Probablllty of Failure Versus bullbbull

-bull DETERMINE UPDATED PROBABILITY OF FAILURE

Pf - P [S gt R]

Pf I P [~ gt Rb I b] f 8 (b) db

Pr - J P [S gt Rb I b] f bull 8 (b) db

bull TOTAL PROBABILITIES OF FAILURE

Total Pf - 1 bull Il [1 middotPf] - l Pf bull

Total Pf - 1 bull Il [1middot Pf] - l Pfmiddot

Total Pr = 1 - n [1middot PfbullJ = l Pfshy

I I I

EXAMPLE APPLICATION bull PHASE I

PLATFORM ST 151 K

bull GULF OF MEXICO

bull WATER DEPTH = 137 FT

bull ANNUAL MAXIMUM HS MEDIAN 175 ft COV 031

bull ASSUMED STORM DURATION 3 hoursmiddot

bull ASSUMED NUMBER OF WAVES 800

bull MEDIAN HS FOR STORM HOURS 1 2 AND 3

bull MEDIAN CURRENT VALUE bull

bull DETERMINE ULTIMATE CAPACITY FOR THREE DIRECTIONS

bull DETERMINE ANNUAL FAILURE PROBABILITIES

Broadsidemiddot Diagonal End-on

0037 0025 0007

0071 0044 0018

0022 0016 0004

DETERMINE TOTAL ANNUAL PROBABILITY OF FAILURE

Pf - 007

Pf - 013

Pr - 004

bull

EXAMPLE USE OF BIAS FACTORS - PHASE II middot

bull DETERMINE PREDICTIVE PROBABILITY OF FAILURE

Case - 1

Pf _ S S Pf (b b2) fa1a2 (b b2) db db2

g bull -1 (b1 R I S) - 1

92 - (b2 R2 I S) - 1

Case -2

Pf s ~ [bJ f11 (b) db

Pf [bJ p [g lt 0)

g - bull 1 (b1 R1 I S) - 1

1

bull1) l

I

BENEFITS OF ANDREW JIP bull PHASE II

lt

bull IMPROVED LOAD AND RESISTANCE RECIPE

bull IMPROVED CALIBRATION PROCEDURE

bull MULTIPLE BIAS FACTORS

bull PROCEDURE TO INCLUDE EFFECT OF OTHER PLATFORMS

bull PROCEDURE TO USE MULTIPLE BIAS FACTORS

  • Structure Bookmarks
    • ANDREW PHASE II JOINT1 INDUSTRY PROJECT
    • PROGRESS MEETING NUMBER 1
    • ANDREW PHASE II JOINT INDUSTRY PROJECT PROGRESS MEETING NUMBER 1 NOVEMBER 16 AGENDA
    • ANDREW PHASE II UST OF PARTICIPANTS
    • ANDREW PHASE I OBJECTIVES AND RESULTS
    • ANDREW PHASE II PROJECT OBJECTIVES
    • SUBSEQUENT PROJECTS
    • ANDREW PHASE II PROJECT TEAM
    • ANDREW PHASE II UPDATE MODELING middotAND ANALYSIS RECIPE
    • -ANDREW PHASE II CONDUCTOR RESISTANCE
    • ANDREW PHASE Il WA VE FORCE PROFILE
    • ANDREW PHASE II WA VE FORCE PROFILEmiddot
    • ANDREW PHASE II BRACE MODELING
    • ANDREW PHASE II FOUNDATIONS
    • ANDREW PHASE II middot middot JOINT MODELING
    • ANDREW PHASE Il JOINT MODELING OPTIONS DISCUSSION
    • ANDREW PHASE II JOINT MODEJJNG OPTIONS ILLUSTRATIONS
    • ANDREW PHASE II JOINT MODELING PHASE II METHOD
    • ANDREW PHASE II ANDREW STORM CONDmONS UPDATED
    • Platform Selection bullbullbull1111111111111
    • BasisCriteria for Selection bullbullbull1111111111111
    • CALIBRATION
    • WHERE TO INTRODUCE BIAS FACTOR (B
    • PHASE I -KEY CONCLUSIONS
    • bull CALIBRATION ISSUES PHASE II
    • PHASE II WORK SCOPE (CALIBRATION)
    • SOFTWARE
    • SENSITIVITY STUDIES POSTERIOR DISTRIBUTION OF B
    • WEIGHTING APPROACH bull
    • INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON
    • CATEGORIZATION OF ADDITIONAL PLATFORMS
    • UPDATING OF CALIBRATION RESULTS
    • APPLICATIONS OF BIAS FACTORS
    • EXAMPLE USE OF BIAS FACTORS
    • EXAMPLE APPLICATION bull PHASE I PLATFORM ST 151 K
    • EXAMPLE USE OF BIAS FACTORS -PHASE II middot
    • BENEFITS OF ANDREW JIP bull PHASE II
Page 43:  · . ' ANDREW PHASE II PROJECT OBJECTIVES . ' • ·UPDATE MODELING AND ANALYSIS RECIPE· JOINTS ' ' ·BRACES WAVE PROFILE FOUNDATIONS . ' ' CONDUCTORS· • ANALVZE PLATFORMS USING

BasisCriteria for Selection bullbullbull1111111111111

0 Benefit for Calibration - Classification (UnexpectedExpected)

- Failure Mode (StructuresFoundations)

- Damaged StructuresKnown response

- Better Representation of Full ~opulation

0 Availability of Data - Existing Models

- Inspection Data

- Damage Data

PREDICTED BEHAVIOR

-

OBSERVED

BEHAVIOR

-SURVIVED DAMAGED FAILED

SURVIVED

ST134W WD90A

-MC311 -

MC397 -

--

-

-

ST151K ST130Q

-

-

-

DAMAGED - -

-

-

--

-

-

ST52 SS139 ST177B -

ST161A

-_

-

FAILED

-

-

-

ST151H ST130A ST72

Operator - Platfonns - Platfonns Survived Damagedamp

- Falled -

bull

-

-

Jle eBDIGIPllDbl -Amocoshy 30 1 Chevron -

143 12 Exxon 72 Mobll 22 1 Phllllps 7 bull

-Shell 39 Unocal 25 1

Total Partlclnts Platfonns 338 15

Other Operators bull- 310 13

-

Total Platfonns - -

648 28

fbml

Platforms Considered 6 7

tie of Total Platfonns 1 25

OVERVIEW OF PLATFORMS SURVIVEDDAMAGEDFAILED

-

shy

v

Platforms Evaluated in Phase I

-

-Platfonn Water Year Number Analysis Avallabillty of Availability Platform

- Name Depth rl of Model DetaDed rl ClassUlcatlon

Installation legs Inspection Soll Report Based on Information Phase I Calibration - -ft

- -- Suaill flalfnrm ~BSi -STISIK 137 1963 8 CAP Yes Yes Un eel Sarvlval

-STIJOO 170 1964 4 CAPmiddotshy Yes ST134data u ST134W 137 1981 4 CAP Yes Yesmiddot E

-

edSarvlval -

middotsurvival

WD90A 184 - 1964 8 CAP Yes E -

- Sarvlvill

MC3ll 343 1978 8 KARMA

MC397 468 1991 4 KARMA -

bull ~reglmmiddot- -

Yes Sure Survival

bull Yes - Sare Sanlval-

-T23STS2 63 1969 4 CAP

T2SSS139 62 1969 4 CAP -

Yes

Yes

Yes Ex

Yes Une

bull SurvivalLikely to DamalI

middot bull SanlvalLlkely to DamalI

ST161A 118 1964 8 USFOS

Eail11a pound1al(bullllm Clss ST177B 142 --1965 8 CAP

STISIH 137 1964 - 8 CAP C

STIJOA 140 19S8 8 CAP - -

Yes

Yes

ollamed

Yes

- Yes E

-

at about 1 mUe

Yes

ST134data

bull SafvlvalLlkelv to Dam- bull- -

E middotrahre

Emeried Failure -

Likely FaUare

T21ST72 61 1969 4 CAP Collamed No Likely FaUure

Platforms Identified for Evaluating Their Likelihood of Falling Under UnexpeCted Survival Category

CllEllATOa

AllB

llLOCll

Sl1lvcnJllE

NAME

YEAR

INSTAUJCD

WATER

IJBPJB

middotmiddotDISTANCE

ftOMSllOllE

(MIUIS)

Ill

(M)

AMOCO BllW

STllI

WDI075

A

bull D

4

n uo 17J

44

32

bull 20

10

bull CllEVllOH STtllO c I 1bull 21 11

STl130 D 0 161 21 11 STllJ4 I 0 137 D - 11

STIUI G I 137 32 11

STIUI I 0 121 32 11

STIUI

J 0 140 32 11

STIUI It 4 137 32 11

STllSI PROl)I 0 137 32 11

STllSI PROD-2 0 137 32 11

STll7 A 0

140 35 11

WDl117 c 65 214 34 10

DXON STshy E 51 14 bull STl165 A t3 bull 34 WD0073

WD0073

WD0073

A _ c

4

62

4

IA IA

17J

22

22

22

bullbullbull WD003-shy E 65 161 Z7

A 0 IN 31 MOBD Bl12I

ssn A bull

50

50

31

7

bull 7

MIJllPllY SS114 B 7 50 7 7

PlllUIPS SSl14t A 55 33 7

SAMEDAN ST17J A 4 117 7 bull 7

SJIELL WD0103 A 65 223 27 WDllOJ bull 65 221 27 WDll04 c 65 221 27 WDOI04 D 67 2 27

UNOCAL

SSl20I

SSl20I

bull SSl20I

SS0209

SS0209

SS0209

SS0209

SS 0253

SS 0253

E

I

PtJllE

A

bull D

G A

c

0

4

4

7J

61

0

0

69

103

97

97

9S

100

95

100

165

175

34

34

34

34

34

34

34

54

54

bullbullbullbullbullbullbullbullbull

Platforms Recommended for Phase II

llalr y_ w- Aot Av- _ I 1 Doplla Model nm bull bull _ bull I I H11 s18-1 Irht lbwl~ bullbullbull(-illtdlel OM lllMtD

II ---

-STISIIC 137 1963 I CAP Yeo Yeo I Yeo

STtshy -170 1964 4 CAP Yeo ST134dlU I Yeo

STl34W 137 Yeo Yeo -1981 4 CAP D Yeo

WD90A 114 1964 I CAP Yeo - D

MC311 30 1971 I KARMA Yeo -sre-

MC397 middot 461 1991 4 ICARMA Yeo Sireshy - - r- -

__ T13ST5l 63 1969 4 CAP Yeo Yeobull middot- ~n- D Yeo _ Tl5SSl39 62 1969 4 CAP Yeo Yeo I Yeo

_ STIT711 142 1965 I CAP Yeo I Yeo middot-1shySTl61A Ill 1964 I USFOS Yeo Yeo - D bulloshy

- r-

STISIH 137 1964 I CAP y D y -_ STl31A 140 1951 I CAP Yeo STl34dlU I Yeo

T21Sll 61 1969 4 CAP No rar-middot I Yeo

~-rmiddot-middot SS209A 95 1967 16 CAP Yeo D -SSU4R

11 UI A 41 -in D SS 2741AIM CAP bull

_ _ -shy - -shy- -shy

CALIBRATION

bull PHASEIAPPROACH

bull CALIBRATION ISSUES

bull PHASE II WORK SCOPE

bull SOFlWARE AVAILABILITY

bull EFFECT OF ADDITIONAL PLATFORMS

bull WEIGHTING APPROACH

bull EXAMPLE APPLICATION middot

(l=Ofce Mode] Wind

~

fcurrent l -middot

~SW ~leldl4

ultimate resistance force I first

design load Illgt

i _

deflection

-

bull

4

Historical Hurrlcane Dariiage and Survival of Platforms

Components Required for Calibration (Petrauskas 1992)

PRIOR Distribution

PLATFORM EXPERIENCE

BAYESIAN Survivals and Failures

UPDATING

Waves Winds and Currents bull

POSTERIOR Distribution

- A_middotJA ~ Survival Data Failure Data

Result Result

Load Resistance

middotfl

-

Bayesian Updaing middotGeneral Description (Petrauskas 1992)

L Imiddot

WHERE TO INTRODUCE BIAS FACTOR (B

On Resistance (Rmue - B (RPREDICreD I

middotOn Loads (Smue middot - B (SPRE01cTEo

On Safety Margin (RSmue - B (RSPREDICTED

CALIBRATION METHODOLOGY

gt

r

FIELD OBSERVATIONS

Observed Behavior

survivals Damages

Failures

r

bull

CAPACITY ANALYSIS

Predicted Load and Resistance Predicted Failure Modes

Survival Damage Failure cases

RELIABILITY ANALYSIS

- Probability of Failure ~

Ukellhood Failure

BAYESIAN UPDATING

Prior Distnbution of B Combined Likelihood Function

Posterior Dlstnbution of B

BIAS FACTOR

Mean Value cov bull

Represents Modeling Uncertainties

Prior Distribution of B Likelihood Functionmiddot Observing Success case STISIK and Posterior Distribution of B

2 00

uo

160

140

120

e 100 L $ uoIshy

0 60 ~ ~ 040

020

0 00

0 00 0 50

_ I 00

b

I 50 2 00

-shy lk [bl STISIK)

--shy Pnor-f(b)

--shy Pos1enor r(bl ISIK)

Prior or B Mean bull 10 COV bull 030

Posterlor of Bbull Meanbull 136COV 0 16

2 50

Likelihood Function and Posterior Distribution Example Platform STISIK

- -

~

- Prior and Posterior Distributions or B Failure cases (ST177B STlSlH ST130A T21) Damage cases (T23 T25 ST161A) Success cases (STlSlK ST130Q ST134W WD90A MC311 MC397)

400

350

300

2 so 0-

I

bull bull

bull bull gt

- bullbull ~ bull bull I bull

_ bull bull bullbull I bull bull bullII V -

bull bull

I bull ~ bull

bullbull bullI bull bullI bull bull bull

middot -~

bull__ ~ middotI middotmiddot

L 2 00

e ~ I SO

-I 00

- 0 so

000 gt

0 00 020 0 40 060 0 80 I 00 120 140 160 110 200

b -

-

shy

Prior-r

- - - - Posterior fLlll failure cases

bull bull bull bull bull bull bull Posterior flall damage cases

- bull bull - bull middot Postenor flall success cases

Posterior flall success fulure damage cases

-

Posterior of B Mean= 119 COV 0101

Posterior Distribution or Blas Factor (8) bull Different Categories and All Cases Combined

shy

l

PHASE I - KEY CONCLUSIONS

bull GLOBAL CORRECTION FACTOR

bull 15 TO 20 CONSERVATISM MEAN)

bull UNEXPECTED SURVIVALS HAVE GREATEST IMPACT

bull DAMAGED CASES MORE INFORMATIVE

bull CALIBRATION ISSUES PHASE II

bull

EFFECT OF IMPROVED CAPACITY ANALYSIS RECIPE

bull APPLICABILITY TO MULTIPLE FAILURE MODES

bull NO FOUNDATION DAMAGE CASE INCLUDED

bull IMPROVED REPRESENTATION OF FULL ANDREW DISTRIBUTION

-bull NUMBER OF PLATFORMS CONSIDERED

I 1 I

I t ) bull ~

PHASE II WORK SCOPE (CALIBRATION)

bull

bull ESTABLISH MULTIPLE BIAS FACTORS

Use Nine Base Case Platforms

middot - Capacity Analysis for Revised recipe

Foundation Blas Factor from Foundation JIP

bull DEVELOP WEIGHTING APPROACH

bull UPDATE MULTIPLE BIAS FACTORS

bullEffect of Platforms Not In the Nine Base Cases

bull EXAMPLE APPLICATION OF BIAS FACTORS I

bullbull

PHASE I PHASE II

CAPACITYbull ANALYSISRu bull

---------------------1----middot

Jc1 - iJbullraquoJ dbull

FttlS frt[f~l ratdhJJ

-

LIKELIHOOD FUNCTION

DEVELOPMENT

lk Jb Jfall1N) bull Jb)

111 Ill Isurvtvoq bull 1 - bl

-middot -------------------shy ~---------------------middot-middot

r111ir8 Cb1 111 Cbl BAYESIAN

UPDATING

~cove

R1R2 bull middot---------------------~----

Jblbull P(g cOJ

1 bull JbRI (MBSI I bull 1

~middot

llcJllJfalluro) bull P(gcOJ Ill 111 b21fallun) bull P ((g1 CO) U 1112 c OD

middot--------------------shy ----------------------shy bull

r111111r8 Cbl 111 Ill i-1nro1

deg ~1b2)laquottfrr (b1b21-lnfo) re Ill bull 1-11middot b2) bull lodegamp

middot

i1 covamp i2 cov~i

-

shy

--

Likelihood Function vs b All Success Case STISI K STl30Q STl34W WD90A MC311 MC397

I 0000

0 9000

OIOOO

0 7000

~ 0amp000

ta osoOO

~ 0 4000

0 3000

0 2000

0 1000

bull 0 0000

bull

I

middot

middot bull I

Ibull

-sibull ~ -middot-middot-middot-middot -middot-

----7----+-----~ I

Ibull bull bull I

rbull

--- Jk ISIK[blsuccess)

- bull - bull - lk 130Q[blsuccess) bull

lk 134W[blsuccess)

Jk-WD90A(blsuccc~I

- - - lk MC397[blsucccss)

--ltgt-- lk middotMC31 l[blsucccss)

--- Jk [bl 6 out of6succcsHascs)

05 I 5 2 250

b

Likelihood Functionsmiddot Success Cases - Phbullsc L

shy

bullbull bull bull bull bull bull -------------------~

bull bull bullbull bullbull bullbull bullbull bullbull bullbull bull

bull bullbull

bullbullbullbull

bullbull

bullbullbullbullbullbullbull

bullbullbullbullbullbull

bullbullbullbullbullbullbull

z 0

t z ) LL

c 0 0c i ~ i

J

~ sect ~--------------~ e l5

I

I

SOFTWARE

bull PF PROGRAM

bull CALREL

bull PF +CALREL

bull RELACS

bull

gt i

l

SENSITIVITY STUDIES

POSTERIOR DISTRIBUTION OF B

bull EFFECT OF CHANGE IN DISTRIBUTION OF PRIOR OF B

bull EFFECT OF REMOVAL OF ONE PLATFORM FROM EACH CATEGORY

bull EFFECT OF REMOVAL OF THREE DAMAGE CASES

bull EFFECT OF ADDITIONAL 13 SIMILAR PLATFORMS

-I

Prior and Posterior Distributions of B Variation In COV of PriorofB fromOl to04 -- -

4500

-4000

-

- -

3500 --middot - bull bull - bull bull Prior-r ICOV =02

3000

- - - - Prior-r ICOV =03 -

bull -~ e 2soo -

bull bull bull bull bull bull bull Pnor-r ICOV =o 4t frJ ~ 2000 Posterior rlCOV =0 2

bull bull Postenor rlCOV =0 3 bull bull 1500 bull middot bull - shy -middot- Posterior rlCOV =04

middot~ I 000 - bull

-middotmiddot ~ ~middotmiddot M -~

- middot 0 500

wbullbullbull _ I middot~ middotmiddotmiddotmiddotmiddot middotmiddotmiddotmiddotmiddot - -~ ~ __ middot_ _ bull -shy0 000

0 000 0200 0 400 0 600 0 800 IOOO 1200 1400 1600 1800 2000 -

- - b

-

=shy

Posterior Distribution of Bias Factor (B) bull Effect of Variation In COV of Prior of B

bull

bullPrior and Posterior Distributions of B Evaluating Effect of Removal of One Platform from Each Category from the Base Case of 13 Platforms

400

~ ~ 350

j IA i

3 00 bull I bull I ---Prior-r I bull l

250 1 I 1

I bull bull ---Posterior r lall 13 casesshy- ~E 2 00 bull bull bull bull bull bull bull Posterior r without STISIK

- ~ 1 bull - bull I - bull bull - bull bull Posterior r Iwithout T2I I 50 I

J-tt~i - bull - bull - Posterior r Iwithout T2S

tOO [-

I I ~ ~ I 0 50

0 00

___ r J i

l~T

~--

0 00 0 20 0 40 0 60 080 I 00 120 140 160 110 2 00

b

zshy

Posterior Distribution of Bias Factor (8) bull Effect of Removal of middotPlatform from Any Category

bullPrior and Posterior Distributions or B Evaluating Effect or Removal or 3 Damage Cases from the Base Case or 13 Platrorms

4

35

3

25

e L 2 ~ bulle shy

I 15

bull I

05

0

0 02 04 06 I 2 I 4 16 IS 2

I I

I

I

I Ibull

J -tmiddot L--

bullI

~

I

fl

II ~

I I v I

middot bull bull

bull L

~ ll- I-~lt

OS I

b

I

bull

1middot bullbullbullbull

~ __

bull

- - - - Prior-r

- - - - Postenor flall failure cases

Postenor flall success cases

--- Posterior flall 13 cases

---Posterior fI success failure cases only

IPosterior or B Mean =t11 COV =0121

-

_I

Posterir Distribution or Bias Factor (B) bull Effect or Removal or 3-Damage Cases

-

bullPrior and Posterior Distributions or B Evaluating Effect or Additional 13 Similar Observations as

Used in Andrew JIP The POsterior or B from Andrew-JIP Considered as New Prior or B

- 500

450

400

350

bull bull

bull bull bull bull J bull

bull bull f ~ I

I ~ Ibull bull

bullbull bull

l bull I -bullbull bull bull

I

I bull bull

bull I I~~bull bull bull

bull ~

I bull bull

~ ~_ ~~~

- - New Prior r bull

Posterior 114 new failure cases ~ 300- bull bull bull bull bull bull bull Posterior f 13 new damage casesl 150 ~

- bull bull - bull bull Posterior 116 new success cases i 2 00

Postenor 1113 new success failure -150 damage cases

I 00

0 50 -

0 00 0 00 0 20 0 40 0 60 0 80 I 00 120 I 40 160 110 2 00

Posterior ol B Mean= 120 COV= 0071 b

shy

shy

_rshy

Posterior Distribution of Bias Factor (B) bull Effect of Additional 13-Similar Observations

1

1 I

WEIGHTING APPROACH bull

bull ASSESSMENT OF PLATFORMS DATA

Select Platforms Most Mappable to the Analyzed Platforms

Unlikely SurvivalDamage Cases Identification

Based on Cut-Off Criteria Agemiddot Hs etc

bull INFORMATION REQUIRED FOR COMPARISON

bull CATEGORIZATION OF ADDITIONALmiddot PLATFORMS

bull UPDATING OF CALIBRATION RESULTS

--

OVERVIEW OF PLATFORMS SURVIVEDDAMAGED -~

Operator -

- Platforms

Survived CS)

Platforms

Damaged CD)

Platforms Used In

Phase IJIP -

JI~ ~Bill~IRBDll -

Amoco -

Chevron -

Exxon

Mob II

Phllllps

Shell

Unocal -

30

143

72 22

7

39 25shy -

1

12

bull

1 bull

1

1(S)+1 (D)

3 (S) +3(D)

2 (S)

bull shy

bull

bull

bull

Total Partlclnants Platforms 338 - 15 10C6S+4D) -

Other Operators

310

13

3 (D)

Total Platforms 648 28 13C6S+7D) -

Summary of Andrew JIP bull Phase I Platforms

Plairshy w v- N-ber Lea-le 1 shy car Hind_ o-w N-bull Deplh fl fl IJpe la bull n Yeadon Jadtet lbelledtf liefalit -- _ -OD

IL H - -shy _ICllshy IL IL - - rQClto ~ _

K STUIK 137 1963 8 365 608S4 Cellardeck

-middot STl-shy 170 1964 4 -~ K 39 62269 Cellardeck - --ST134W 137 1911 4 43 608S4 Smvlnlshy- _ - -

WD90A 184 1964 8 K 267Sbull34 83 SOS31 SUboftllar deck SmvMI

MC311 343 1978 8 x S7 62 - MC397 468 1991 4 x shy S33bull6SO 65911 -shySmeSmvlvol -- gt -

r-- T23ST52 63 1969 4 ton- shyKshy 52594 S03 -T25SSl39 62 1969 4 Kshy SOS94 S063 - middot ton-shy__ -

bullbull a ton--shyST161A 118 1964 8 K - 347S 51495 Cellarclecti

r~

ST1771 142 1965 8 K 39 60167 CeDordeck failme

STISIH 137 1964 8 K 395 608S4 CeDardeck Failme

K ST130A 140 1958 a SS 6095 I lblv Pai1uno - T21ST72 61 1969 4 Kshy S4S94 4974 I lhlv Pai1uno -

- -

JE

INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON

bull WATER DEPTH

bull bull YEAR OF INSTALLATIONDESIGN

bull PLATFORM ORIENTATION

bull PLATFORM LATITUDE AND LONGITUDE

bull KEY SKETCHES OF PLATFORM STRUCTURAL FRAMING

Number of Legs Plies Leg Batter Number of Conductors bull Deck Elevations (Including sub-cellar deck If any) Jacket Framing Elevations Brace type In Vertical Frames Sizes of Primary Members (leg pile vertical frames braces primary horizontals) Joint Details (can gap etc) LegPiie Annulus Grouting Status

bull middotPOST ANDREW INSPECTION REPORT Significant Wave Height Hs (ft)

bull INFORMATION ON DAMAGES PRIOR TO ANDREW

bull RESULTS FROM IN-HOUSE STUDIES IF AVAILABLE

bull ~RIGINAL DESIGN CRITERIA INFORMATION IF AVAILABLE

bull SOIL SHEAR STRENGTH PROFILE IF AVAILABLE

Platforms Identified for Evaluating Therr Likelihood of Falling Under Unexpected Survival Category

OPlRATOR

middotADA

BLOClt

llmucnJllE

NAME

YEAll

INSTAUJD

WA1ER

DBPTll

DISTANCE

JlllOM SHOllE

(MILU)

Bl

(M)

AMOCO 111113

STl11

WDI075

A

bull D

St

17

120

1n

44

20

10 I

CllllVllQfl STlllO c 61 110 21 11 STIUO D Q 1110 21 11

STl134 p Q 137 29 11

ST0151 G 61 137 11

ST0151 I 3 121 11

ST0151 J Q 140 11

STl151 It 137 11

ST0151 PJl001 Q UT 11

ST0151 PROD-2 Q UT 11

ST017 A 3 140 35 11

WDIUT c 65 214 34 10

llXXON STI055 II 51 6T 14 I

STl15 A t3 34

WDI073

WD0073

Amiddotshy Q

1111 1111

22

22

I

I

WDI073 c 1n 22 I

- WDOCl93-shy II

65 1110 2T A 3 194 31 MOBU 111121

SS IOl7

A B

501

50

50

31

T

I

T 6

MtlllPllY SS114 B T 50 T T

PlllllIPS SS0149 A 5 55 33 T

SAMllDAN STln A 107 T T

SBllLL WDl103 A 65 223 2T WDl103 B 65 221 2T WD0104 c 65 221 2T WD0104 D 6T 20 2T

UNOCAL SS020I

SS 020I SSl20I

SS 0209

SS0209

II p

PURll

A

B

3

n 61

103

97

97

95

100

34

34

34

34 34

bullbull bullbullbull

- SS0209 D u 95 34 I

- SS 1209 SSl253

SS0253

G A

c

6T

2

100

165

1T5

34

54

54

bullbull I

I I

I I

SOUTH MARCH

~LAND

N

SOUTH PaTOI ENE ISWltD AREA

MMS Defined Limits ror Offshore Structures Requiring Inspections Following Andrew

t---Jbulln3 T25bull I bullTtl

bull1$1Htl1Abull bull1SIK1779bull

~

middot-DaTA AREA~T

CRANO rsLE

AREA bull1300

bullll4W

=shy

bullPath of Hurricane Andrew with Platforms Used In Calibration

Likelihood Function vs b All Success Case STISI K STIJOQ ST134W WD90A MCJll MCJ97

I 0000

0 9000

0 8000

07000

-Bo6000

tiiii 0 5000 -

04000

0 3000

0 2000

01000

0 0000

0 05 25

ri1rt middot-middot-middot-middot-middot-

I 5 2

b

---11 middot151K(blsu~ccsI

- bull - bull - fk 130Q(bfU~lCSS)

lie bull134W(blsuccc]

lk-WD90A[blsultCS]

- - - lk middotMC397(bfUtlCS]

--o-- lk-MCJll(bbuctcsJ

--- ll (bl 6 out of 6 Mlttcs case]

_

( Likelihood Functions middotSuccess Cases - p - 1 c I

CATEGORIZATION OF ADDITIONAL PLATFORMS

I I I

bull CATEGORIES

Unexpected Survivals

Expected Survivals

bull APPROACH

Identify Corresponding Platform(s) Evaluated In the JIP

Determine Max Wave Height at Platform Location

Evaluate Whether Wave Hit the Deck

middot - Evaluate Likelihood of R and S to Differ from the Base Cases

Review Joint Information

Assess Special Features

bull IDENTIFY NUMBER OF PLATFORMS IN EACH CATEGORY

I

UPDATING OF CALIBRATION RESULTS

bull INCLUDE EFFECT OF ADDITIONAL PLATFORMS ON bullBbull

Consider Increased 1Uncertalntles In R and S

Develop bullNeW Ukellhood Functions for Different Categories

middotLet bullxbull Number of Ukely Unexpected Survival Cases and y Number of Ukely Expected Survival Cases

Determine Combined Likelihood Function lk (b Iadditional platforms) middot

bull REVISED LIKELIHOOD FUNCTION

lk(b I all cases) = lk(b I 9 platforms) middot x lk(b Ix platforms) x lk(b I y platforms)

bull ESTABLISH REVISED BIAS FACTOR B

f8 (b) u f8 (b) lk (b I all cases)

II

t t ~

shy

APPLICATIONS OF BIAS FACTORS

ii APPLICABLE TO PARTICULAR LOAD AND RESISTANCE RECIPE

bull USE BY THE API COMMITTEES AND MMS

Criteria Revisions

Recommended Design Guidelines Revisions bull

bullbull REASSESSMENT STUDIES AND DECISION MAKING

middot

bull RISK ASSESSMENT STUDIES BY THE OPERATORS

Individual Platforms

Fleet of Platforms

middot IMR Program Implementation

bullbull NOT TO BE USED IN DESIGN

I I fl J

EXAMPLE USE OF BIAS FACTORS

bull PHASE I EXAMPLE APPLICATION

bull EVALUATION OF ANNUAL PROBABILITY OF FAILURE

Establish Annual Maximum Seastate Data

Determine Coefficients to Define Base Shear

Determine Ultimate Capacity of Platform

Determine Probablllty of Failure Versus bullbbull

-bull DETERMINE UPDATED PROBABILITY OF FAILURE

Pf - P [S gt R]

Pf I P [~ gt Rb I b] f 8 (b) db

Pr - J P [S gt Rb I b] f bull 8 (b) db

bull TOTAL PROBABILITIES OF FAILURE

Total Pf - 1 bull Il [1 middotPf] - l Pf bull

Total Pf - 1 bull Il [1middot Pf] - l Pfmiddot

Total Pr = 1 - n [1middot PfbullJ = l Pfshy

I I I

EXAMPLE APPLICATION bull PHASE I

PLATFORM ST 151 K

bull GULF OF MEXICO

bull WATER DEPTH = 137 FT

bull ANNUAL MAXIMUM HS MEDIAN 175 ft COV 031

bull ASSUMED STORM DURATION 3 hoursmiddot

bull ASSUMED NUMBER OF WAVES 800

bull MEDIAN HS FOR STORM HOURS 1 2 AND 3

bull MEDIAN CURRENT VALUE bull

bull DETERMINE ULTIMATE CAPACITY FOR THREE DIRECTIONS

bull DETERMINE ANNUAL FAILURE PROBABILITIES

Broadsidemiddot Diagonal End-on

0037 0025 0007

0071 0044 0018

0022 0016 0004

DETERMINE TOTAL ANNUAL PROBABILITY OF FAILURE

Pf - 007

Pf - 013

Pr - 004

bull

EXAMPLE USE OF BIAS FACTORS - PHASE II middot

bull DETERMINE PREDICTIVE PROBABILITY OF FAILURE

Case - 1

Pf _ S S Pf (b b2) fa1a2 (b b2) db db2

g bull -1 (b1 R I S) - 1

92 - (b2 R2 I S) - 1

Case -2

Pf s ~ [bJ f11 (b) db

Pf [bJ p [g lt 0)

g - bull 1 (b1 R1 I S) - 1

1

bull1) l

I

BENEFITS OF ANDREW JIP bull PHASE II

lt

bull IMPROVED LOAD AND RESISTANCE RECIPE

bull IMPROVED CALIBRATION PROCEDURE

bull MULTIPLE BIAS FACTORS

bull PROCEDURE TO INCLUDE EFFECT OF OTHER PLATFORMS

bull PROCEDURE TO USE MULTIPLE BIAS FACTORS

  • Structure Bookmarks
    • ANDREW PHASE II JOINT1 INDUSTRY PROJECT
    • PROGRESS MEETING NUMBER 1
    • ANDREW PHASE II JOINT INDUSTRY PROJECT PROGRESS MEETING NUMBER 1 NOVEMBER 16 AGENDA
    • ANDREW PHASE II UST OF PARTICIPANTS
    • ANDREW PHASE I OBJECTIVES AND RESULTS
    • ANDREW PHASE II PROJECT OBJECTIVES
    • SUBSEQUENT PROJECTS
    • ANDREW PHASE II PROJECT TEAM
    • ANDREW PHASE II UPDATE MODELING middotAND ANALYSIS RECIPE
    • -ANDREW PHASE II CONDUCTOR RESISTANCE
    • ANDREW PHASE Il WA VE FORCE PROFILE
    • ANDREW PHASE II WA VE FORCE PROFILEmiddot
    • ANDREW PHASE II BRACE MODELING
    • ANDREW PHASE II FOUNDATIONS
    • ANDREW PHASE II middot middot JOINT MODELING
    • ANDREW PHASE Il JOINT MODELING OPTIONS DISCUSSION
    • ANDREW PHASE II JOINT MODEJJNG OPTIONS ILLUSTRATIONS
    • ANDREW PHASE II JOINT MODELING PHASE II METHOD
    • ANDREW PHASE II ANDREW STORM CONDmONS UPDATED
    • Platform Selection bullbullbull1111111111111
    • BasisCriteria for Selection bullbullbull1111111111111
    • CALIBRATION
    • WHERE TO INTRODUCE BIAS FACTOR (B
    • PHASE I -KEY CONCLUSIONS
    • bull CALIBRATION ISSUES PHASE II
    • PHASE II WORK SCOPE (CALIBRATION)
    • SOFTWARE
    • SENSITIVITY STUDIES POSTERIOR DISTRIBUTION OF B
    • WEIGHTING APPROACH bull
    • INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON
    • CATEGORIZATION OF ADDITIONAL PLATFORMS
    • UPDATING OF CALIBRATION RESULTS
    • APPLICATIONS OF BIAS FACTORS
    • EXAMPLE USE OF BIAS FACTORS
    • EXAMPLE APPLICATION bull PHASE I PLATFORM ST 151 K
    • EXAMPLE USE OF BIAS FACTORS -PHASE II middot
    • BENEFITS OF ANDREW JIP bull PHASE II
Page 44:  · . ' ANDREW PHASE II PROJECT OBJECTIVES . ' • ·UPDATE MODELING AND ANALYSIS RECIPE· JOINTS ' ' ·BRACES WAVE PROFILE FOUNDATIONS . ' ' CONDUCTORS· • ANALVZE PLATFORMS USING

PREDICTED BEHAVIOR

-

OBSERVED

BEHAVIOR

-SURVIVED DAMAGED FAILED

SURVIVED

ST134W WD90A

-MC311 -

MC397 -

--

-

-

ST151K ST130Q

-

-

-

DAMAGED - -

-

-

--

-

-

ST52 SS139 ST177B -

ST161A

-_

-

FAILED

-

-

-

ST151H ST130A ST72

Operator - Platfonns - Platfonns Survived Damagedamp

- Falled -

bull

-

-

Jle eBDIGIPllDbl -Amocoshy 30 1 Chevron -

143 12 Exxon 72 Mobll 22 1 Phllllps 7 bull

-Shell 39 Unocal 25 1

Total Partlclnts Platfonns 338 15

Other Operators bull- 310 13

-

Total Platfonns - -

648 28

fbml

Platforms Considered 6 7

tie of Total Platfonns 1 25

OVERVIEW OF PLATFORMS SURVIVEDDAMAGEDFAILED

-

shy

v

Platforms Evaluated in Phase I

-

-Platfonn Water Year Number Analysis Avallabillty of Availability Platform

- Name Depth rl of Model DetaDed rl ClassUlcatlon

Installation legs Inspection Soll Report Based on Information Phase I Calibration - -ft

- -- Suaill flalfnrm ~BSi -STISIK 137 1963 8 CAP Yes Yes Un eel Sarvlval

-STIJOO 170 1964 4 CAPmiddotshy Yes ST134data u ST134W 137 1981 4 CAP Yes Yesmiddot E

-

edSarvlval -

middotsurvival

WD90A 184 - 1964 8 CAP Yes E -

- Sarvlvill

MC3ll 343 1978 8 KARMA

MC397 468 1991 4 KARMA -

bull ~reglmmiddot- -

Yes Sure Survival

bull Yes - Sare Sanlval-

-T23STS2 63 1969 4 CAP

T2SSS139 62 1969 4 CAP -

Yes

Yes

Yes Ex

Yes Une

bull SurvivalLikely to DamalI

middot bull SanlvalLlkely to DamalI

ST161A 118 1964 8 USFOS

Eail11a pound1al(bullllm Clss ST177B 142 --1965 8 CAP

STISIH 137 1964 - 8 CAP C

STIJOA 140 19S8 8 CAP - -

Yes

Yes

ollamed

Yes

- Yes E

-

at about 1 mUe

Yes

ST134data

bull SafvlvalLlkelv to Dam- bull- -

E middotrahre

Emeried Failure -

Likely FaUare

T21ST72 61 1969 4 CAP Collamed No Likely FaUure

Platforms Identified for Evaluating Their Likelihood of Falling Under UnexpeCted Survival Category

CllEllATOa

AllB

llLOCll

Sl1lvcnJllE

NAME

YEAR

INSTAUJCD

WATER

IJBPJB

middotmiddotDISTANCE

ftOMSllOllE

(MIUIS)

Ill

(M)

AMOCO BllW

STllI

WDI075

A

bull D

4

n uo 17J

44

32

bull 20

10

bull CllEVllOH STtllO c I 1bull 21 11

STl130 D 0 161 21 11 STllJ4 I 0 137 D - 11

STIUI G I 137 32 11

STIUI I 0 121 32 11

STIUI

J 0 140 32 11

STIUI It 4 137 32 11

STllSI PROl)I 0 137 32 11

STllSI PROD-2 0 137 32 11

STll7 A 0

140 35 11

WDl117 c 65 214 34 10

DXON STshy E 51 14 bull STl165 A t3 bull 34 WD0073

WD0073

WD0073

A _ c

4

62

4

IA IA

17J

22

22

22

bullbullbull WD003-shy E 65 161 Z7

A 0 IN 31 MOBD Bl12I

ssn A bull

50

50

31

7

bull 7

MIJllPllY SS114 B 7 50 7 7

PlllUIPS SSl14t A 55 33 7

SAMEDAN ST17J A 4 117 7 bull 7

SJIELL WD0103 A 65 223 27 WDllOJ bull 65 221 27 WDll04 c 65 221 27 WDOI04 D 67 2 27

UNOCAL

SSl20I

SSl20I

bull SSl20I

SS0209

SS0209

SS0209

SS0209

SS 0253

SS 0253

E

I

PtJllE

A

bull D

G A

c

0

4

4

7J

61

0

0

69

103

97

97

9S

100

95

100

165

175

34

34

34

34

34

34

34

54

54

bullbullbullbullbullbullbullbullbull

Platforms Recommended for Phase II

llalr y_ w- Aot Av- _ I 1 Doplla Model nm bull bull _ bull I I H11 s18-1 Irht lbwl~ bullbullbull(-illtdlel OM lllMtD

II ---

-STISIIC 137 1963 I CAP Yeo Yeo I Yeo

STtshy -170 1964 4 CAP Yeo ST134dlU I Yeo

STl34W 137 Yeo Yeo -1981 4 CAP D Yeo

WD90A 114 1964 I CAP Yeo - D

MC311 30 1971 I KARMA Yeo -sre-

MC397 middot 461 1991 4 ICARMA Yeo Sireshy - - r- -

__ T13ST5l 63 1969 4 CAP Yeo Yeobull middot- ~n- D Yeo _ Tl5SSl39 62 1969 4 CAP Yeo Yeo I Yeo

_ STIT711 142 1965 I CAP Yeo I Yeo middot-1shySTl61A Ill 1964 I USFOS Yeo Yeo - D bulloshy

- r-

STISIH 137 1964 I CAP y D y -_ STl31A 140 1951 I CAP Yeo STl34dlU I Yeo

T21Sll 61 1969 4 CAP No rar-middot I Yeo

~-rmiddot-middot SS209A 95 1967 16 CAP Yeo D -SSU4R

11 UI A 41 -in D SS 2741AIM CAP bull

_ _ -shy - -shy- -shy

CALIBRATION

bull PHASEIAPPROACH

bull CALIBRATION ISSUES

bull PHASE II WORK SCOPE

bull SOFlWARE AVAILABILITY

bull EFFECT OF ADDITIONAL PLATFORMS

bull WEIGHTING APPROACH

bull EXAMPLE APPLICATION middot

(l=Ofce Mode] Wind

~

fcurrent l -middot

~SW ~leldl4

ultimate resistance force I first

design load Illgt

i _

deflection

-

bull

4

Historical Hurrlcane Dariiage and Survival of Platforms

Components Required for Calibration (Petrauskas 1992)

PRIOR Distribution

PLATFORM EXPERIENCE

BAYESIAN Survivals and Failures

UPDATING

Waves Winds and Currents bull

POSTERIOR Distribution

- A_middotJA ~ Survival Data Failure Data

Result Result

Load Resistance

middotfl

-

Bayesian Updaing middotGeneral Description (Petrauskas 1992)

L Imiddot

WHERE TO INTRODUCE BIAS FACTOR (B

On Resistance (Rmue - B (RPREDICreD I

middotOn Loads (Smue middot - B (SPRE01cTEo

On Safety Margin (RSmue - B (RSPREDICTED

CALIBRATION METHODOLOGY

gt

r

FIELD OBSERVATIONS

Observed Behavior

survivals Damages

Failures

r

bull

CAPACITY ANALYSIS

Predicted Load and Resistance Predicted Failure Modes

Survival Damage Failure cases

RELIABILITY ANALYSIS

- Probability of Failure ~

Ukellhood Failure

BAYESIAN UPDATING

Prior Distnbution of B Combined Likelihood Function

Posterior Dlstnbution of B

BIAS FACTOR

Mean Value cov bull

Represents Modeling Uncertainties

Prior Distribution of B Likelihood Functionmiddot Observing Success case STISIK and Posterior Distribution of B

2 00

uo

160

140

120

e 100 L $ uoIshy

0 60 ~ ~ 040

020

0 00

0 00 0 50

_ I 00

b

I 50 2 00

-shy lk [bl STISIK)

--shy Pnor-f(b)

--shy Pos1enor r(bl ISIK)

Prior or B Mean bull 10 COV bull 030

Posterlor of Bbull Meanbull 136COV 0 16

2 50

Likelihood Function and Posterior Distribution Example Platform STISIK

- -

~

- Prior and Posterior Distributions or B Failure cases (ST177B STlSlH ST130A T21) Damage cases (T23 T25 ST161A) Success cases (STlSlK ST130Q ST134W WD90A MC311 MC397)

400

350

300

2 so 0-

I

bull bull

bull bull gt

- bullbull ~ bull bull I bull

_ bull bull bullbull I bull bull bullII V -

bull bull

I bull ~ bull

bullbull bullI bull bullI bull bull bull

middot -~

bull__ ~ middotI middotmiddot

L 2 00

e ~ I SO

-I 00

- 0 so

000 gt

0 00 020 0 40 060 0 80 I 00 120 140 160 110 200

b -

-

shy

Prior-r

- - - - Posterior fLlll failure cases

bull bull bull bull bull bull bull Posterior flall damage cases

- bull bull - bull middot Postenor flall success cases

Posterior flall success fulure damage cases

-

Posterior of B Mean= 119 COV 0101

Posterior Distribution or Blas Factor (8) bull Different Categories and All Cases Combined

shy

l

PHASE I - KEY CONCLUSIONS

bull GLOBAL CORRECTION FACTOR

bull 15 TO 20 CONSERVATISM MEAN)

bull UNEXPECTED SURVIVALS HAVE GREATEST IMPACT

bull DAMAGED CASES MORE INFORMATIVE

bull CALIBRATION ISSUES PHASE II

bull

EFFECT OF IMPROVED CAPACITY ANALYSIS RECIPE

bull APPLICABILITY TO MULTIPLE FAILURE MODES

bull NO FOUNDATION DAMAGE CASE INCLUDED

bull IMPROVED REPRESENTATION OF FULL ANDREW DISTRIBUTION

-bull NUMBER OF PLATFORMS CONSIDERED

I 1 I

I t ) bull ~

PHASE II WORK SCOPE (CALIBRATION)

bull

bull ESTABLISH MULTIPLE BIAS FACTORS

Use Nine Base Case Platforms

middot - Capacity Analysis for Revised recipe

Foundation Blas Factor from Foundation JIP

bull DEVELOP WEIGHTING APPROACH

bull UPDATE MULTIPLE BIAS FACTORS

bullEffect of Platforms Not In the Nine Base Cases

bull EXAMPLE APPLICATION OF BIAS FACTORS I

bullbull

PHASE I PHASE II

CAPACITYbull ANALYSISRu bull

---------------------1----middot

Jc1 - iJbullraquoJ dbull

FttlS frt[f~l ratdhJJ

-

LIKELIHOOD FUNCTION

DEVELOPMENT

lk Jb Jfall1N) bull Jb)

111 Ill Isurvtvoq bull 1 - bl

-middot -------------------shy ~---------------------middot-middot

r111ir8 Cb1 111 Cbl BAYESIAN

UPDATING

~cove

R1R2 bull middot---------------------~----

Jblbull P(g cOJ

1 bull JbRI (MBSI I bull 1

~middot

llcJllJfalluro) bull P(gcOJ Ill 111 b21fallun) bull P ((g1 CO) U 1112 c OD

middot--------------------shy ----------------------shy bull

r111111r8 Cbl 111 Ill i-1nro1

deg ~1b2)laquottfrr (b1b21-lnfo) re Ill bull 1-11middot b2) bull lodegamp

middot

i1 covamp i2 cov~i

-

shy

--

Likelihood Function vs b All Success Case STISI K STl30Q STl34W WD90A MC311 MC397

I 0000

0 9000

OIOOO

0 7000

~ 0amp000

ta osoOO

~ 0 4000

0 3000

0 2000

0 1000

bull 0 0000

bull

I

middot

middot bull I

Ibull

-sibull ~ -middot-middot-middot-middot -middot-

----7----+-----~ I

Ibull bull bull I

rbull

--- Jk ISIK[blsuccess)

- bull - bull - lk 130Q[blsuccess) bull

lk 134W[blsuccess)

Jk-WD90A(blsuccc~I

- - - lk MC397[blsucccss)

--ltgt-- lk middotMC31 l[blsucccss)

--- Jk [bl 6 out of6succcsHascs)

05 I 5 2 250

b

Likelihood Functionsmiddot Success Cases - Phbullsc L

shy

bullbull bull bull bull bull bull -------------------~

bull bull bullbull bullbull bullbull bullbull bullbull bullbull bull

bull bullbull

bullbullbullbull

bullbull

bullbullbullbullbullbullbull

bullbullbullbullbullbull

bullbullbullbullbullbullbull

z 0

t z ) LL

c 0 0c i ~ i

J

~ sect ~--------------~ e l5

I

I

SOFTWARE

bull PF PROGRAM

bull CALREL

bull PF +CALREL

bull RELACS

bull

gt i

l

SENSITIVITY STUDIES

POSTERIOR DISTRIBUTION OF B

bull EFFECT OF CHANGE IN DISTRIBUTION OF PRIOR OF B

bull EFFECT OF REMOVAL OF ONE PLATFORM FROM EACH CATEGORY

bull EFFECT OF REMOVAL OF THREE DAMAGE CASES

bull EFFECT OF ADDITIONAL 13 SIMILAR PLATFORMS

-I

Prior and Posterior Distributions of B Variation In COV of PriorofB fromOl to04 -- -

4500

-4000

-

- -

3500 --middot - bull bull - bull bull Prior-r ICOV =02

3000

- - - - Prior-r ICOV =03 -

bull -~ e 2soo -

bull bull bull bull bull bull bull Pnor-r ICOV =o 4t frJ ~ 2000 Posterior rlCOV =0 2

bull bull Postenor rlCOV =0 3 bull bull 1500 bull middot bull - shy -middot- Posterior rlCOV =04

middot~ I 000 - bull

-middotmiddot ~ ~middotmiddot M -~

- middot 0 500

wbullbullbull _ I middot~ middotmiddotmiddotmiddotmiddot middotmiddotmiddotmiddotmiddot - -~ ~ __ middot_ _ bull -shy0 000

0 000 0200 0 400 0 600 0 800 IOOO 1200 1400 1600 1800 2000 -

- - b

-

=shy

Posterior Distribution of Bias Factor (B) bull Effect of Variation In COV of Prior of B

bull

bullPrior and Posterior Distributions of B Evaluating Effect of Removal of One Platform from Each Category from the Base Case of 13 Platforms

400

~ ~ 350

j IA i

3 00 bull I bull I ---Prior-r I bull l

250 1 I 1

I bull bull ---Posterior r lall 13 casesshy- ~E 2 00 bull bull bull bull bull bull bull Posterior r without STISIK

- ~ 1 bull - bull I - bull bull - bull bull Posterior r Iwithout T2I I 50 I

J-tt~i - bull - bull - Posterior r Iwithout T2S

tOO [-

I I ~ ~ I 0 50

0 00

___ r J i

l~T

~--

0 00 0 20 0 40 0 60 080 I 00 120 140 160 110 2 00

b

zshy

Posterior Distribution of Bias Factor (8) bull Effect of Removal of middotPlatform from Any Category

bullPrior and Posterior Distributions or B Evaluating Effect or Removal or 3 Damage Cases from the Base Case or 13 Platrorms

4

35

3

25

e L 2 ~ bulle shy

I 15

bull I

05

0

0 02 04 06 I 2 I 4 16 IS 2

I I

I

I

I Ibull

J -tmiddot L--

bullI

~

I

fl

II ~

I I v I

middot bull bull

bull L

~ ll- I-~lt

OS I

b

I

bull

1middot bullbullbullbull

~ __

bull

- - - - Prior-r

- - - - Postenor flall failure cases

Postenor flall success cases

--- Posterior flall 13 cases

---Posterior fI success failure cases only

IPosterior or B Mean =t11 COV =0121

-

_I

Posterir Distribution or Bias Factor (B) bull Effect or Removal or 3-Damage Cases

-

bullPrior and Posterior Distributions or B Evaluating Effect or Additional 13 Similar Observations as

Used in Andrew JIP The POsterior or B from Andrew-JIP Considered as New Prior or B

- 500

450

400

350

bull bull

bull bull bull bull J bull

bull bull f ~ I

I ~ Ibull bull

bullbull bull

l bull I -bullbull bull bull

I

I bull bull

bull I I~~bull bull bull

bull ~

I bull bull

~ ~_ ~~~

- - New Prior r bull

Posterior 114 new failure cases ~ 300- bull bull bull bull bull bull bull Posterior f 13 new damage casesl 150 ~

- bull bull - bull bull Posterior 116 new success cases i 2 00

Postenor 1113 new success failure -150 damage cases

I 00

0 50 -

0 00 0 00 0 20 0 40 0 60 0 80 I 00 120 I 40 160 110 2 00

Posterior ol B Mean= 120 COV= 0071 b

shy

shy

_rshy

Posterior Distribution of Bias Factor (B) bull Effect of Additional 13-Similar Observations

1

1 I

WEIGHTING APPROACH bull

bull ASSESSMENT OF PLATFORMS DATA

Select Platforms Most Mappable to the Analyzed Platforms

Unlikely SurvivalDamage Cases Identification

Based on Cut-Off Criteria Agemiddot Hs etc

bull INFORMATION REQUIRED FOR COMPARISON

bull CATEGORIZATION OF ADDITIONALmiddot PLATFORMS

bull UPDATING OF CALIBRATION RESULTS

--

OVERVIEW OF PLATFORMS SURVIVEDDAMAGED -~

Operator -

- Platforms

Survived CS)

Platforms

Damaged CD)

Platforms Used In

Phase IJIP -

JI~ ~Bill~IRBDll -

Amoco -

Chevron -

Exxon

Mob II

Phllllps

Shell

Unocal -

30

143

72 22

7

39 25shy -

1

12

bull

1 bull

1

1(S)+1 (D)

3 (S) +3(D)

2 (S)

bull shy

bull

bull

bull

Total Partlclnants Platforms 338 - 15 10C6S+4D) -

Other Operators

310

13

3 (D)

Total Platforms 648 28 13C6S+7D) -

Summary of Andrew JIP bull Phase I Platforms

Plairshy w v- N-ber Lea-le 1 shy car Hind_ o-w N-bull Deplh fl fl IJpe la bull n Yeadon Jadtet lbelledtf liefalit -- _ -OD

IL H - -shy _ICllshy IL IL - - rQClto ~ _

K STUIK 137 1963 8 365 608S4 Cellardeck

-middot STl-shy 170 1964 4 -~ K 39 62269 Cellardeck - --ST134W 137 1911 4 43 608S4 Smvlnlshy- _ - -

WD90A 184 1964 8 K 267Sbull34 83 SOS31 SUboftllar deck SmvMI

MC311 343 1978 8 x S7 62 - MC397 468 1991 4 x shy S33bull6SO 65911 -shySmeSmvlvol -- gt -

r-- T23ST52 63 1969 4 ton- shyKshy 52594 S03 -T25SSl39 62 1969 4 Kshy SOS94 S063 - middot ton-shy__ -

bullbull a ton--shyST161A 118 1964 8 K - 347S 51495 Cellarclecti

r~

ST1771 142 1965 8 K 39 60167 CeDordeck failme

STISIH 137 1964 8 K 395 608S4 CeDardeck Failme

K ST130A 140 1958 a SS 6095 I lblv Pai1uno - T21ST72 61 1969 4 Kshy S4S94 4974 I lhlv Pai1uno -

- -

JE

INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON

bull WATER DEPTH

bull bull YEAR OF INSTALLATIONDESIGN

bull PLATFORM ORIENTATION

bull PLATFORM LATITUDE AND LONGITUDE

bull KEY SKETCHES OF PLATFORM STRUCTURAL FRAMING

Number of Legs Plies Leg Batter Number of Conductors bull Deck Elevations (Including sub-cellar deck If any) Jacket Framing Elevations Brace type In Vertical Frames Sizes of Primary Members (leg pile vertical frames braces primary horizontals) Joint Details (can gap etc) LegPiie Annulus Grouting Status

bull middotPOST ANDREW INSPECTION REPORT Significant Wave Height Hs (ft)

bull INFORMATION ON DAMAGES PRIOR TO ANDREW

bull RESULTS FROM IN-HOUSE STUDIES IF AVAILABLE

bull ~RIGINAL DESIGN CRITERIA INFORMATION IF AVAILABLE

bull SOIL SHEAR STRENGTH PROFILE IF AVAILABLE

Platforms Identified for Evaluating Therr Likelihood of Falling Under Unexpected Survival Category

OPlRATOR

middotADA

BLOClt

llmucnJllE

NAME

YEAll

INSTAUJD

WA1ER

DBPTll

DISTANCE

JlllOM SHOllE

(MILU)

Bl

(M)

AMOCO 111113

STl11

WDI075

A

bull D

St

17

120

1n

44

20

10 I

CllllVllQfl STlllO c 61 110 21 11 STIUO D Q 1110 21 11

STl134 p Q 137 29 11

ST0151 G 61 137 11

ST0151 I 3 121 11

ST0151 J Q 140 11

STl151 It 137 11

ST0151 PJl001 Q UT 11

ST0151 PROD-2 Q UT 11

ST017 A 3 140 35 11

WDIUT c 65 214 34 10

llXXON STI055 II 51 6T 14 I

STl15 A t3 34

WDI073

WD0073

Amiddotshy Q

1111 1111

22

22

I

I

WDI073 c 1n 22 I

- WDOCl93-shy II

65 1110 2T A 3 194 31 MOBU 111121

SS IOl7

A B

501

50

50

31

T

I

T 6

MtlllPllY SS114 B T 50 T T

PlllllIPS SS0149 A 5 55 33 T

SAMllDAN STln A 107 T T

SBllLL WDl103 A 65 223 2T WDl103 B 65 221 2T WD0104 c 65 221 2T WD0104 D 6T 20 2T

UNOCAL SS020I

SS 020I SSl20I

SS 0209

SS0209

II p

PURll

A

B

3

n 61

103

97

97

95

100

34

34

34

34 34

bullbull bullbullbull

- SS0209 D u 95 34 I

- SS 1209 SSl253

SS0253

G A

c

6T

2

100

165

1T5

34

54

54

bullbull I

I I

I I

SOUTH MARCH

~LAND

N

SOUTH PaTOI ENE ISWltD AREA

MMS Defined Limits ror Offshore Structures Requiring Inspections Following Andrew

t---Jbulln3 T25bull I bullTtl

bull1$1Htl1Abull bull1SIK1779bull

~

middot-DaTA AREA~T

CRANO rsLE

AREA bull1300

bullll4W

=shy

bullPath of Hurricane Andrew with Platforms Used In Calibration

Likelihood Function vs b All Success Case STISI K STIJOQ ST134W WD90A MCJll MCJ97

I 0000

0 9000

0 8000

07000

-Bo6000

tiiii 0 5000 -

04000

0 3000

0 2000

01000

0 0000

0 05 25

ri1rt middot-middot-middot-middot-middot-

I 5 2

b

---11 middot151K(blsu~ccsI

- bull - bull - fk 130Q(bfU~lCSS)

lie bull134W(blsuccc]

lk-WD90A[blsultCS]

- - - lk middotMC397(bfUtlCS]

--o-- lk-MCJll(bbuctcsJ

--- ll (bl 6 out of 6 Mlttcs case]

_

( Likelihood Functions middotSuccess Cases - p - 1 c I

CATEGORIZATION OF ADDITIONAL PLATFORMS

I I I

bull CATEGORIES

Unexpected Survivals

Expected Survivals

bull APPROACH

Identify Corresponding Platform(s) Evaluated In the JIP

Determine Max Wave Height at Platform Location

Evaluate Whether Wave Hit the Deck

middot - Evaluate Likelihood of R and S to Differ from the Base Cases

Review Joint Information

Assess Special Features

bull IDENTIFY NUMBER OF PLATFORMS IN EACH CATEGORY

I

UPDATING OF CALIBRATION RESULTS

bull INCLUDE EFFECT OF ADDITIONAL PLATFORMS ON bullBbull

Consider Increased 1Uncertalntles In R and S

Develop bullNeW Ukellhood Functions for Different Categories

middotLet bullxbull Number of Ukely Unexpected Survival Cases and y Number of Ukely Expected Survival Cases

Determine Combined Likelihood Function lk (b Iadditional platforms) middot

bull REVISED LIKELIHOOD FUNCTION

lk(b I all cases) = lk(b I 9 platforms) middot x lk(b Ix platforms) x lk(b I y platforms)

bull ESTABLISH REVISED BIAS FACTOR B

f8 (b) u f8 (b) lk (b I all cases)

II

t t ~

shy

APPLICATIONS OF BIAS FACTORS

ii APPLICABLE TO PARTICULAR LOAD AND RESISTANCE RECIPE

bull USE BY THE API COMMITTEES AND MMS

Criteria Revisions

Recommended Design Guidelines Revisions bull

bullbull REASSESSMENT STUDIES AND DECISION MAKING

middot

bull RISK ASSESSMENT STUDIES BY THE OPERATORS

Individual Platforms

Fleet of Platforms

middot IMR Program Implementation

bullbull NOT TO BE USED IN DESIGN

I I fl J

EXAMPLE USE OF BIAS FACTORS

bull PHASE I EXAMPLE APPLICATION

bull EVALUATION OF ANNUAL PROBABILITY OF FAILURE

Establish Annual Maximum Seastate Data

Determine Coefficients to Define Base Shear

Determine Ultimate Capacity of Platform

Determine Probablllty of Failure Versus bullbbull

-bull DETERMINE UPDATED PROBABILITY OF FAILURE

Pf - P [S gt R]

Pf I P [~ gt Rb I b] f 8 (b) db

Pr - J P [S gt Rb I b] f bull 8 (b) db

bull TOTAL PROBABILITIES OF FAILURE

Total Pf - 1 bull Il [1 middotPf] - l Pf bull

Total Pf - 1 bull Il [1middot Pf] - l Pfmiddot

Total Pr = 1 - n [1middot PfbullJ = l Pfshy

I I I

EXAMPLE APPLICATION bull PHASE I

PLATFORM ST 151 K

bull GULF OF MEXICO

bull WATER DEPTH = 137 FT

bull ANNUAL MAXIMUM HS MEDIAN 175 ft COV 031

bull ASSUMED STORM DURATION 3 hoursmiddot

bull ASSUMED NUMBER OF WAVES 800

bull MEDIAN HS FOR STORM HOURS 1 2 AND 3

bull MEDIAN CURRENT VALUE bull

bull DETERMINE ULTIMATE CAPACITY FOR THREE DIRECTIONS

bull DETERMINE ANNUAL FAILURE PROBABILITIES

Broadsidemiddot Diagonal End-on

0037 0025 0007

0071 0044 0018

0022 0016 0004

DETERMINE TOTAL ANNUAL PROBABILITY OF FAILURE

Pf - 007

Pf - 013

Pr - 004

bull

EXAMPLE USE OF BIAS FACTORS - PHASE II middot

bull DETERMINE PREDICTIVE PROBABILITY OF FAILURE

Case - 1

Pf _ S S Pf (b b2) fa1a2 (b b2) db db2

g bull -1 (b1 R I S) - 1

92 - (b2 R2 I S) - 1

Case -2

Pf s ~ [bJ f11 (b) db

Pf [bJ p [g lt 0)

g - bull 1 (b1 R1 I S) - 1

1

bull1) l

I

BENEFITS OF ANDREW JIP bull PHASE II

lt

bull IMPROVED LOAD AND RESISTANCE RECIPE

bull IMPROVED CALIBRATION PROCEDURE

bull MULTIPLE BIAS FACTORS

bull PROCEDURE TO INCLUDE EFFECT OF OTHER PLATFORMS

bull PROCEDURE TO USE MULTIPLE BIAS FACTORS

  • Structure Bookmarks
    • ANDREW PHASE II JOINT1 INDUSTRY PROJECT
    • PROGRESS MEETING NUMBER 1
    • ANDREW PHASE II JOINT INDUSTRY PROJECT PROGRESS MEETING NUMBER 1 NOVEMBER 16 AGENDA
    • ANDREW PHASE II UST OF PARTICIPANTS
    • ANDREW PHASE I OBJECTIVES AND RESULTS
    • ANDREW PHASE II PROJECT OBJECTIVES
    • SUBSEQUENT PROJECTS
    • ANDREW PHASE II PROJECT TEAM
    • ANDREW PHASE II UPDATE MODELING middotAND ANALYSIS RECIPE
    • -ANDREW PHASE II CONDUCTOR RESISTANCE
    • ANDREW PHASE Il WA VE FORCE PROFILE
    • ANDREW PHASE II WA VE FORCE PROFILEmiddot
    • ANDREW PHASE II BRACE MODELING
    • ANDREW PHASE II FOUNDATIONS
    • ANDREW PHASE II middot middot JOINT MODELING
    • ANDREW PHASE Il JOINT MODELING OPTIONS DISCUSSION
    • ANDREW PHASE II JOINT MODEJJNG OPTIONS ILLUSTRATIONS
    • ANDREW PHASE II JOINT MODELING PHASE II METHOD
    • ANDREW PHASE II ANDREW STORM CONDmONS UPDATED
    • Platform Selection bullbullbull1111111111111
    • BasisCriteria for Selection bullbullbull1111111111111
    • CALIBRATION
    • WHERE TO INTRODUCE BIAS FACTOR (B
    • PHASE I -KEY CONCLUSIONS
    • bull CALIBRATION ISSUES PHASE II
    • PHASE II WORK SCOPE (CALIBRATION)
    • SOFTWARE
    • SENSITIVITY STUDIES POSTERIOR DISTRIBUTION OF B
    • WEIGHTING APPROACH bull
    • INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON
    • CATEGORIZATION OF ADDITIONAL PLATFORMS
    • UPDATING OF CALIBRATION RESULTS
    • APPLICATIONS OF BIAS FACTORS
    • EXAMPLE USE OF BIAS FACTORS
    • EXAMPLE APPLICATION bull PHASE I PLATFORM ST 151 K
    • EXAMPLE USE OF BIAS FACTORS -PHASE II middot
    • BENEFITS OF ANDREW JIP bull PHASE II
Page 45:  · . ' ANDREW PHASE II PROJECT OBJECTIVES . ' • ·UPDATE MODELING AND ANALYSIS RECIPE· JOINTS ' ' ·BRACES WAVE PROFILE FOUNDATIONS . ' ' CONDUCTORS· • ANALVZE PLATFORMS USING

Operator - Platfonns - Platfonns Survived Damagedamp

- Falled -

bull

-

-

Jle eBDIGIPllDbl -Amocoshy 30 1 Chevron -

143 12 Exxon 72 Mobll 22 1 Phllllps 7 bull

-Shell 39 Unocal 25 1

Total Partlclnts Platfonns 338 15

Other Operators bull- 310 13

-

Total Platfonns - -

648 28

fbml

Platforms Considered 6 7

tie of Total Platfonns 1 25

OVERVIEW OF PLATFORMS SURVIVEDDAMAGEDFAILED

-

shy

v

Platforms Evaluated in Phase I

-

-Platfonn Water Year Number Analysis Avallabillty of Availability Platform

- Name Depth rl of Model DetaDed rl ClassUlcatlon

Installation legs Inspection Soll Report Based on Information Phase I Calibration - -ft

- -- Suaill flalfnrm ~BSi -STISIK 137 1963 8 CAP Yes Yes Un eel Sarvlval

-STIJOO 170 1964 4 CAPmiddotshy Yes ST134data u ST134W 137 1981 4 CAP Yes Yesmiddot E

-

edSarvlval -

middotsurvival

WD90A 184 - 1964 8 CAP Yes E -

- Sarvlvill

MC3ll 343 1978 8 KARMA

MC397 468 1991 4 KARMA -

bull ~reglmmiddot- -

Yes Sure Survival

bull Yes - Sare Sanlval-

-T23STS2 63 1969 4 CAP

T2SSS139 62 1969 4 CAP -

Yes

Yes

Yes Ex

Yes Une

bull SurvivalLikely to DamalI

middot bull SanlvalLlkely to DamalI

ST161A 118 1964 8 USFOS

Eail11a pound1al(bullllm Clss ST177B 142 --1965 8 CAP

STISIH 137 1964 - 8 CAP C

STIJOA 140 19S8 8 CAP - -

Yes

Yes

ollamed

Yes

- Yes E

-

at about 1 mUe

Yes

ST134data

bull SafvlvalLlkelv to Dam- bull- -

E middotrahre

Emeried Failure -

Likely FaUare

T21ST72 61 1969 4 CAP Collamed No Likely FaUure

Platforms Identified for Evaluating Their Likelihood of Falling Under UnexpeCted Survival Category

CllEllATOa

AllB

llLOCll

Sl1lvcnJllE

NAME

YEAR

INSTAUJCD

WATER

IJBPJB

middotmiddotDISTANCE

ftOMSllOllE

(MIUIS)

Ill

(M)

AMOCO BllW

STllI

WDI075

A

bull D

4

n uo 17J

44

32

bull 20

10

bull CllEVllOH STtllO c I 1bull 21 11

STl130 D 0 161 21 11 STllJ4 I 0 137 D - 11

STIUI G I 137 32 11

STIUI I 0 121 32 11

STIUI

J 0 140 32 11

STIUI It 4 137 32 11

STllSI PROl)I 0 137 32 11

STllSI PROD-2 0 137 32 11

STll7 A 0

140 35 11

WDl117 c 65 214 34 10

DXON STshy E 51 14 bull STl165 A t3 bull 34 WD0073

WD0073

WD0073

A _ c

4

62

4

IA IA

17J

22

22

22

bullbullbull WD003-shy E 65 161 Z7

A 0 IN 31 MOBD Bl12I

ssn A bull

50

50

31

7

bull 7

MIJllPllY SS114 B 7 50 7 7

PlllUIPS SSl14t A 55 33 7

SAMEDAN ST17J A 4 117 7 bull 7

SJIELL WD0103 A 65 223 27 WDllOJ bull 65 221 27 WDll04 c 65 221 27 WDOI04 D 67 2 27

UNOCAL

SSl20I

SSl20I

bull SSl20I

SS0209

SS0209

SS0209

SS0209

SS 0253

SS 0253

E

I

PtJllE

A

bull D

G A

c

0

4

4

7J

61

0

0

69

103

97

97

9S

100

95

100

165

175

34

34

34

34

34

34

34

54

54

bullbullbullbullbullbullbullbullbull

Platforms Recommended for Phase II

llalr y_ w- Aot Av- _ I 1 Doplla Model nm bull bull _ bull I I H11 s18-1 Irht lbwl~ bullbullbull(-illtdlel OM lllMtD

II ---

-STISIIC 137 1963 I CAP Yeo Yeo I Yeo

STtshy -170 1964 4 CAP Yeo ST134dlU I Yeo

STl34W 137 Yeo Yeo -1981 4 CAP D Yeo

WD90A 114 1964 I CAP Yeo - D

MC311 30 1971 I KARMA Yeo -sre-

MC397 middot 461 1991 4 ICARMA Yeo Sireshy - - r- -

__ T13ST5l 63 1969 4 CAP Yeo Yeobull middot- ~n- D Yeo _ Tl5SSl39 62 1969 4 CAP Yeo Yeo I Yeo

_ STIT711 142 1965 I CAP Yeo I Yeo middot-1shySTl61A Ill 1964 I USFOS Yeo Yeo - D bulloshy

- r-

STISIH 137 1964 I CAP y D y -_ STl31A 140 1951 I CAP Yeo STl34dlU I Yeo

T21Sll 61 1969 4 CAP No rar-middot I Yeo

~-rmiddot-middot SS209A 95 1967 16 CAP Yeo D -SSU4R

11 UI A 41 -in D SS 2741AIM CAP bull

_ _ -shy - -shy- -shy

CALIBRATION

bull PHASEIAPPROACH

bull CALIBRATION ISSUES

bull PHASE II WORK SCOPE

bull SOFlWARE AVAILABILITY

bull EFFECT OF ADDITIONAL PLATFORMS

bull WEIGHTING APPROACH

bull EXAMPLE APPLICATION middot

(l=Ofce Mode] Wind

~

fcurrent l -middot

~SW ~leldl4

ultimate resistance force I first

design load Illgt

i _

deflection

-

bull

4

Historical Hurrlcane Dariiage and Survival of Platforms

Components Required for Calibration (Petrauskas 1992)

PRIOR Distribution

PLATFORM EXPERIENCE

BAYESIAN Survivals and Failures

UPDATING

Waves Winds and Currents bull

POSTERIOR Distribution

- A_middotJA ~ Survival Data Failure Data

Result Result

Load Resistance

middotfl

-

Bayesian Updaing middotGeneral Description (Petrauskas 1992)

L Imiddot

WHERE TO INTRODUCE BIAS FACTOR (B

On Resistance (Rmue - B (RPREDICreD I

middotOn Loads (Smue middot - B (SPRE01cTEo

On Safety Margin (RSmue - B (RSPREDICTED

CALIBRATION METHODOLOGY

gt

r

FIELD OBSERVATIONS

Observed Behavior

survivals Damages

Failures

r

bull

CAPACITY ANALYSIS

Predicted Load and Resistance Predicted Failure Modes

Survival Damage Failure cases

RELIABILITY ANALYSIS

- Probability of Failure ~

Ukellhood Failure

BAYESIAN UPDATING

Prior Distnbution of B Combined Likelihood Function

Posterior Dlstnbution of B

BIAS FACTOR

Mean Value cov bull

Represents Modeling Uncertainties

Prior Distribution of B Likelihood Functionmiddot Observing Success case STISIK and Posterior Distribution of B

2 00

uo

160

140

120

e 100 L $ uoIshy

0 60 ~ ~ 040

020

0 00

0 00 0 50

_ I 00

b

I 50 2 00

-shy lk [bl STISIK)

--shy Pnor-f(b)

--shy Pos1enor r(bl ISIK)

Prior or B Mean bull 10 COV bull 030

Posterlor of Bbull Meanbull 136COV 0 16

2 50

Likelihood Function and Posterior Distribution Example Platform STISIK

- -

~

- Prior and Posterior Distributions or B Failure cases (ST177B STlSlH ST130A T21) Damage cases (T23 T25 ST161A) Success cases (STlSlK ST130Q ST134W WD90A MC311 MC397)

400

350

300

2 so 0-

I

bull bull

bull bull gt

- bullbull ~ bull bull I bull

_ bull bull bullbull I bull bull bullII V -

bull bull

I bull ~ bull

bullbull bullI bull bullI bull bull bull

middot -~

bull__ ~ middotI middotmiddot

L 2 00

e ~ I SO

-I 00

- 0 so

000 gt

0 00 020 0 40 060 0 80 I 00 120 140 160 110 200

b -

-

shy

Prior-r

- - - - Posterior fLlll failure cases

bull bull bull bull bull bull bull Posterior flall damage cases

- bull bull - bull middot Postenor flall success cases

Posterior flall success fulure damage cases

-

Posterior of B Mean= 119 COV 0101

Posterior Distribution or Blas Factor (8) bull Different Categories and All Cases Combined

shy

l

PHASE I - KEY CONCLUSIONS

bull GLOBAL CORRECTION FACTOR

bull 15 TO 20 CONSERVATISM MEAN)

bull UNEXPECTED SURVIVALS HAVE GREATEST IMPACT

bull DAMAGED CASES MORE INFORMATIVE

bull CALIBRATION ISSUES PHASE II

bull

EFFECT OF IMPROVED CAPACITY ANALYSIS RECIPE

bull APPLICABILITY TO MULTIPLE FAILURE MODES

bull NO FOUNDATION DAMAGE CASE INCLUDED

bull IMPROVED REPRESENTATION OF FULL ANDREW DISTRIBUTION

-bull NUMBER OF PLATFORMS CONSIDERED

I 1 I

I t ) bull ~

PHASE II WORK SCOPE (CALIBRATION)

bull

bull ESTABLISH MULTIPLE BIAS FACTORS

Use Nine Base Case Platforms

middot - Capacity Analysis for Revised recipe

Foundation Blas Factor from Foundation JIP

bull DEVELOP WEIGHTING APPROACH

bull UPDATE MULTIPLE BIAS FACTORS

bullEffect of Platforms Not In the Nine Base Cases

bull EXAMPLE APPLICATION OF BIAS FACTORS I

bullbull

PHASE I PHASE II

CAPACITYbull ANALYSISRu bull

---------------------1----middot

Jc1 - iJbullraquoJ dbull

FttlS frt[f~l ratdhJJ

-

LIKELIHOOD FUNCTION

DEVELOPMENT

lk Jb Jfall1N) bull Jb)

111 Ill Isurvtvoq bull 1 - bl

-middot -------------------shy ~---------------------middot-middot

r111ir8 Cb1 111 Cbl BAYESIAN

UPDATING

~cove

R1R2 bull middot---------------------~----

Jblbull P(g cOJ

1 bull JbRI (MBSI I bull 1

~middot

llcJllJfalluro) bull P(gcOJ Ill 111 b21fallun) bull P ((g1 CO) U 1112 c OD

middot--------------------shy ----------------------shy bull

r111111r8 Cbl 111 Ill i-1nro1

deg ~1b2)laquottfrr (b1b21-lnfo) re Ill bull 1-11middot b2) bull lodegamp

middot

i1 covamp i2 cov~i

-

shy

--

Likelihood Function vs b All Success Case STISI K STl30Q STl34W WD90A MC311 MC397

I 0000

0 9000

OIOOO

0 7000

~ 0amp000

ta osoOO

~ 0 4000

0 3000

0 2000

0 1000

bull 0 0000

bull

I

middot

middot bull I

Ibull

-sibull ~ -middot-middot-middot-middot -middot-

----7----+-----~ I

Ibull bull bull I

rbull

--- Jk ISIK[blsuccess)

- bull - bull - lk 130Q[blsuccess) bull

lk 134W[blsuccess)

Jk-WD90A(blsuccc~I

- - - lk MC397[blsucccss)

--ltgt-- lk middotMC31 l[blsucccss)

--- Jk [bl 6 out of6succcsHascs)

05 I 5 2 250

b

Likelihood Functionsmiddot Success Cases - Phbullsc L

shy

bullbull bull bull bull bull bull -------------------~

bull bull bullbull bullbull bullbull bullbull bullbull bullbull bull

bull bullbull

bullbullbullbull

bullbull

bullbullbullbullbullbullbull

bullbullbullbullbullbull

bullbullbullbullbullbullbull

z 0

t z ) LL

c 0 0c i ~ i

J

~ sect ~--------------~ e l5

I

I

SOFTWARE

bull PF PROGRAM

bull CALREL

bull PF +CALREL

bull RELACS

bull

gt i

l

SENSITIVITY STUDIES

POSTERIOR DISTRIBUTION OF B

bull EFFECT OF CHANGE IN DISTRIBUTION OF PRIOR OF B

bull EFFECT OF REMOVAL OF ONE PLATFORM FROM EACH CATEGORY

bull EFFECT OF REMOVAL OF THREE DAMAGE CASES

bull EFFECT OF ADDITIONAL 13 SIMILAR PLATFORMS

-I

Prior and Posterior Distributions of B Variation In COV of PriorofB fromOl to04 -- -

4500

-4000

-

- -

3500 --middot - bull bull - bull bull Prior-r ICOV =02

3000

- - - - Prior-r ICOV =03 -

bull -~ e 2soo -

bull bull bull bull bull bull bull Pnor-r ICOV =o 4t frJ ~ 2000 Posterior rlCOV =0 2

bull bull Postenor rlCOV =0 3 bull bull 1500 bull middot bull - shy -middot- Posterior rlCOV =04

middot~ I 000 - bull

-middotmiddot ~ ~middotmiddot M -~

- middot 0 500

wbullbullbull _ I middot~ middotmiddotmiddotmiddotmiddot middotmiddotmiddotmiddotmiddot - -~ ~ __ middot_ _ bull -shy0 000

0 000 0200 0 400 0 600 0 800 IOOO 1200 1400 1600 1800 2000 -

- - b

-

=shy

Posterior Distribution of Bias Factor (B) bull Effect of Variation In COV of Prior of B

bull

bullPrior and Posterior Distributions of B Evaluating Effect of Removal of One Platform from Each Category from the Base Case of 13 Platforms

400

~ ~ 350

j IA i

3 00 bull I bull I ---Prior-r I bull l

250 1 I 1

I bull bull ---Posterior r lall 13 casesshy- ~E 2 00 bull bull bull bull bull bull bull Posterior r without STISIK

- ~ 1 bull - bull I - bull bull - bull bull Posterior r Iwithout T2I I 50 I

J-tt~i - bull - bull - Posterior r Iwithout T2S

tOO [-

I I ~ ~ I 0 50

0 00

___ r J i

l~T

~--

0 00 0 20 0 40 0 60 080 I 00 120 140 160 110 2 00

b

zshy

Posterior Distribution of Bias Factor (8) bull Effect of Removal of middotPlatform from Any Category

bullPrior and Posterior Distributions or B Evaluating Effect or Removal or 3 Damage Cases from the Base Case or 13 Platrorms

4

35

3

25

e L 2 ~ bulle shy

I 15

bull I

05

0

0 02 04 06 I 2 I 4 16 IS 2

I I

I

I

I Ibull

J -tmiddot L--

bullI

~

I

fl

II ~

I I v I

middot bull bull

bull L

~ ll- I-~lt

OS I

b

I

bull

1middot bullbullbullbull

~ __

bull

- - - - Prior-r

- - - - Postenor flall failure cases

Postenor flall success cases

--- Posterior flall 13 cases

---Posterior fI success failure cases only

IPosterior or B Mean =t11 COV =0121

-

_I

Posterir Distribution or Bias Factor (B) bull Effect or Removal or 3-Damage Cases

-

bullPrior and Posterior Distributions or B Evaluating Effect or Additional 13 Similar Observations as

Used in Andrew JIP The POsterior or B from Andrew-JIP Considered as New Prior or B

- 500

450

400

350

bull bull

bull bull bull bull J bull

bull bull f ~ I

I ~ Ibull bull

bullbull bull

l bull I -bullbull bull bull

I

I bull bull

bull I I~~bull bull bull

bull ~

I bull bull

~ ~_ ~~~

- - New Prior r bull

Posterior 114 new failure cases ~ 300- bull bull bull bull bull bull bull Posterior f 13 new damage casesl 150 ~

- bull bull - bull bull Posterior 116 new success cases i 2 00

Postenor 1113 new success failure -150 damage cases

I 00

0 50 -

0 00 0 00 0 20 0 40 0 60 0 80 I 00 120 I 40 160 110 2 00

Posterior ol B Mean= 120 COV= 0071 b

shy

shy

_rshy

Posterior Distribution of Bias Factor (B) bull Effect of Additional 13-Similar Observations

1

1 I

WEIGHTING APPROACH bull

bull ASSESSMENT OF PLATFORMS DATA

Select Platforms Most Mappable to the Analyzed Platforms

Unlikely SurvivalDamage Cases Identification

Based on Cut-Off Criteria Agemiddot Hs etc

bull INFORMATION REQUIRED FOR COMPARISON

bull CATEGORIZATION OF ADDITIONALmiddot PLATFORMS

bull UPDATING OF CALIBRATION RESULTS

--

OVERVIEW OF PLATFORMS SURVIVEDDAMAGED -~

Operator -

- Platforms

Survived CS)

Platforms

Damaged CD)

Platforms Used In

Phase IJIP -

JI~ ~Bill~IRBDll -

Amoco -

Chevron -

Exxon

Mob II

Phllllps

Shell

Unocal -

30

143

72 22

7

39 25shy -

1

12

bull

1 bull

1

1(S)+1 (D)

3 (S) +3(D)

2 (S)

bull shy

bull

bull

bull

Total Partlclnants Platforms 338 - 15 10C6S+4D) -

Other Operators

310

13

3 (D)

Total Platforms 648 28 13C6S+7D) -

Summary of Andrew JIP bull Phase I Platforms

Plairshy w v- N-ber Lea-le 1 shy car Hind_ o-w N-bull Deplh fl fl IJpe la bull n Yeadon Jadtet lbelledtf liefalit -- _ -OD

IL H - -shy _ICllshy IL IL - - rQClto ~ _

K STUIK 137 1963 8 365 608S4 Cellardeck

-middot STl-shy 170 1964 4 -~ K 39 62269 Cellardeck - --ST134W 137 1911 4 43 608S4 Smvlnlshy- _ - -

WD90A 184 1964 8 K 267Sbull34 83 SOS31 SUboftllar deck SmvMI

MC311 343 1978 8 x S7 62 - MC397 468 1991 4 x shy S33bull6SO 65911 -shySmeSmvlvol -- gt -

r-- T23ST52 63 1969 4 ton- shyKshy 52594 S03 -T25SSl39 62 1969 4 Kshy SOS94 S063 - middot ton-shy__ -

bullbull a ton--shyST161A 118 1964 8 K - 347S 51495 Cellarclecti

r~

ST1771 142 1965 8 K 39 60167 CeDordeck failme

STISIH 137 1964 8 K 395 608S4 CeDardeck Failme

K ST130A 140 1958 a SS 6095 I lblv Pai1uno - T21ST72 61 1969 4 Kshy S4S94 4974 I lhlv Pai1uno -

- -

JE

INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON

bull WATER DEPTH

bull bull YEAR OF INSTALLATIONDESIGN

bull PLATFORM ORIENTATION

bull PLATFORM LATITUDE AND LONGITUDE

bull KEY SKETCHES OF PLATFORM STRUCTURAL FRAMING

Number of Legs Plies Leg Batter Number of Conductors bull Deck Elevations (Including sub-cellar deck If any) Jacket Framing Elevations Brace type In Vertical Frames Sizes of Primary Members (leg pile vertical frames braces primary horizontals) Joint Details (can gap etc) LegPiie Annulus Grouting Status

bull middotPOST ANDREW INSPECTION REPORT Significant Wave Height Hs (ft)

bull INFORMATION ON DAMAGES PRIOR TO ANDREW

bull RESULTS FROM IN-HOUSE STUDIES IF AVAILABLE

bull ~RIGINAL DESIGN CRITERIA INFORMATION IF AVAILABLE

bull SOIL SHEAR STRENGTH PROFILE IF AVAILABLE

Platforms Identified for Evaluating Therr Likelihood of Falling Under Unexpected Survival Category

OPlRATOR

middotADA

BLOClt

llmucnJllE

NAME

YEAll

INSTAUJD

WA1ER

DBPTll

DISTANCE

JlllOM SHOllE

(MILU)

Bl

(M)

AMOCO 111113

STl11

WDI075

A

bull D

St

17

120

1n

44

20

10 I

CllllVllQfl STlllO c 61 110 21 11 STIUO D Q 1110 21 11

STl134 p Q 137 29 11

ST0151 G 61 137 11

ST0151 I 3 121 11

ST0151 J Q 140 11

STl151 It 137 11

ST0151 PJl001 Q UT 11

ST0151 PROD-2 Q UT 11

ST017 A 3 140 35 11

WDIUT c 65 214 34 10

llXXON STI055 II 51 6T 14 I

STl15 A t3 34

WDI073

WD0073

Amiddotshy Q

1111 1111

22

22

I

I

WDI073 c 1n 22 I

- WDOCl93-shy II

65 1110 2T A 3 194 31 MOBU 111121

SS IOl7

A B

501

50

50

31

T

I

T 6

MtlllPllY SS114 B T 50 T T

PlllllIPS SS0149 A 5 55 33 T

SAMllDAN STln A 107 T T

SBllLL WDl103 A 65 223 2T WDl103 B 65 221 2T WD0104 c 65 221 2T WD0104 D 6T 20 2T

UNOCAL SS020I

SS 020I SSl20I

SS 0209

SS0209

II p

PURll

A

B

3

n 61

103

97

97

95

100

34

34

34

34 34

bullbull bullbullbull

- SS0209 D u 95 34 I

- SS 1209 SSl253

SS0253

G A

c

6T

2

100

165

1T5

34

54

54

bullbull I

I I

I I

SOUTH MARCH

~LAND

N

SOUTH PaTOI ENE ISWltD AREA

MMS Defined Limits ror Offshore Structures Requiring Inspections Following Andrew

t---Jbulln3 T25bull I bullTtl

bull1$1Htl1Abull bull1SIK1779bull

~

middot-DaTA AREA~T

CRANO rsLE

AREA bull1300

bullll4W

=shy

bullPath of Hurricane Andrew with Platforms Used In Calibration

Likelihood Function vs b All Success Case STISI K STIJOQ ST134W WD90A MCJll MCJ97

I 0000

0 9000

0 8000

07000

-Bo6000

tiiii 0 5000 -

04000

0 3000

0 2000

01000

0 0000

0 05 25

ri1rt middot-middot-middot-middot-middot-

I 5 2

b

---11 middot151K(blsu~ccsI

- bull - bull - fk 130Q(bfU~lCSS)

lie bull134W(blsuccc]

lk-WD90A[blsultCS]

- - - lk middotMC397(bfUtlCS]

--o-- lk-MCJll(bbuctcsJ

--- ll (bl 6 out of 6 Mlttcs case]

_

( Likelihood Functions middotSuccess Cases - p - 1 c I

CATEGORIZATION OF ADDITIONAL PLATFORMS

I I I

bull CATEGORIES

Unexpected Survivals

Expected Survivals

bull APPROACH

Identify Corresponding Platform(s) Evaluated In the JIP

Determine Max Wave Height at Platform Location

Evaluate Whether Wave Hit the Deck

middot - Evaluate Likelihood of R and S to Differ from the Base Cases

Review Joint Information

Assess Special Features

bull IDENTIFY NUMBER OF PLATFORMS IN EACH CATEGORY

I

UPDATING OF CALIBRATION RESULTS

bull INCLUDE EFFECT OF ADDITIONAL PLATFORMS ON bullBbull

Consider Increased 1Uncertalntles In R and S

Develop bullNeW Ukellhood Functions for Different Categories

middotLet bullxbull Number of Ukely Unexpected Survival Cases and y Number of Ukely Expected Survival Cases

Determine Combined Likelihood Function lk (b Iadditional platforms) middot

bull REVISED LIKELIHOOD FUNCTION

lk(b I all cases) = lk(b I 9 platforms) middot x lk(b Ix platforms) x lk(b I y platforms)

bull ESTABLISH REVISED BIAS FACTOR B

f8 (b) u f8 (b) lk (b I all cases)

II

t t ~

shy

APPLICATIONS OF BIAS FACTORS

ii APPLICABLE TO PARTICULAR LOAD AND RESISTANCE RECIPE

bull USE BY THE API COMMITTEES AND MMS

Criteria Revisions

Recommended Design Guidelines Revisions bull

bullbull REASSESSMENT STUDIES AND DECISION MAKING

middot

bull RISK ASSESSMENT STUDIES BY THE OPERATORS

Individual Platforms

Fleet of Platforms

middot IMR Program Implementation

bullbull NOT TO BE USED IN DESIGN

I I fl J

EXAMPLE USE OF BIAS FACTORS

bull PHASE I EXAMPLE APPLICATION

bull EVALUATION OF ANNUAL PROBABILITY OF FAILURE

Establish Annual Maximum Seastate Data

Determine Coefficients to Define Base Shear

Determine Ultimate Capacity of Platform

Determine Probablllty of Failure Versus bullbbull

-bull DETERMINE UPDATED PROBABILITY OF FAILURE

Pf - P [S gt R]

Pf I P [~ gt Rb I b] f 8 (b) db

Pr - J P [S gt Rb I b] f bull 8 (b) db

bull TOTAL PROBABILITIES OF FAILURE

Total Pf - 1 bull Il [1 middotPf] - l Pf bull

Total Pf - 1 bull Il [1middot Pf] - l Pfmiddot

Total Pr = 1 - n [1middot PfbullJ = l Pfshy

I I I

EXAMPLE APPLICATION bull PHASE I

PLATFORM ST 151 K

bull GULF OF MEXICO

bull WATER DEPTH = 137 FT

bull ANNUAL MAXIMUM HS MEDIAN 175 ft COV 031

bull ASSUMED STORM DURATION 3 hoursmiddot

bull ASSUMED NUMBER OF WAVES 800

bull MEDIAN HS FOR STORM HOURS 1 2 AND 3

bull MEDIAN CURRENT VALUE bull

bull DETERMINE ULTIMATE CAPACITY FOR THREE DIRECTIONS

bull DETERMINE ANNUAL FAILURE PROBABILITIES

Broadsidemiddot Diagonal End-on

0037 0025 0007

0071 0044 0018

0022 0016 0004

DETERMINE TOTAL ANNUAL PROBABILITY OF FAILURE

Pf - 007

Pf - 013

Pr - 004

bull

EXAMPLE USE OF BIAS FACTORS - PHASE II middot

bull DETERMINE PREDICTIVE PROBABILITY OF FAILURE

Case - 1

Pf _ S S Pf (b b2) fa1a2 (b b2) db db2

g bull -1 (b1 R I S) - 1

92 - (b2 R2 I S) - 1

Case -2

Pf s ~ [bJ f11 (b) db

Pf [bJ p [g lt 0)

g - bull 1 (b1 R1 I S) - 1

1

bull1) l

I

BENEFITS OF ANDREW JIP bull PHASE II

lt

bull IMPROVED LOAD AND RESISTANCE RECIPE

bull IMPROVED CALIBRATION PROCEDURE

bull MULTIPLE BIAS FACTORS

bull PROCEDURE TO INCLUDE EFFECT OF OTHER PLATFORMS

bull PROCEDURE TO USE MULTIPLE BIAS FACTORS

  • Structure Bookmarks
    • ANDREW PHASE II JOINT1 INDUSTRY PROJECT
    • PROGRESS MEETING NUMBER 1
    • ANDREW PHASE II JOINT INDUSTRY PROJECT PROGRESS MEETING NUMBER 1 NOVEMBER 16 AGENDA
    • ANDREW PHASE II UST OF PARTICIPANTS
    • ANDREW PHASE I OBJECTIVES AND RESULTS
    • ANDREW PHASE II PROJECT OBJECTIVES
    • SUBSEQUENT PROJECTS
    • ANDREW PHASE II PROJECT TEAM
    • ANDREW PHASE II UPDATE MODELING middotAND ANALYSIS RECIPE
    • -ANDREW PHASE II CONDUCTOR RESISTANCE
    • ANDREW PHASE Il WA VE FORCE PROFILE
    • ANDREW PHASE II WA VE FORCE PROFILEmiddot
    • ANDREW PHASE II BRACE MODELING
    • ANDREW PHASE II FOUNDATIONS
    • ANDREW PHASE II middot middot JOINT MODELING
    • ANDREW PHASE Il JOINT MODELING OPTIONS DISCUSSION
    • ANDREW PHASE II JOINT MODEJJNG OPTIONS ILLUSTRATIONS
    • ANDREW PHASE II JOINT MODELING PHASE II METHOD
    • ANDREW PHASE II ANDREW STORM CONDmONS UPDATED
    • Platform Selection bullbullbull1111111111111
    • BasisCriteria for Selection bullbullbull1111111111111
    • CALIBRATION
    • WHERE TO INTRODUCE BIAS FACTOR (B
    • PHASE I -KEY CONCLUSIONS
    • bull CALIBRATION ISSUES PHASE II
    • PHASE II WORK SCOPE (CALIBRATION)
    • SOFTWARE
    • SENSITIVITY STUDIES POSTERIOR DISTRIBUTION OF B
    • WEIGHTING APPROACH bull
    • INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON
    • CATEGORIZATION OF ADDITIONAL PLATFORMS
    • UPDATING OF CALIBRATION RESULTS
    • APPLICATIONS OF BIAS FACTORS
    • EXAMPLE USE OF BIAS FACTORS
    • EXAMPLE APPLICATION bull PHASE I PLATFORM ST 151 K
    • EXAMPLE USE OF BIAS FACTORS -PHASE II middot
    • BENEFITS OF ANDREW JIP bull PHASE II
Page 46:  · . ' ANDREW PHASE II PROJECT OBJECTIVES . ' • ·UPDATE MODELING AND ANALYSIS RECIPE· JOINTS ' ' ·BRACES WAVE PROFILE FOUNDATIONS . ' ' CONDUCTORS· • ANALVZE PLATFORMS USING

Platforms Evaluated in Phase I

-

-Platfonn Water Year Number Analysis Avallabillty of Availability Platform

- Name Depth rl of Model DetaDed rl ClassUlcatlon

Installation legs Inspection Soll Report Based on Information Phase I Calibration - -ft

- -- Suaill flalfnrm ~BSi -STISIK 137 1963 8 CAP Yes Yes Un eel Sarvlval

-STIJOO 170 1964 4 CAPmiddotshy Yes ST134data u ST134W 137 1981 4 CAP Yes Yesmiddot E

-

edSarvlval -

middotsurvival

WD90A 184 - 1964 8 CAP Yes E -

- Sarvlvill

MC3ll 343 1978 8 KARMA

MC397 468 1991 4 KARMA -

bull ~reglmmiddot- -

Yes Sure Survival

bull Yes - Sare Sanlval-

-T23STS2 63 1969 4 CAP

T2SSS139 62 1969 4 CAP -

Yes

Yes

Yes Ex

Yes Une

bull SurvivalLikely to DamalI

middot bull SanlvalLlkely to DamalI

ST161A 118 1964 8 USFOS

Eail11a pound1al(bullllm Clss ST177B 142 --1965 8 CAP

STISIH 137 1964 - 8 CAP C

STIJOA 140 19S8 8 CAP - -

Yes

Yes

ollamed

Yes

- Yes E

-

at about 1 mUe

Yes

ST134data

bull SafvlvalLlkelv to Dam- bull- -

E middotrahre

Emeried Failure -

Likely FaUare

T21ST72 61 1969 4 CAP Collamed No Likely FaUure

Platforms Identified for Evaluating Their Likelihood of Falling Under UnexpeCted Survival Category

CllEllATOa

AllB

llLOCll

Sl1lvcnJllE

NAME

YEAR

INSTAUJCD

WATER

IJBPJB

middotmiddotDISTANCE

ftOMSllOllE

(MIUIS)

Ill

(M)

AMOCO BllW

STllI

WDI075

A

bull D

4

n uo 17J

44

32

bull 20

10

bull CllEVllOH STtllO c I 1bull 21 11

STl130 D 0 161 21 11 STllJ4 I 0 137 D - 11

STIUI G I 137 32 11

STIUI I 0 121 32 11

STIUI

J 0 140 32 11

STIUI It 4 137 32 11

STllSI PROl)I 0 137 32 11

STllSI PROD-2 0 137 32 11

STll7 A 0

140 35 11

WDl117 c 65 214 34 10

DXON STshy E 51 14 bull STl165 A t3 bull 34 WD0073

WD0073

WD0073

A _ c

4

62

4

IA IA

17J

22

22

22

bullbullbull WD003-shy E 65 161 Z7

A 0 IN 31 MOBD Bl12I

ssn A bull

50

50

31

7

bull 7

MIJllPllY SS114 B 7 50 7 7

PlllUIPS SSl14t A 55 33 7

SAMEDAN ST17J A 4 117 7 bull 7

SJIELL WD0103 A 65 223 27 WDllOJ bull 65 221 27 WDll04 c 65 221 27 WDOI04 D 67 2 27

UNOCAL

SSl20I

SSl20I

bull SSl20I

SS0209

SS0209

SS0209

SS0209

SS 0253

SS 0253

E

I

PtJllE

A

bull D

G A

c

0

4

4

7J

61

0

0

69

103

97

97

9S

100

95

100

165

175

34

34

34

34

34

34

34

54

54

bullbullbullbullbullbullbullbullbull

Platforms Recommended for Phase II

llalr y_ w- Aot Av- _ I 1 Doplla Model nm bull bull _ bull I I H11 s18-1 Irht lbwl~ bullbullbull(-illtdlel OM lllMtD

II ---

-STISIIC 137 1963 I CAP Yeo Yeo I Yeo

STtshy -170 1964 4 CAP Yeo ST134dlU I Yeo

STl34W 137 Yeo Yeo -1981 4 CAP D Yeo

WD90A 114 1964 I CAP Yeo - D

MC311 30 1971 I KARMA Yeo -sre-

MC397 middot 461 1991 4 ICARMA Yeo Sireshy - - r- -

__ T13ST5l 63 1969 4 CAP Yeo Yeobull middot- ~n- D Yeo _ Tl5SSl39 62 1969 4 CAP Yeo Yeo I Yeo

_ STIT711 142 1965 I CAP Yeo I Yeo middot-1shySTl61A Ill 1964 I USFOS Yeo Yeo - D bulloshy

- r-

STISIH 137 1964 I CAP y D y -_ STl31A 140 1951 I CAP Yeo STl34dlU I Yeo

T21Sll 61 1969 4 CAP No rar-middot I Yeo

~-rmiddot-middot SS209A 95 1967 16 CAP Yeo D -SSU4R

11 UI A 41 -in D SS 2741AIM CAP bull

_ _ -shy - -shy- -shy

CALIBRATION

bull PHASEIAPPROACH

bull CALIBRATION ISSUES

bull PHASE II WORK SCOPE

bull SOFlWARE AVAILABILITY

bull EFFECT OF ADDITIONAL PLATFORMS

bull WEIGHTING APPROACH

bull EXAMPLE APPLICATION middot

(l=Ofce Mode] Wind

~

fcurrent l -middot

~SW ~leldl4

ultimate resistance force I first

design load Illgt

i _

deflection

-

bull

4

Historical Hurrlcane Dariiage and Survival of Platforms

Components Required for Calibration (Petrauskas 1992)

PRIOR Distribution

PLATFORM EXPERIENCE

BAYESIAN Survivals and Failures

UPDATING

Waves Winds and Currents bull

POSTERIOR Distribution

- A_middotJA ~ Survival Data Failure Data

Result Result

Load Resistance

middotfl

-

Bayesian Updaing middotGeneral Description (Petrauskas 1992)

L Imiddot

WHERE TO INTRODUCE BIAS FACTOR (B

On Resistance (Rmue - B (RPREDICreD I

middotOn Loads (Smue middot - B (SPRE01cTEo

On Safety Margin (RSmue - B (RSPREDICTED

CALIBRATION METHODOLOGY

gt

r

FIELD OBSERVATIONS

Observed Behavior

survivals Damages

Failures

r

bull

CAPACITY ANALYSIS

Predicted Load and Resistance Predicted Failure Modes

Survival Damage Failure cases

RELIABILITY ANALYSIS

- Probability of Failure ~

Ukellhood Failure

BAYESIAN UPDATING

Prior Distnbution of B Combined Likelihood Function

Posterior Dlstnbution of B

BIAS FACTOR

Mean Value cov bull

Represents Modeling Uncertainties

Prior Distribution of B Likelihood Functionmiddot Observing Success case STISIK and Posterior Distribution of B

2 00

uo

160

140

120

e 100 L $ uoIshy

0 60 ~ ~ 040

020

0 00

0 00 0 50

_ I 00

b

I 50 2 00

-shy lk [bl STISIK)

--shy Pnor-f(b)

--shy Pos1enor r(bl ISIK)

Prior or B Mean bull 10 COV bull 030

Posterlor of Bbull Meanbull 136COV 0 16

2 50

Likelihood Function and Posterior Distribution Example Platform STISIK

- -

~

- Prior and Posterior Distributions or B Failure cases (ST177B STlSlH ST130A T21) Damage cases (T23 T25 ST161A) Success cases (STlSlK ST130Q ST134W WD90A MC311 MC397)

400

350

300

2 so 0-

I

bull bull

bull bull gt

- bullbull ~ bull bull I bull

_ bull bull bullbull I bull bull bullII V -

bull bull

I bull ~ bull

bullbull bullI bull bullI bull bull bull

middot -~

bull__ ~ middotI middotmiddot

L 2 00

e ~ I SO

-I 00

- 0 so

000 gt

0 00 020 0 40 060 0 80 I 00 120 140 160 110 200

b -

-

shy

Prior-r

- - - - Posterior fLlll failure cases

bull bull bull bull bull bull bull Posterior flall damage cases

- bull bull - bull middot Postenor flall success cases

Posterior flall success fulure damage cases

-

Posterior of B Mean= 119 COV 0101

Posterior Distribution or Blas Factor (8) bull Different Categories and All Cases Combined

shy

l

PHASE I - KEY CONCLUSIONS

bull GLOBAL CORRECTION FACTOR

bull 15 TO 20 CONSERVATISM MEAN)

bull UNEXPECTED SURVIVALS HAVE GREATEST IMPACT

bull DAMAGED CASES MORE INFORMATIVE

bull CALIBRATION ISSUES PHASE II

bull

EFFECT OF IMPROVED CAPACITY ANALYSIS RECIPE

bull APPLICABILITY TO MULTIPLE FAILURE MODES

bull NO FOUNDATION DAMAGE CASE INCLUDED

bull IMPROVED REPRESENTATION OF FULL ANDREW DISTRIBUTION

-bull NUMBER OF PLATFORMS CONSIDERED

I 1 I

I t ) bull ~

PHASE II WORK SCOPE (CALIBRATION)

bull

bull ESTABLISH MULTIPLE BIAS FACTORS

Use Nine Base Case Platforms

middot - Capacity Analysis for Revised recipe

Foundation Blas Factor from Foundation JIP

bull DEVELOP WEIGHTING APPROACH

bull UPDATE MULTIPLE BIAS FACTORS

bullEffect of Platforms Not In the Nine Base Cases

bull EXAMPLE APPLICATION OF BIAS FACTORS I

bullbull

PHASE I PHASE II

CAPACITYbull ANALYSISRu bull

---------------------1----middot

Jc1 - iJbullraquoJ dbull

FttlS frt[f~l ratdhJJ

-

LIKELIHOOD FUNCTION

DEVELOPMENT

lk Jb Jfall1N) bull Jb)

111 Ill Isurvtvoq bull 1 - bl

-middot -------------------shy ~---------------------middot-middot

r111ir8 Cb1 111 Cbl BAYESIAN

UPDATING

~cove

R1R2 bull middot---------------------~----

Jblbull P(g cOJ

1 bull JbRI (MBSI I bull 1

~middot

llcJllJfalluro) bull P(gcOJ Ill 111 b21fallun) bull P ((g1 CO) U 1112 c OD

middot--------------------shy ----------------------shy bull

r111111r8 Cbl 111 Ill i-1nro1

deg ~1b2)laquottfrr (b1b21-lnfo) re Ill bull 1-11middot b2) bull lodegamp

middot

i1 covamp i2 cov~i

-

shy

--

Likelihood Function vs b All Success Case STISI K STl30Q STl34W WD90A MC311 MC397

I 0000

0 9000

OIOOO

0 7000

~ 0amp000

ta osoOO

~ 0 4000

0 3000

0 2000

0 1000

bull 0 0000

bull

I

middot

middot bull I

Ibull

-sibull ~ -middot-middot-middot-middot -middot-

----7----+-----~ I

Ibull bull bull I

rbull

--- Jk ISIK[blsuccess)

- bull - bull - lk 130Q[blsuccess) bull

lk 134W[blsuccess)

Jk-WD90A(blsuccc~I

- - - lk MC397[blsucccss)

--ltgt-- lk middotMC31 l[blsucccss)

--- Jk [bl 6 out of6succcsHascs)

05 I 5 2 250

b

Likelihood Functionsmiddot Success Cases - Phbullsc L

shy

bullbull bull bull bull bull bull -------------------~

bull bull bullbull bullbull bullbull bullbull bullbull bullbull bull

bull bullbull

bullbullbullbull

bullbull

bullbullbullbullbullbullbull

bullbullbullbullbullbull

bullbullbullbullbullbullbull

z 0

t z ) LL

c 0 0c i ~ i

J

~ sect ~--------------~ e l5

I

I

SOFTWARE

bull PF PROGRAM

bull CALREL

bull PF +CALREL

bull RELACS

bull

gt i

l

SENSITIVITY STUDIES

POSTERIOR DISTRIBUTION OF B

bull EFFECT OF CHANGE IN DISTRIBUTION OF PRIOR OF B

bull EFFECT OF REMOVAL OF ONE PLATFORM FROM EACH CATEGORY

bull EFFECT OF REMOVAL OF THREE DAMAGE CASES

bull EFFECT OF ADDITIONAL 13 SIMILAR PLATFORMS

-I

Prior and Posterior Distributions of B Variation In COV of PriorofB fromOl to04 -- -

4500

-4000

-

- -

3500 --middot - bull bull - bull bull Prior-r ICOV =02

3000

- - - - Prior-r ICOV =03 -

bull -~ e 2soo -

bull bull bull bull bull bull bull Pnor-r ICOV =o 4t frJ ~ 2000 Posterior rlCOV =0 2

bull bull Postenor rlCOV =0 3 bull bull 1500 bull middot bull - shy -middot- Posterior rlCOV =04

middot~ I 000 - bull

-middotmiddot ~ ~middotmiddot M -~

- middot 0 500

wbullbullbull _ I middot~ middotmiddotmiddotmiddotmiddot middotmiddotmiddotmiddotmiddot - -~ ~ __ middot_ _ bull -shy0 000

0 000 0200 0 400 0 600 0 800 IOOO 1200 1400 1600 1800 2000 -

- - b

-

=shy

Posterior Distribution of Bias Factor (B) bull Effect of Variation In COV of Prior of B

bull

bullPrior and Posterior Distributions of B Evaluating Effect of Removal of One Platform from Each Category from the Base Case of 13 Platforms

400

~ ~ 350

j IA i

3 00 bull I bull I ---Prior-r I bull l

250 1 I 1

I bull bull ---Posterior r lall 13 casesshy- ~E 2 00 bull bull bull bull bull bull bull Posterior r without STISIK

- ~ 1 bull - bull I - bull bull - bull bull Posterior r Iwithout T2I I 50 I

J-tt~i - bull - bull - Posterior r Iwithout T2S

tOO [-

I I ~ ~ I 0 50

0 00

___ r J i

l~T

~--

0 00 0 20 0 40 0 60 080 I 00 120 140 160 110 2 00

b

zshy

Posterior Distribution of Bias Factor (8) bull Effect of Removal of middotPlatform from Any Category

bullPrior and Posterior Distributions or B Evaluating Effect or Removal or 3 Damage Cases from the Base Case or 13 Platrorms

4

35

3

25

e L 2 ~ bulle shy

I 15

bull I

05

0

0 02 04 06 I 2 I 4 16 IS 2

I I

I

I

I Ibull

J -tmiddot L--

bullI

~

I

fl

II ~

I I v I

middot bull bull

bull L

~ ll- I-~lt

OS I

b

I

bull

1middot bullbullbullbull

~ __

bull

- - - - Prior-r

- - - - Postenor flall failure cases

Postenor flall success cases

--- Posterior flall 13 cases

---Posterior fI success failure cases only

IPosterior or B Mean =t11 COV =0121

-

_I

Posterir Distribution or Bias Factor (B) bull Effect or Removal or 3-Damage Cases

-

bullPrior and Posterior Distributions or B Evaluating Effect or Additional 13 Similar Observations as

Used in Andrew JIP The POsterior or B from Andrew-JIP Considered as New Prior or B

- 500

450

400

350

bull bull

bull bull bull bull J bull

bull bull f ~ I

I ~ Ibull bull

bullbull bull

l bull I -bullbull bull bull

I

I bull bull

bull I I~~bull bull bull

bull ~

I bull bull

~ ~_ ~~~

- - New Prior r bull

Posterior 114 new failure cases ~ 300- bull bull bull bull bull bull bull Posterior f 13 new damage casesl 150 ~

- bull bull - bull bull Posterior 116 new success cases i 2 00

Postenor 1113 new success failure -150 damage cases

I 00

0 50 -

0 00 0 00 0 20 0 40 0 60 0 80 I 00 120 I 40 160 110 2 00

Posterior ol B Mean= 120 COV= 0071 b

shy

shy

_rshy

Posterior Distribution of Bias Factor (B) bull Effect of Additional 13-Similar Observations

1

1 I

WEIGHTING APPROACH bull

bull ASSESSMENT OF PLATFORMS DATA

Select Platforms Most Mappable to the Analyzed Platforms

Unlikely SurvivalDamage Cases Identification

Based on Cut-Off Criteria Agemiddot Hs etc

bull INFORMATION REQUIRED FOR COMPARISON

bull CATEGORIZATION OF ADDITIONALmiddot PLATFORMS

bull UPDATING OF CALIBRATION RESULTS

--

OVERVIEW OF PLATFORMS SURVIVEDDAMAGED -~

Operator -

- Platforms

Survived CS)

Platforms

Damaged CD)

Platforms Used In

Phase IJIP -

JI~ ~Bill~IRBDll -

Amoco -

Chevron -

Exxon

Mob II

Phllllps

Shell

Unocal -

30

143

72 22

7

39 25shy -

1

12

bull

1 bull

1

1(S)+1 (D)

3 (S) +3(D)

2 (S)

bull shy

bull

bull

bull

Total Partlclnants Platforms 338 - 15 10C6S+4D) -

Other Operators

310

13

3 (D)

Total Platforms 648 28 13C6S+7D) -

Summary of Andrew JIP bull Phase I Platforms

Plairshy w v- N-ber Lea-le 1 shy car Hind_ o-w N-bull Deplh fl fl IJpe la bull n Yeadon Jadtet lbelledtf liefalit -- _ -OD

IL H - -shy _ICllshy IL IL - - rQClto ~ _

K STUIK 137 1963 8 365 608S4 Cellardeck

-middot STl-shy 170 1964 4 -~ K 39 62269 Cellardeck - --ST134W 137 1911 4 43 608S4 Smvlnlshy- _ - -

WD90A 184 1964 8 K 267Sbull34 83 SOS31 SUboftllar deck SmvMI

MC311 343 1978 8 x S7 62 - MC397 468 1991 4 x shy S33bull6SO 65911 -shySmeSmvlvol -- gt -

r-- T23ST52 63 1969 4 ton- shyKshy 52594 S03 -T25SSl39 62 1969 4 Kshy SOS94 S063 - middot ton-shy__ -

bullbull a ton--shyST161A 118 1964 8 K - 347S 51495 Cellarclecti

r~

ST1771 142 1965 8 K 39 60167 CeDordeck failme

STISIH 137 1964 8 K 395 608S4 CeDardeck Failme

K ST130A 140 1958 a SS 6095 I lblv Pai1uno - T21ST72 61 1969 4 Kshy S4S94 4974 I lhlv Pai1uno -

- -

JE

INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON

bull WATER DEPTH

bull bull YEAR OF INSTALLATIONDESIGN

bull PLATFORM ORIENTATION

bull PLATFORM LATITUDE AND LONGITUDE

bull KEY SKETCHES OF PLATFORM STRUCTURAL FRAMING

Number of Legs Plies Leg Batter Number of Conductors bull Deck Elevations (Including sub-cellar deck If any) Jacket Framing Elevations Brace type In Vertical Frames Sizes of Primary Members (leg pile vertical frames braces primary horizontals) Joint Details (can gap etc) LegPiie Annulus Grouting Status

bull middotPOST ANDREW INSPECTION REPORT Significant Wave Height Hs (ft)

bull INFORMATION ON DAMAGES PRIOR TO ANDREW

bull RESULTS FROM IN-HOUSE STUDIES IF AVAILABLE

bull ~RIGINAL DESIGN CRITERIA INFORMATION IF AVAILABLE

bull SOIL SHEAR STRENGTH PROFILE IF AVAILABLE

Platforms Identified for Evaluating Therr Likelihood of Falling Under Unexpected Survival Category

OPlRATOR

middotADA

BLOClt

llmucnJllE

NAME

YEAll

INSTAUJD

WA1ER

DBPTll

DISTANCE

JlllOM SHOllE

(MILU)

Bl

(M)

AMOCO 111113

STl11

WDI075

A

bull D

St

17

120

1n

44

20

10 I

CllllVllQfl STlllO c 61 110 21 11 STIUO D Q 1110 21 11

STl134 p Q 137 29 11

ST0151 G 61 137 11

ST0151 I 3 121 11

ST0151 J Q 140 11

STl151 It 137 11

ST0151 PJl001 Q UT 11

ST0151 PROD-2 Q UT 11

ST017 A 3 140 35 11

WDIUT c 65 214 34 10

llXXON STI055 II 51 6T 14 I

STl15 A t3 34

WDI073

WD0073

Amiddotshy Q

1111 1111

22

22

I

I

WDI073 c 1n 22 I

- WDOCl93-shy II

65 1110 2T A 3 194 31 MOBU 111121

SS IOl7

A B

501

50

50

31

T

I

T 6

MtlllPllY SS114 B T 50 T T

PlllllIPS SS0149 A 5 55 33 T

SAMllDAN STln A 107 T T

SBllLL WDl103 A 65 223 2T WDl103 B 65 221 2T WD0104 c 65 221 2T WD0104 D 6T 20 2T

UNOCAL SS020I

SS 020I SSl20I

SS 0209

SS0209

II p

PURll

A

B

3

n 61

103

97

97

95

100

34

34

34

34 34

bullbull bullbullbull

- SS0209 D u 95 34 I

- SS 1209 SSl253

SS0253

G A

c

6T

2

100

165

1T5

34

54

54

bullbull I

I I

I I

SOUTH MARCH

~LAND

N

SOUTH PaTOI ENE ISWltD AREA

MMS Defined Limits ror Offshore Structures Requiring Inspections Following Andrew

t---Jbulln3 T25bull I bullTtl

bull1$1Htl1Abull bull1SIK1779bull

~

middot-DaTA AREA~T

CRANO rsLE

AREA bull1300

bullll4W

=shy

bullPath of Hurricane Andrew with Platforms Used In Calibration

Likelihood Function vs b All Success Case STISI K STIJOQ ST134W WD90A MCJll MCJ97

I 0000

0 9000

0 8000

07000

-Bo6000

tiiii 0 5000 -

04000

0 3000

0 2000

01000

0 0000

0 05 25

ri1rt middot-middot-middot-middot-middot-

I 5 2

b

---11 middot151K(blsu~ccsI

- bull - bull - fk 130Q(bfU~lCSS)

lie bull134W(blsuccc]

lk-WD90A[blsultCS]

- - - lk middotMC397(bfUtlCS]

--o-- lk-MCJll(bbuctcsJ

--- ll (bl 6 out of 6 Mlttcs case]

_

( Likelihood Functions middotSuccess Cases - p - 1 c I

CATEGORIZATION OF ADDITIONAL PLATFORMS

I I I

bull CATEGORIES

Unexpected Survivals

Expected Survivals

bull APPROACH

Identify Corresponding Platform(s) Evaluated In the JIP

Determine Max Wave Height at Platform Location

Evaluate Whether Wave Hit the Deck

middot - Evaluate Likelihood of R and S to Differ from the Base Cases

Review Joint Information

Assess Special Features

bull IDENTIFY NUMBER OF PLATFORMS IN EACH CATEGORY

I

UPDATING OF CALIBRATION RESULTS

bull INCLUDE EFFECT OF ADDITIONAL PLATFORMS ON bullBbull

Consider Increased 1Uncertalntles In R and S

Develop bullNeW Ukellhood Functions for Different Categories

middotLet bullxbull Number of Ukely Unexpected Survival Cases and y Number of Ukely Expected Survival Cases

Determine Combined Likelihood Function lk (b Iadditional platforms) middot

bull REVISED LIKELIHOOD FUNCTION

lk(b I all cases) = lk(b I 9 platforms) middot x lk(b Ix platforms) x lk(b I y platforms)

bull ESTABLISH REVISED BIAS FACTOR B

f8 (b) u f8 (b) lk (b I all cases)

II

t t ~

shy

APPLICATIONS OF BIAS FACTORS

ii APPLICABLE TO PARTICULAR LOAD AND RESISTANCE RECIPE

bull USE BY THE API COMMITTEES AND MMS

Criteria Revisions

Recommended Design Guidelines Revisions bull

bullbull REASSESSMENT STUDIES AND DECISION MAKING

middot

bull RISK ASSESSMENT STUDIES BY THE OPERATORS

Individual Platforms

Fleet of Platforms

middot IMR Program Implementation

bullbull NOT TO BE USED IN DESIGN

I I fl J

EXAMPLE USE OF BIAS FACTORS

bull PHASE I EXAMPLE APPLICATION

bull EVALUATION OF ANNUAL PROBABILITY OF FAILURE

Establish Annual Maximum Seastate Data

Determine Coefficients to Define Base Shear

Determine Ultimate Capacity of Platform

Determine Probablllty of Failure Versus bullbbull

-bull DETERMINE UPDATED PROBABILITY OF FAILURE

Pf - P [S gt R]

Pf I P [~ gt Rb I b] f 8 (b) db

Pr - J P [S gt Rb I b] f bull 8 (b) db

bull TOTAL PROBABILITIES OF FAILURE

Total Pf - 1 bull Il [1 middotPf] - l Pf bull

Total Pf - 1 bull Il [1middot Pf] - l Pfmiddot

Total Pr = 1 - n [1middot PfbullJ = l Pfshy

I I I

EXAMPLE APPLICATION bull PHASE I

PLATFORM ST 151 K

bull GULF OF MEXICO

bull WATER DEPTH = 137 FT

bull ANNUAL MAXIMUM HS MEDIAN 175 ft COV 031

bull ASSUMED STORM DURATION 3 hoursmiddot

bull ASSUMED NUMBER OF WAVES 800

bull MEDIAN HS FOR STORM HOURS 1 2 AND 3

bull MEDIAN CURRENT VALUE bull

bull DETERMINE ULTIMATE CAPACITY FOR THREE DIRECTIONS

bull DETERMINE ANNUAL FAILURE PROBABILITIES

Broadsidemiddot Diagonal End-on

0037 0025 0007

0071 0044 0018

0022 0016 0004

DETERMINE TOTAL ANNUAL PROBABILITY OF FAILURE

Pf - 007

Pf - 013

Pr - 004

bull

EXAMPLE USE OF BIAS FACTORS - PHASE II middot

bull DETERMINE PREDICTIVE PROBABILITY OF FAILURE

Case - 1

Pf _ S S Pf (b b2) fa1a2 (b b2) db db2

g bull -1 (b1 R I S) - 1

92 - (b2 R2 I S) - 1

Case -2

Pf s ~ [bJ f11 (b) db

Pf [bJ p [g lt 0)

g - bull 1 (b1 R1 I S) - 1

1

bull1) l

I

BENEFITS OF ANDREW JIP bull PHASE II

lt

bull IMPROVED LOAD AND RESISTANCE RECIPE

bull IMPROVED CALIBRATION PROCEDURE

bull MULTIPLE BIAS FACTORS

bull PROCEDURE TO INCLUDE EFFECT OF OTHER PLATFORMS

bull PROCEDURE TO USE MULTIPLE BIAS FACTORS

  • Structure Bookmarks
    • ANDREW PHASE II JOINT1 INDUSTRY PROJECT
    • PROGRESS MEETING NUMBER 1
    • ANDREW PHASE II JOINT INDUSTRY PROJECT PROGRESS MEETING NUMBER 1 NOVEMBER 16 AGENDA
    • ANDREW PHASE II UST OF PARTICIPANTS
    • ANDREW PHASE I OBJECTIVES AND RESULTS
    • ANDREW PHASE II PROJECT OBJECTIVES
    • SUBSEQUENT PROJECTS
    • ANDREW PHASE II PROJECT TEAM
    • ANDREW PHASE II UPDATE MODELING middotAND ANALYSIS RECIPE
    • -ANDREW PHASE II CONDUCTOR RESISTANCE
    • ANDREW PHASE Il WA VE FORCE PROFILE
    • ANDREW PHASE II WA VE FORCE PROFILEmiddot
    • ANDREW PHASE II BRACE MODELING
    • ANDREW PHASE II FOUNDATIONS
    • ANDREW PHASE II middot middot JOINT MODELING
    • ANDREW PHASE Il JOINT MODELING OPTIONS DISCUSSION
    • ANDREW PHASE II JOINT MODEJJNG OPTIONS ILLUSTRATIONS
    • ANDREW PHASE II JOINT MODELING PHASE II METHOD
    • ANDREW PHASE II ANDREW STORM CONDmONS UPDATED
    • Platform Selection bullbullbull1111111111111
    • BasisCriteria for Selection bullbullbull1111111111111
    • CALIBRATION
    • WHERE TO INTRODUCE BIAS FACTOR (B
    • PHASE I -KEY CONCLUSIONS
    • bull CALIBRATION ISSUES PHASE II
    • PHASE II WORK SCOPE (CALIBRATION)
    • SOFTWARE
    • SENSITIVITY STUDIES POSTERIOR DISTRIBUTION OF B
    • WEIGHTING APPROACH bull
    • INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON
    • CATEGORIZATION OF ADDITIONAL PLATFORMS
    • UPDATING OF CALIBRATION RESULTS
    • APPLICATIONS OF BIAS FACTORS
    • EXAMPLE USE OF BIAS FACTORS
    • EXAMPLE APPLICATION bull PHASE I PLATFORM ST 151 K
    • EXAMPLE USE OF BIAS FACTORS -PHASE II middot
    • BENEFITS OF ANDREW JIP bull PHASE II
Page 47:  · . ' ANDREW PHASE II PROJECT OBJECTIVES . ' • ·UPDATE MODELING AND ANALYSIS RECIPE· JOINTS ' ' ·BRACES WAVE PROFILE FOUNDATIONS . ' ' CONDUCTORS· • ANALVZE PLATFORMS USING

Platforms Identified for Evaluating Their Likelihood of Falling Under UnexpeCted Survival Category

CllEllATOa

AllB

llLOCll

Sl1lvcnJllE

NAME

YEAR

INSTAUJCD

WATER

IJBPJB

middotmiddotDISTANCE

ftOMSllOllE

(MIUIS)

Ill

(M)

AMOCO BllW

STllI

WDI075

A

bull D

4

n uo 17J

44

32

bull 20

10

bull CllEVllOH STtllO c I 1bull 21 11

STl130 D 0 161 21 11 STllJ4 I 0 137 D - 11

STIUI G I 137 32 11

STIUI I 0 121 32 11

STIUI

J 0 140 32 11

STIUI It 4 137 32 11

STllSI PROl)I 0 137 32 11

STllSI PROD-2 0 137 32 11

STll7 A 0

140 35 11

WDl117 c 65 214 34 10

DXON STshy E 51 14 bull STl165 A t3 bull 34 WD0073

WD0073

WD0073

A _ c

4

62

4

IA IA

17J

22

22

22

bullbullbull WD003-shy E 65 161 Z7

A 0 IN 31 MOBD Bl12I

ssn A bull

50

50

31

7

bull 7

MIJllPllY SS114 B 7 50 7 7

PlllUIPS SSl14t A 55 33 7

SAMEDAN ST17J A 4 117 7 bull 7

SJIELL WD0103 A 65 223 27 WDllOJ bull 65 221 27 WDll04 c 65 221 27 WDOI04 D 67 2 27

UNOCAL

SSl20I

SSl20I

bull SSl20I

SS0209

SS0209

SS0209

SS0209

SS 0253

SS 0253

E

I

PtJllE

A

bull D

G A

c

0

4

4

7J

61

0

0

69

103

97

97

9S

100

95

100

165

175

34

34

34

34

34

34

34

54

54

bullbullbullbullbullbullbullbullbull

Platforms Recommended for Phase II

llalr y_ w- Aot Av- _ I 1 Doplla Model nm bull bull _ bull I I H11 s18-1 Irht lbwl~ bullbullbull(-illtdlel OM lllMtD

II ---

-STISIIC 137 1963 I CAP Yeo Yeo I Yeo

STtshy -170 1964 4 CAP Yeo ST134dlU I Yeo

STl34W 137 Yeo Yeo -1981 4 CAP D Yeo

WD90A 114 1964 I CAP Yeo - D

MC311 30 1971 I KARMA Yeo -sre-

MC397 middot 461 1991 4 ICARMA Yeo Sireshy - - r- -

__ T13ST5l 63 1969 4 CAP Yeo Yeobull middot- ~n- D Yeo _ Tl5SSl39 62 1969 4 CAP Yeo Yeo I Yeo

_ STIT711 142 1965 I CAP Yeo I Yeo middot-1shySTl61A Ill 1964 I USFOS Yeo Yeo - D bulloshy

- r-

STISIH 137 1964 I CAP y D y -_ STl31A 140 1951 I CAP Yeo STl34dlU I Yeo

T21Sll 61 1969 4 CAP No rar-middot I Yeo

~-rmiddot-middot SS209A 95 1967 16 CAP Yeo D -SSU4R

11 UI A 41 -in D SS 2741AIM CAP bull

_ _ -shy - -shy- -shy

CALIBRATION

bull PHASEIAPPROACH

bull CALIBRATION ISSUES

bull PHASE II WORK SCOPE

bull SOFlWARE AVAILABILITY

bull EFFECT OF ADDITIONAL PLATFORMS

bull WEIGHTING APPROACH

bull EXAMPLE APPLICATION middot

(l=Ofce Mode] Wind

~

fcurrent l -middot

~SW ~leldl4

ultimate resistance force I first

design load Illgt

i _

deflection

-

bull

4

Historical Hurrlcane Dariiage and Survival of Platforms

Components Required for Calibration (Petrauskas 1992)

PRIOR Distribution

PLATFORM EXPERIENCE

BAYESIAN Survivals and Failures

UPDATING

Waves Winds and Currents bull

POSTERIOR Distribution

- A_middotJA ~ Survival Data Failure Data

Result Result

Load Resistance

middotfl

-

Bayesian Updaing middotGeneral Description (Petrauskas 1992)

L Imiddot

WHERE TO INTRODUCE BIAS FACTOR (B

On Resistance (Rmue - B (RPREDICreD I

middotOn Loads (Smue middot - B (SPRE01cTEo

On Safety Margin (RSmue - B (RSPREDICTED

CALIBRATION METHODOLOGY

gt

r

FIELD OBSERVATIONS

Observed Behavior

survivals Damages

Failures

r

bull

CAPACITY ANALYSIS

Predicted Load and Resistance Predicted Failure Modes

Survival Damage Failure cases

RELIABILITY ANALYSIS

- Probability of Failure ~

Ukellhood Failure

BAYESIAN UPDATING

Prior Distnbution of B Combined Likelihood Function

Posterior Dlstnbution of B

BIAS FACTOR

Mean Value cov bull

Represents Modeling Uncertainties

Prior Distribution of B Likelihood Functionmiddot Observing Success case STISIK and Posterior Distribution of B

2 00

uo

160

140

120

e 100 L $ uoIshy

0 60 ~ ~ 040

020

0 00

0 00 0 50

_ I 00

b

I 50 2 00

-shy lk [bl STISIK)

--shy Pnor-f(b)

--shy Pos1enor r(bl ISIK)

Prior or B Mean bull 10 COV bull 030

Posterlor of Bbull Meanbull 136COV 0 16

2 50

Likelihood Function and Posterior Distribution Example Platform STISIK

- -

~

- Prior and Posterior Distributions or B Failure cases (ST177B STlSlH ST130A T21) Damage cases (T23 T25 ST161A) Success cases (STlSlK ST130Q ST134W WD90A MC311 MC397)

400

350

300

2 so 0-

I

bull bull

bull bull gt

- bullbull ~ bull bull I bull

_ bull bull bullbull I bull bull bullII V -

bull bull

I bull ~ bull

bullbull bullI bull bullI bull bull bull

middot -~

bull__ ~ middotI middotmiddot

L 2 00

e ~ I SO

-I 00

- 0 so

000 gt

0 00 020 0 40 060 0 80 I 00 120 140 160 110 200

b -

-

shy

Prior-r

- - - - Posterior fLlll failure cases

bull bull bull bull bull bull bull Posterior flall damage cases

- bull bull - bull middot Postenor flall success cases

Posterior flall success fulure damage cases

-

Posterior of B Mean= 119 COV 0101

Posterior Distribution or Blas Factor (8) bull Different Categories and All Cases Combined

shy

l

PHASE I - KEY CONCLUSIONS

bull GLOBAL CORRECTION FACTOR

bull 15 TO 20 CONSERVATISM MEAN)

bull UNEXPECTED SURVIVALS HAVE GREATEST IMPACT

bull DAMAGED CASES MORE INFORMATIVE

bull CALIBRATION ISSUES PHASE II

bull

EFFECT OF IMPROVED CAPACITY ANALYSIS RECIPE

bull APPLICABILITY TO MULTIPLE FAILURE MODES

bull NO FOUNDATION DAMAGE CASE INCLUDED

bull IMPROVED REPRESENTATION OF FULL ANDREW DISTRIBUTION

-bull NUMBER OF PLATFORMS CONSIDERED

I 1 I

I t ) bull ~

PHASE II WORK SCOPE (CALIBRATION)

bull

bull ESTABLISH MULTIPLE BIAS FACTORS

Use Nine Base Case Platforms

middot - Capacity Analysis for Revised recipe

Foundation Blas Factor from Foundation JIP

bull DEVELOP WEIGHTING APPROACH

bull UPDATE MULTIPLE BIAS FACTORS

bullEffect of Platforms Not In the Nine Base Cases

bull EXAMPLE APPLICATION OF BIAS FACTORS I

bullbull

PHASE I PHASE II

CAPACITYbull ANALYSISRu bull

---------------------1----middot

Jc1 - iJbullraquoJ dbull

FttlS frt[f~l ratdhJJ

-

LIKELIHOOD FUNCTION

DEVELOPMENT

lk Jb Jfall1N) bull Jb)

111 Ill Isurvtvoq bull 1 - bl

-middot -------------------shy ~---------------------middot-middot

r111ir8 Cb1 111 Cbl BAYESIAN

UPDATING

~cove

R1R2 bull middot---------------------~----

Jblbull P(g cOJ

1 bull JbRI (MBSI I bull 1

~middot

llcJllJfalluro) bull P(gcOJ Ill 111 b21fallun) bull P ((g1 CO) U 1112 c OD

middot--------------------shy ----------------------shy bull

r111111r8 Cbl 111 Ill i-1nro1

deg ~1b2)laquottfrr (b1b21-lnfo) re Ill bull 1-11middot b2) bull lodegamp

middot

i1 covamp i2 cov~i

-

shy

--

Likelihood Function vs b All Success Case STISI K STl30Q STl34W WD90A MC311 MC397

I 0000

0 9000

OIOOO

0 7000

~ 0amp000

ta osoOO

~ 0 4000

0 3000

0 2000

0 1000

bull 0 0000

bull

I

middot

middot bull I

Ibull

-sibull ~ -middot-middot-middot-middot -middot-

----7----+-----~ I

Ibull bull bull I

rbull

--- Jk ISIK[blsuccess)

- bull - bull - lk 130Q[blsuccess) bull

lk 134W[blsuccess)

Jk-WD90A(blsuccc~I

- - - lk MC397[blsucccss)

--ltgt-- lk middotMC31 l[blsucccss)

--- Jk [bl 6 out of6succcsHascs)

05 I 5 2 250

b

Likelihood Functionsmiddot Success Cases - Phbullsc L

shy

bullbull bull bull bull bull bull -------------------~

bull bull bullbull bullbull bullbull bullbull bullbull bullbull bull

bull bullbull

bullbullbullbull

bullbull

bullbullbullbullbullbullbull

bullbullbullbullbullbull

bullbullbullbullbullbullbull

z 0

t z ) LL

c 0 0c i ~ i

J

~ sect ~--------------~ e l5

I

I

SOFTWARE

bull PF PROGRAM

bull CALREL

bull PF +CALREL

bull RELACS

bull

gt i

l

SENSITIVITY STUDIES

POSTERIOR DISTRIBUTION OF B

bull EFFECT OF CHANGE IN DISTRIBUTION OF PRIOR OF B

bull EFFECT OF REMOVAL OF ONE PLATFORM FROM EACH CATEGORY

bull EFFECT OF REMOVAL OF THREE DAMAGE CASES

bull EFFECT OF ADDITIONAL 13 SIMILAR PLATFORMS

-I

Prior and Posterior Distributions of B Variation In COV of PriorofB fromOl to04 -- -

4500

-4000

-

- -

3500 --middot - bull bull - bull bull Prior-r ICOV =02

3000

- - - - Prior-r ICOV =03 -

bull -~ e 2soo -

bull bull bull bull bull bull bull Pnor-r ICOV =o 4t frJ ~ 2000 Posterior rlCOV =0 2

bull bull Postenor rlCOV =0 3 bull bull 1500 bull middot bull - shy -middot- Posterior rlCOV =04

middot~ I 000 - bull

-middotmiddot ~ ~middotmiddot M -~

- middot 0 500

wbullbullbull _ I middot~ middotmiddotmiddotmiddotmiddot middotmiddotmiddotmiddotmiddot - -~ ~ __ middot_ _ bull -shy0 000

0 000 0200 0 400 0 600 0 800 IOOO 1200 1400 1600 1800 2000 -

- - b

-

=shy

Posterior Distribution of Bias Factor (B) bull Effect of Variation In COV of Prior of B

bull

bullPrior and Posterior Distributions of B Evaluating Effect of Removal of One Platform from Each Category from the Base Case of 13 Platforms

400

~ ~ 350

j IA i

3 00 bull I bull I ---Prior-r I bull l

250 1 I 1

I bull bull ---Posterior r lall 13 casesshy- ~E 2 00 bull bull bull bull bull bull bull Posterior r without STISIK

- ~ 1 bull - bull I - bull bull - bull bull Posterior r Iwithout T2I I 50 I

J-tt~i - bull - bull - Posterior r Iwithout T2S

tOO [-

I I ~ ~ I 0 50

0 00

___ r J i

l~T

~--

0 00 0 20 0 40 0 60 080 I 00 120 140 160 110 2 00

b

zshy

Posterior Distribution of Bias Factor (8) bull Effect of Removal of middotPlatform from Any Category

bullPrior and Posterior Distributions or B Evaluating Effect or Removal or 3 Damage Cases from the Base Case or 13 Platrorms

4

35

3

25

e L 2 ~ bulle shy

I 15

bull I

05

0

0 02 04 06 I 2 I 4 16 IS 2

I I

I

I

I Ibull

J -tmiddot L--

bullI

~

I

fl

II ~

I I v I

middot bull bull

bull L

~ ll- I-~lt

OS I

b

I

bull

1middot bullbullbullbull

~ __

bull

- - - - Prior-r

- - - - Postenor flall failure cases

Postenor flall success cases

--- Posterior flall 13 cases

---Posterior fI success failure cases only

IPosterior or B Mean =t11 COV =0121

-

_I

Posterir Distribution or Bias Factor (B) bull Effect or Removal or 3-Damage Cases

-

bullPrior and Posterior Distributions or B Evaluating Effect or Additional 13 Similar Observations as

Used in Andrew JIP The POsterior or B from Andrew-JIP Considered as New Prior or B

- 500

450

400

350

bull bull

bull bull bull bull J bull

bull bull f ~ I

I ~ Ibull bull

bullbull bull

l bull I -bullbull bull bull

I

I bull bull

bull I I~~bull bull bull

bull ~

I bull bull

~ ~_ ~~~

- - New Prior r bull

Posterior 114 new failure cases ~ 300- bull bull bull bull bull bull bull Posterior f 13 new damage casesl 150 ~

- bull bull - bull bull Posterior 116 new success cases i 2 00

Postenor 1113 new success failure -150 damage cases

I 00

0 50 -

0 00 0 00 0 20 0 40 0 60 0 80 I 00 120 I 40 160 110 2 00

Posterior ol B Mean= 120 COV= 0071 b

shy

shy

_rshy

Posterior Distribution of Bias Factor (B) bull Effect of Additional 13-Similar Observations

1

1 I

WEIGHTING APPROACH bull

bull ASSESSMENT OF PLATFORMS DATA

Select Platforms Most Mappable to the Analyzed Platforms

Unlikely SurvivalDamage Cases Identification

Based on Cut-Off Criteria Agemiddot Hs etc

bull INFORMATION REQUIRED FOR COMPARISON

bull CATEGORIZATION OF ADDITIONALmiddot PLATFORMS

bull UPDATING OF CALIBRATION RESULTS

--

OVERVIEW OF PLATFORMS SURVIVEDDAMAGED -~

Operator -

- Platforms

Survived CS)

Platforms

Damaged CD)

Platforms Used In

Phase IJIP -

JI~ ~Bill~IRBDll -

Amoco -

Chevron -

Exxon

Mob II

Phllllps

Shell

Unocal -

30

143

72 22

7

39 25shy -

1

12

bull

1 bull

1

1(S)+1 (D)

3 (S) +3(D)

2 (S)

bull shy

bull

bull

bull

Total Partlclnants Platforms 338 - 15 10C6S+4D) -

Other Operators

310

13

3 (D)

Total Platforms 648 28 13C6S+7D) -

Summary of Andrew JIP bull Phase I Platforms

Plairshy w v- N-ber Lea-le 1 shy car Hind_ o-w N-bull Deplh fl fl IJpe la bull n Yeadon Jadtet lbelledtf liefalit -- _ -OD

IL H - -shy _ICllshy IL IL - - rQClto ~ _

K STUIK 137 1963 8 365 608S4 Cellardeck

-middot STl-shy 170 1964 4 -~ K 39 62269 Cellardeck - --ST134W 137 1911 4 43 608S4 Smvlnlshy- _ - -

WD90A 184 1964 8 K 267Sbull34 83 SOS31 SUboftllar deck SmvMI

MC311 343 1978 8 x S7 62 - MC397 468 1991 4 x shy S33bull6SO 65911 -shySmeSmvlvol -- gt -

r-- T23ST52 63 1969 4 ton- shyKshy 52594 S03 -T25SSl39 62 1969 4 Kshy SOS94 S063 - middot ton-shy__ -

bullbull a ton--shyST161A 118 1964 8 K - 347S 51495 Cellarclecti

r~

ST1771 142 1965 8 K 39 60167 CeDordeck failme

STISIH 137 1964 8 K 395 608S4 CeDardeck Failme

K ST130A 140 1958 a SS 6095 I lblv Pai1uno - T21ST72 61 1969 4 Kshy S4S94 4974 I lhlv Pai1uno -

- -

JE

INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON

bull WATER DEPTH

bull bull YEAR OF INSTALLATIONDESIGN

bull PLATFORM ORIENTATION

bull PLATFORM LATITUDE AND LONGITUDE

bull KEY SKETCHES OF PLATFORM STRUCTURAL FRAMING

Number of Legs Plies Leg Batter Number of Conductors bull Deck Elevations (Including sub-cellar deck If any) Jacket Framing Elevations Brace type In Vertical Frames Sizes of Primary Members (leg pile vertical frames braces primary horizontals) Joint Details (can gap etc) LegPiie Annulus Grouting Status

bull middotPOST ANDREW INSPECTION REPORT Significant Wave Height Hs (ft)

bull INFORMATION ON DAMAGES PRIOR TO ANDREW

bull RESULTS FROM IN-HOUSE STUDIES IF AVAILABLE

bull ~RIGINAL DESIGN CRITERIA INFORMATION IF AVAILABLE

bull SOIL SHEAR STRENGTH PROFILE IF AVAILABLE

Platforms Identified for Evaluating Therr Likelihood of Falling Under Unexpected Survival Category

OPlRATOR

middotADA

BLOClt

llmucnJllE

NAME

YEAll

INSTAUJD

WA1ER

DBPTll

DISTANCE

JlllOM SHOllE

(MILU)

Bl

(M)

AMOCO 111113

STl11

WDI075

A

bull D

St

17

120

1n

44

20

10 I

CllllVllQfl STlllO c 61 110 21 11 STIUO D Q 1110 21 11

STl134 p Q 137 29 11

ST0151 G 61 137 11

ST0151 I 3 121 11

ST0151 J Q 140 11

STl151 It 137 11

ST0151 PJl001 Q UT 11

ST0151 PROD-2 Q UT 11

ST017 A 3 140 35 11

WDIUT c 65 214 34 10

llXXON STI055 II 51 6T 14 I

STl15 A t3 34

WDI073

WD0073

Amiddotshy Q

1111 1111

22

22

I

I

WDI073 c 1n 22 I

- WDOCl93-shy II

65 1110 2T A 3 194 31 MOBU 111121

SS IOl7

A B

501

50

50

31

T

I

T 6

MtlllPllY SS114 B T 50 T T

PlllllIPS SS0149 A 5 55 33 T

SAMllDAN STln A 107 T T

SBllLL WDl103 A 65 223 2T WDl103 B 65 221 2T WD0104 c 65 221 2T WD0104 D 6T 20 2T

UNOCAL SS020I

SS 020I SSl20I

SS 0209

SS0209

II p

PURll

A

B

3

n 61

103

97

97

95

100

34

34

34

34 34

bullbull bullbullbull

- SS0209 D u 95 34 I

- SS 1209 SSl253

SS0253

G A

c

6T

2

100

165

1T5

34

54

54

bullbull I

I I

I I

SOUTH MARCH

~LAND

N

SOUTH PaTOI ENE ISWltD AREA

MMS Defined Limits ror Offshore Structures Requiring Inspections Following Andrew

t---Jbulln3 T25bull I bullTtl

bull1$1Htl1Abull bull1SIK1779bull

~

middot-DaTA AREA~T

CRANO rsLE

AREA bull1300

bullll4W

=shy

bullPath of Hurricane Andrew with Platforms Used In Calibration

Likelihood Function vs b All Success Case STISI K STIJOQ ST134W WD90A MCJll MCJ97

I 0000

0 9000

0 8000

07000

-Bo6000

tiiii 0 5000 -

04000

0 3000

0 2000

01000

0 0000

0 05 25

ri1rt middot-middot-middot-middot-middot-

I 5 2

b

---11 middot151K(blsu~ccsI

- bull - bull - fk 130Q(bfU~lCSS)

lie bull134W(blsuccc]

lk-WD90A[blsultCS]

- - - lk middotMC397(bfUtlCS]

--o-- lk-MCJll(bbuctcsJ

--- ll (bl 6 out of 6 Mlttcs case]

_

( Likelihood Functions middotSuccess Cases - p - 1 c I

CATEGORIZATION OF ADDITIONAL PLATFORMS

I I I

bull CATEGORIES

Unexpected Survivals

Expected Survivals

bull APPROACH

Identify Corresponding Platform(s) Evaluated In the JIP

Determine Max Wave Height at Platform Location

Evaluate Whether Wave Hit the Deck

middot - Evaluate Likelihood of R and S to Differ from the Base Cases

Review Joint Information

Assess Special Features

bull IDENTIFY NUMBER OF PLATFORMS IN EACH CATEGORY

I

UPDATING OF CALIBRATION RESULTS

bull INCLUDE EFFECT OF ADDITIONAL PLATFORMS ON bullBbull

Consider Increased 1Uncertalntles In R and S

Develop bullNeW Ukellhood Functions for Different Categories

middotLet bullxbull Number of Ukely Unexpected Survival Cases and y Number of Ukely Expected Survival Cases

Determine Combined Likelihood Function lk (b Iadditional platforms) middot

bull REVISED LIKELIHOOD FUNCTION

lk(b I all cases) = lk(b I 9 platforms) middot x lk(b Ix platforms) x lk(b I y platforms)

bull ESTABLISH REVISED BIAS FACTOR B

f8 (b) u f8 (b) lk (b I all cases)

II

t t ~

shy

APPLICATIONS OF BIAS FACTORS

ii APPLICABLE TO PARTICULAR LOAD AND RESISTANCE RECIPE

bull USE BY THE API COMMITTEES AND MMS

Criteria Revisions

Recommended Design Guidelines Revisions bull

bullbull REASSESSMENT STUDIES AND DECISION MAKING

middot

bull RISK ASSESSMENT STUDIES BY THE OPERATORS

Individual Platforms

Fleet of Platforms

middot IMR Program Implementation

bullbull NOT TO BE USED IN DESIGN

I I fl J

EXAMPLE USE OF BIAS FACTORS

bull PHASE I EXAMPLE APPLICATION

bull EVALUATION OF ANNUAL PROBABILITY OF FAILURE

Establish Annual Maximum Seastate Data

Determine Coefficients to Define Base Shear

Determine Ultimate Capacity of Platform

Determine Probablllty of Failure Versus bullbbull

-bull DETERMINE UPDATED PROBABILITY OF FAILURE

Pf - P [S gt R]

Pf I P [~ gt Rb I b] f 8 (b) db

Pr - J P [S gt Rb I b] f bull 8 (b) db

bull TOTAL PROBABILITIES OF FAILURE

Total Pf - 1 bull Il [1 middotPf] - l Pf bull

Total Pf - 1 bull Il [1middot Pf] - l Pfmiddot

Total Pr = 1 - n [1middot PfbullJ = l Pfshy

I I I

EXAMPLE APPLICATION bull PHASE I

PLATFORM ST 151 K

bull GULF OF MEXICO

bull WATER DEPTH = 137 FT

bull ANNUAL MAXIMUM HS MEDIAN 175 ft COV 031

bull ASSUMED STORM DURATION 3 hoursmiddot

bull ASSUMED NUMBER OF WAVES 800

bull MEDIAN HS FOR STORM HOURS 1 2 AND 3

bull MEDIAN CURRENT VALUE bull

bull DETERMINE ULTIMATE CAPACITY FOR THREE DIRECTIONS

bull DETERMINE ANNUAL FAILURE PROBABILITIES

Broadsidemiddot Diagonal End-on

0037 0025 0007

0071 0044 0018

0022 0016 0004

DETERMINE TOTAL ANNUAL PROBABILITY OF FAILURE

Pf - 007

Pf - 013

Pr - 004

bull

EXAMPLE USE OF BIAS FACTORS - PHASE II middot

bull DETERMINE PREDICTIVE PROBABILITY OF FAILURE

Case - 1

Pf _ S S Pf (b b2) fa1a2 (b b2) db db2

g bull -1 (b1 R I S) - 1

92 - (b2 R2 I S) - 1

Case -2

Pf s ~ [bJ f11 (b) db

Pf [bJ p [g lt 0)

g - bull 1 (b1 R1 I S) - 1

1

bull1) l

I

BENEFITS OF ANDREW JIP bull PHASE II

lt

bull IMPROVED LOAD AND RESISTANCE RECIPE

bull IMPROVED CALIBRATION PROCEDURE

bull MULTIPLE BIAS FACTORS

bull PROCEDURE TO INCLUDE EFFECT OF OTHER PLATFORMS

bull PROCEDURE TO USE MULTIPLE BIAS FACTORS

  • Structure Bookmarks
    • ANDREW PHASE II JOINT1 INDUSTRY PROJECT
    • PROGRESS MEETING NUMBER 1
    • ANDREW PHASE II JOINT INDUSTRY PROJECT PROGRESS MEETING NUMBER 1 NOVEMBER 16 AGENDA
    • ANDREW PHASE II UST OF PARTICIPANTS
    • ANDREW PHASE I OBJECTIVES AND RESULTS
    • ANDREW PHASE II PROJECT OBJECTIVES
    • SUBSEQUENT PROJECTS
    • ANDREW PHASE II PROJECT TEAM
    • ANDREW PHASE II UPDATE MODELING middotAND ANALYSIS RECIPE
    • -ANDREW PHASE II CONDUCTOR RESISTANCE
    • ANDREW PHASE Il WA VE FORCE PROFILE
    • ANDREW PHASE II WA VE FORCE PROFILEmiddot
    • ANDREW PHASE II BRACE MODELING
    • ANDREW PHASE II FOUNDATIONS
    • ANDREW PHASE II middot middot JOINT MODELING
    • ANDREW PHASE Il JOINT MODELING OPTIONS DISCUSSION
    • ANDREW PHASE II JOINT MODEJJNG OPTIONS ILLUSTRATIONS
    • ANDREW PHASE II JOINT MODELING PHASE II METHOD
    • ANDREW PHASE II ANDREW STORM CONDmONS UPDATED
    • Platform Selection bullbullbull1111111111111
    • BasisCriteria for Selection bullbullbull1111111111111
    • CALIBRATION
    • WHERE TO INTRODUCE BIAS FACTOR (B
    • PHASE I -KEY CONCLUSIONS
    • bull CALIBRATION ISSUES PHASE II
    • PHASE II WORK SCOPE (CALIBRATION)
    • SOFTWARE
    • SENSITIVITY STUDIES POSTERIOR DISTRIBUTION OF B
    • WEIGHTING APPROACH bull
    • INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON
    • CATEGORIZATION OF ADDITIONAL PLATFORMS
    • UPDATING OF CALIBRATION RESULTS
    • APPLICATIONS OF BIAS FACTORS
    • EXAMPLE USE OF BIAS FACTORS
    • EXAMPLE APPLICATION bull PHASE I PLATFORM ST 151 K
    • EXAMPLE USE OF BIAS FACTORS -PHASE II middot
    • BENEFITS OF ANDREW JIP bull PHASE II
Page 48:  · . ' ANDREW PHASE II PROJECT OBJECTIVES . ' • ·UPDATE MODELING AND ANALYSIS RECIPE· JOINTS ' ' ·BRACES WAVE PROFILE FOUNDATIONS . ' ' CONDUCTORS· • ANALVZE PLATFORMS USING

Platforms Recommended for Phase II

llalr y_ w- Aot Av- _ I 1 Doplla Model nm bull bull _ bull I I H11 s18-1 Irht lbwl~ bullbullbull(-illtdlel OM lllMtD

II ---

-STISIIC 137 1963 I CAP Yeo Yeo I Yeo

STtshy -170 1964 4 CAP Yeo ST134dlU I Yeo

STl34W 137 Yeo Yeo -1981 4 CAP D Yeo

WD90A 114 1964 I CAP Yeo - D

MC311 30 1971 I KARMA Yeo -sre-

MC397 middot 461 1991 4 ICARMA Yeo Sireshy - - r- -

__ T13ST5l 63 1969 4 CAP Yeo Yeobull middot- ~n- D Yeo _ Tl5SSl39 62 1969 4 CAP Yeo Yeo I Yeo

_ STIT711 142 1965 I CAP Yeo I Yeo middot-1shySTl61A Ill 1964 I USFOS Yeo Yeo - D bulloshy

- r-

STISIH 137 1964 I CAP y D y -_ STl31A 140 1951 I CAP Yeo STl34dlU I Yeo

T21Sll 61 1969 4 CAP No rar-middot I Yeo

~-rmiddot-middot SS209A 95 1967 16 CAP Yeo D -SSU4R

11 UI A 41 -in D SS 2741AIM CAP bull

_ _ -shy - -shy- -shy

CALIBRATION

bull PHASEIAPPROACH

bull CALIBRATION ISSUES

bull PHASE II WORK SCOPE

bull SOFlWARE AVAILABILITY

bull EFFECT OF ADDITIONAL PLATFORMS

bull WEIGHTING APPROACH

bull EXAMPLE APPLICATION middot

(l=Ofce Mode] Wind

~

fcurrent l -middot

~SW ~leldl4

ultimate resistance force I first

design load Illgt

i _

deflection

-

bull

4

Historical Hurrlcane Dariiage and Survival of Platforms

Components Required for Calibration (Petrauskas 1992)

PRIOR Distribution

PLATFORM EXPERIENCE

BAYESIAN Survivals and Failures

UPDATING

Waves Winds and Currents bull

POSTERIOR Distribution

- A_middotJA ~ Survival Data Failure Data

Result Result

Load Resistance

middotfl

-

Bayesian Updaing middotGeneral Description (Petrauskas 1992)

L Imiddot

WHERE TO INTRODUCE BIAS FACTOR (B

On Resistance (Rmue - B (RPREDICreD I

middotOn Loads (Smue middot - B (SPRE01cTEo

On Safety Margin (RSmue - B (RSPREDICTED

CALIBRATION METHODOLOGY

gt

r

FIELD OBSERVATIONS

Observed Behavior

survivals Damages

Failures

r

bull

CAPACITY ANALYSIS

Predicted Load and Resistance Predicted Failure Modes

Survival Damage Failure cases

RELIABILITY ANALYSIS

- Probability of Failure ~

Ukellhood Failure

BAYESIAN UPDATING

Prior Distnbution of B Combined Likelihood Function

Posterior Dlstnbution of B

BIAS FACTOR

Mean Value cov bull

Represents Modeling Uncertainties

Prior Distribution of B Likelihood Functionmiddot Observing Success case STISIK and Posterior Distribution of B

2 00

uo

160

140

120

e 100 L $ uoIshy

0 60 ~ ~ 040

020

0 00

0 00 0 50

_ I 00

b

I 50 2 00

-shy lk [bl STISIK)

--shy Pnor-f(b)

--shy Pos1enor r(bl ISIK)

Prior or B Mean bull 10 COV bull 030

Posterlor of Bbull Meanbull 136COV 0 16

2 50

Likelihood Function and Posterior Distribution Example Platform STISIK

- -

~

- Prior and Posterior Distributions or B Failure cases (ST177B STlSlH ST130A T21) Damage cases (T23 T25 ST161A) Success cases (STlSlK ST130Q ST134W WD90A MC311 MC397)

400

350

300

2 so 0-

I

bull bull

bull bull gt

- bullbull ~ bull bull I bull

_ bull bull bullbull I bull bull bullII V -

bull bull

I bull ~ bull

bullbull bullI bull bullI bull bull bull

middot -~

bull__ ~ middotI middotmiddot

L 2 00

e ~ I SO

-I 00

- 0 so

000 gt

0 00 020 0 40 060 0 80 I 00 120 140 160 110 200

b -

-

shy

Prior-r

- - - - Posterior fLlll failure cases

bull bull bull bull bull bull bull Posterior flall damage cases

- bull bull - bull middot Postenor flall success cases

Posterior flall success fulure damage cases

-

Posterior of B Mean= 119 COV 0101

Posterior Distribution or Blas Factor (8) bull Different Categories and All Cases Combined

shy

l

PHASE I - KEY CONCLUSIONS

bull GLOBAL CORRECTION FACTOR

bull 15 TO 20 CONSERVATISM MEAN)

bull UNEXPECTED SURVIVALS HAVE GREATEST IMPACT

bull DAMAGED CASES MORE INFORMATIVE

bull CALIBRATION ISSUES PHASE II

bull

EFFECT OF IMPROVED CAPACITY ANALYSIS RECIPE

bull APPLICABILITY TO MULTIPLE FAILURE MODES

bull NO FOUNDATION DAMAGE CASE INCLUDED

bull IMPROVED REPRESENTATION OF FULL ANDREW DISTRIBUTION

-bull NUMBER OF PLATFORMS CONSIDERED

I 1 I

I t ) bull ~

PHASE II WORK SCOPE (CALIBRATION)

bull

bull ESTABLISH MULTIPLE BIAS FACTORS

Use Nine Base Case Platforms

middot - Capacity Analysis for Revised recipe

Foundation Blas Factor from Foundation JIP

bull DEVELOP WEIGHTING APPROACH

bull UPDATE MULTIPLE BIAS FACTORS

bullEffect of Platforms Not In the Nine Base Cases

bull EXAMPLE APPLICATION OF BIAS FACTORS I

bullbull

PHASE I PHASE II

CAPACITYbull ANALYSISRu bull

---------------------1----middot

Jc1 - iJbullraquoJ dbull

FttlS frt[f~l ratdhJJ

-

LIKELIHOOD FUNCTION

DEVELOPMENT

lk Jb Jfall1N) bull Jb)

111 Ill Isurvtvoq bull 1 - bl

-middot -------------------shy ~---------------------middot-middot

r111ir8 Cb1 111 Cbl BAYESIAN

UPDATING

~cove

R1R2 bull middot---------------------~----

Jblbull P(g cOJ

1 bull JbRI (MBSI I bull 1

~middot

llcJllJfalluro) bull P(gcOJ Ill 111 b21fallun) bull P ((g1 CO) U 1112 c OD

middot--------------------shy ----------------------shy bull

r111111r8 Cbl 111 Ill i-1nro1

deg ~1b2)laquottfrr (b1b21-lnfo) re Ill bull 1-11middot b2) bull lodegamp

middot

i1 covamp i2 cov~i

-

shy

--

Likelihood Function vs b All Success Case STISI K STl30Q STl34W WD90A MC311 MC397

I 0000

0 9000

OIOOO

0 7000

~ 0amp000

ta osoOO

~ 0 4000

0 3000

0 2000

0 1000

bull 0 0000

bull

I

middot

middot bull I

Ibull

-sibull ~ -middot-middot-middot-middot -middot-

----7----+-----~ I

Ibull bull bull I

rbull

--- Jk ISIK[blsuccess)

- bull - bull - lk 130Q[blsuccess) bull

lk 134W[blsuccess)

Jk-WD90A(blsuccc~I

- - - lk MC397[blsucccss)

--ltgt-- lk middotMC31 l[blsucccss)

--- Jk [bl 6 out of6succcsHascs)

05 I 5 2 250

b

Likelihood Functionsmiddot Success Cases - Phbullsc L

shy

bullbull bull bull bull bull bull -------------------~

bull bull bullbull bullbull bullbull bullbull bullbull bullbull bull

bull bullbull

bullbullbullbull

bullbull

bullbullbullbullbullbullbull

bullbullbullbullbullbull

bullbullbullbullbullbullbull

z 0

t z ) LL

c 0 0c i ~ i

J

~ sect ~--------------~ e l5

I

I

SOFTWARE

bull PF PROGRAM

bull CALREL

bull PF +CALREL

bull RELACS

bull

gt i

l

SENSITIVITY STUDIES

POSTERIOR DISTRIBUTION OF B

bull EFFECT OF CHANGE IN DISTRIBUTION OF PRIOR OF B

bull EFFECT OF REMOVAL OF ONE PLATFORM FROM EACH CATEGORY

bull EFFECT OF REMOVAL OF THREE DAMAGE CASES

bull EFFECT OF ADDITIONAL 13 SIMILAR PLATFORMS

-I

Prior and Posterior Distributions of B Variation In COV of PriorofB fromOl to04 -- -

4500

-4000

-

- -

3500 --middot - bull bull - bull bull Prior-r ICOV =02

3000

- - - - Prior-r ICOV =03 -

bull -~ e 2soo -

bull bull bull bull bull bull bull Pnor-r ICOV =o 4t frJ ~ 2000 Posterior rlCOV =0 2

bull bull Postenor rlCOV =0 3 bull bull 1500 bull middot bull - shy -middot- Posterior rlCOV =04

middot~ I 000 - bull

-middotmiddot ~ ~middotmiddot M -~

- middot 0 500

wbullbullbull _ I middot~ middotmiddotmiddotmiddotmiddot middotmiddotmiddotmiddotmiddot - -~ ~ __ middot_ _ bull -shy0 000

0 000 0200 0 400 0 600 0 800 IOOO 1200 1400 1600 1800 2000 -

- - b

-

=shy

Posterior Distribution of Bias Factor (B) bull Effect of Variation In COV of Prior of B

bull

bullPrior and Posterior Distributions of B Evaluating Effect of Removal of One Platform from Each Category from the Base Case of 13 Platforms

400

~ ~ 350

j IA i

3 00 bull I bull I ---Prior-r I bull l

250 1 I 1

I bull bull ---Posterior r lall 13 casesshy- ~E 2 00 bull bull bull bull bull bull bull Posterior r without STISIK

- ~ 1 bull - bull I - bull bull - bull bull Posterior r Iwithout T2I I 50 I

J-tt~i - bull - bull - Posterior r Iwithout T2S

tOO [-

I I ~ ~ I 0 50

0 00

___ r J i

l~T

~--

0 00 0 20 0 40 0 60 080 I 00 120 140 160 110 2 00

b

zshy

Posterior Distribution of Bias Factor (8) bull Effect of Removal of middotPlatform from Any Category

bullPrior and Posterior Distributions or B Evaluating Effect or Removal or 3 Damage Cases from the Base Case or 13 Platrorms

4

35

3

25

e L 2 ~ bulle shy

I 15

bull I

05

0

0 02 04 06 I 2 I 4 16 IS 2

I I

I

I

I Ibull

J -tmiddot L--

bullI

~

I

fl

II ~

I I v I

middot bull bull

bull L

~ ll- I-~lt

OS I

b

I

bull

1middot bullbullbullbull

~ __

bull

- - - - Prior-r

- - - - Postenor flall failure cases

Postenor flall success cases

--- Posterior flall 13 cases

---Posterior fI success failure cases only

IPosterior or B Mean =t11 COV =0121

-

_I

Posterir Distribution or Bias Factor (B) bull Effect or Removal or 3-Damage Cases

-

bullPrior and Posterior Distributions or B Evaluating Effect or Additional 13 Similar Observations as

Used in Andrew JIP The POsterior or B from Andrew-JIP Considered as New Prior or B

- 500

450

400

350

bull bull

bull bull bull bull J bull

bull bull f ~ I

I ~ Ibull bull

bullbull bull

l bull I -bullbull bull bull

I

I bull bull

bull I I~~bull bull bull

bull ~

I bull bull

~ ~_ ~~~

- - New Prior r bull

Posterior 114 new failure cases ~ 300- bull bull bull bull bull bull bull Posterior f 13 new damage casesl 150 ~

- bull bull - bull bull Posterior 116 new success cases i 2 00

Postenor 1113 new success failure -150 damage cases

I 00

0 50 -

0 00 0 00 0 20 0 40 0 60 0 80 I 00 120 I 40 160 110 2 00

Posterior ol B Mean= 120 COV= 0071 b

shy

shy

_rshy

Posterior Distribution of Bias Factor (B) bull Effect of Additional 13-Similar Observations

1

1 I

WEIGHTING APPROACH bull

bull ASSESSMENT OF PLATFORMS DATA

Select Platforms Most Mappable to the Analyzed Platforms

Unlikely SurvivalDamage Cases Identification

Based on Cut-Off Criteria Agemiddot Hs etc

bull INFORMATION REQUIRED FOR COMPARISON

bull CATEGORIZATION OF ADDITIONALmiddot PLATFORMS

bull UPDATING OF CALIBRATION RESULTS

--

OVERVIEW OF PLATFORMS SURVIVEDDAMAGED -~

Operator -

- Platforms

Survived CS)

Platforms

Damaged CD)

Platforms Used In

Phase IJIP -

JI~ ~Bill~IRBDll -

Amoco -

Chevron -

Exxon

Mob II

Phllllps

Shell

Unocal -

30

143

72 22

7

39 25shy -

1

12

bull

1 bull

1

1(S)+1 (D)

3 (S) +3(D)

2 (S)

bull shy

bull

bull

bull

Total Partlclnants Platforms 338 - 15 10C6S+4D) -

Other Operators

310

13

3 (D)

Total Platforms 648 28 13C6S+7D) -

Summary of Andrew JIP bull Phase I Platforms

Plairshy w v- N-ber Lea-le 1 shy car Hind_ o-w N-bull Deplh fl fl IJpe la bull n Yeadon Jadtet lbelledtf liefalit -- _ -OD

IL H - -shy _ICllshy IL IL - - rQClto ~ _

K STUIK 137 1963 8 365 608S4 Cellardeck

-middot STl-shy 170 1964 4 -~ K 39 62269 Cellardeck - --ST134W 137 1911 4 43 608S4 Smvlnlshy- _ - -

WD90A 184 1964 8 K 267Sbull34 83 SOS31 SUboftllar deck SmvMI

MC311 343 1978 8 x S7 62 - MC397 468 1991 4 x shy S33bull6SO 65911 -shySmeSmvlvol -- gt -

r-- T23ST52 63 1969 4 ton- shyKshy 52594 S03 -T25SSl39 62 1969 4 Kshy SOS94 S063 - middot ton-shy__ -

bullbull a ton--shyST161A 118 1964 8 K - 347S 51495 Cellarclecti

r~

ST1771 142 1965 8 K 39 60167 CeDordeck failme

STISIH 137 1964 8 K 395 608S4 CeDardeck Failme

K ST130A 140 1958 a SS 6095 I lblv Pai1uno - T21ST72 61 1969 4 Kshy S4S94 4974 I lhlv Pai1uno -

- -

JE

INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON

bull WATER DEPTH

bull bull YEAR OF INSTALLATIONDESIGN

bull PLATFORM ORIENTATION

bull PLATFORM LATITUDE AND LONGITUDE

bull KEY SKETCHES OF PLATFORM STRUCTURAL FRAMING

Number of Legs Plies Leg Batter Number of Conductors bull Deck Elevations (Including sub-cellar deck If any) Jacket Framing Elevations Brace type In Vertical Frames Sizes of Primary Members (leg pile vertical frames braces primary horizontals) Joint Details (can gap etc) LegPiie Annulus Grouting Status

bull middotPOST ANDREW INSPECTION REPORT Significant Wave Height Hs (ft)

bull INFORMATION ON DAMAGES PRIOR TO ANDREW

bull RESULTS FROM IN-HOUSE STUDIES IF AVAILABLE

bull ~RIGINAL DESIGN CRITERIA INFORMATION IF AVAILABLE

bull SOIL SHEAR STRENGTH PROFILE IF AVAILABLE

Platforms Identified for Evaluating Therr Likelihood of Falling Under Unexpected Survival Category

OPlRATOR

middotADA

BLOClt

llmucnJllE

NAME

YEAll

INSTAUJD

WA1ER

DBPTll

DISTANCE

JlllOM SHOllE

(MILU)

Bl

(M)

AMOCO 111113

STl11

WDI075

A

bull D

St

17

120

1n

44

20

10 I

CllllVllQfl STlllO c 61 110 21 11 STIUO D Q 1110 21 11

STl134 p Q 137 29 11

ST0151 G 61 137 11

ST0151 I 3 121 11

ST0151 J Q 140 11

STl151 It 137 11

ST0151 PJl001 Q UT 11

ST0151 PROD-2 Q UT 11

ST017 A 3 140 35 11

WDIUT c 65 214 34 10

llXXON STI055 II 51 6T 14 I

STl15 A t3 34

WDI073

WD0073

Amiddotshy Q

1111 1111

22

22

I

I

WDI073 c 1n 22 I

- WDOCl93-shy II

65 1110 2T A 3 194 31 MOBU 111121

SS IOl7

A B

501

50

50

31

T

I

T 6

MtlllPllY SS114 B T 50 T T

PlllllIPS SS0149 A 5 55 33 T

SAMllDAN STln A 107 T T

SBllLL WDl103 A 65 223 2T WDl103 B 65 221 2T WD0104 c 65 221 2T WD0104 D 6T 20 2T

UNOCAL SS020I

SS 020I SSl20I

SS 0209

SS0209

II p

PURll

A

B

3

n 61

103

97

97

95

100

34

34

34

34 34

bullbull bullbullbull

- SS0209 D u 95 34 I

- SS 1209 SSl253

SS0253

G A

c

6T

2

100

165

1T5

34

54

54

bullbull I

I I

I I

SOUTH MARCH

~LAND

N

SOUTH PaTOI ENE ISWltD AREA

MMS Defined Limits ror Offshore Structures Requiring Inspections Following Andrew

t---Jbulln3 T25bull I bullTtl

bull1$1Htl1Abull bull1SIK1779bull

~

middot-DaTA AREA~T

CRANO rsLE

AREA bull1300

bullll4W

=shy

bullPath of Hurricane Andrew with Platforms Used In Calibration

Likelihood Function vs b All Success Case STISI K STIJOQ ST134W WD90A MCJll MCJ97

I 0000

0 9000

0 8000

07000

-Bo6000

tiiii 0 5000 -

04000

0 3000

0 2000

01000

0 0000

0 05 25

ri1rt middot-middot-middot-middot-middot-

I 5 2

b

---11 middot151K(blsu~ccsI

- bull - bull - fk 130Q(bfU~lCSS)

lie bull134W(blsuccc]

lk-WD90A[blsultCS]

- - - lk middotMC397(bfUtlCS]

--o-- lk-MCJll(bbuctcsJ

--- ll (bl 6 out of 6 Mlttcs case]

_

( Likelihood Functions middotSuccess Cases - p - 1 c I

CATEGORIZATION OF ADDITIONAL PLATFORMS

I I I

bull CATEGORIES

Unexpected Survivals

Expected Survivals

bull APPROACH

Identify Corresponding Platform(s) Evaluated In the JIP

Determine Max Wave Height at Platform Location

Evaluate Whether Wave Hit the Deck

middot - Evaluate Likelihood of R and S to Differ from the Base Cases

Review Joint Information

Assess Special Features

bull IDENTIFY NUMBER OF PLATFORMS IN EACH CATEGORY

I

UPDATING OF CALIBRATION RESULTS

bull INCLUDE EFFECT OF ADDITIONAL PLATFORMS ON bullBbull

Consider Increased 1Uncertalntles In R and S

Develop bullNeW Ukellhood Functions for Different Categories

middotLet bullxbull Number of Ukely Unexpected Survival Cases and y Number of Ukely Expected Survival Cases

Determine Combined Likelihood Function lk (b Iadditional platforms) middot

bull REVISED LIKELIHOOD FUNCTION

lk(b I all cases) = lk(b I 9 platforms) middot x lk(b Ix platforms) x lk(b I y platforms)

bull ESTABLISH REVISED BIAS FACTOR B

f8 (b) u f8 (b) lk (b I all cases)

II

t t ~

shy

APPLICATIONS OF BIAS FACTORS

ii APPLICABLE TO PARTICULAR LOAD AND RESISTANCE RECIPE

bull USE BY THE API COMMITTEES AND MMS

Criteria Revisions

Recommended Design Guidelines Revisions bull

bullbull REASSESSMENT STUDIES AND DECISION MAKING

middot

bull RISK ASSESSMENT STUDIES BY THE OPERATORS

Individual Platforms

Fleet of Platforms

middot IMR Program Implementation

bullbull NOT TO BE USED IN DESIGN

I I fl J

EXAMPLE USE OF BIAS FACTORS

bull PHASE I EXAMPLE APPLICATION

bull EVALUATION OF ANNUAL PROBABILITY OF FAILURE

Establish Annual Maximum Seastate Data

Determine Coefficients to Define Base Shear

Determine Ultimate Capacity of Platform

Determine Probablllty of Failure Versus bullbbull

-bull DETERMINE UPDATED PROBABILITY OF FAILURE

Pf - P [S gt R]

Pf I P [~ gt Rb I b] f 8 (b) db

Pr - J P [S gt Rb I b] f bull 8 (b) db

bull TOTAL PROBABILITIES OF FAILURE

Total Pf - 1 bull Il [1 middotPf] - l Pf bull

Total Pf - 1 bull Il [1middot Pf] - l Pfmiddot

Total Pr = 1 - n [1middot PfbullJ = l Pfshy

I I I

EXAMPLE APPLICATION bull PHASE I

PLATFORM ST 151 K

bull GULF OF MEXICO

bull WATER DEPTH = 137 FT

bull ANNUAL MAXIMUM HS MEDIAN 175 ft COV 031

bull ASSUMED STORM DURATION 3 hoursmiddot

bull ASSUMED NUMBER OF WAVES 800

bull MEDIAN HS FOR STORM HOURS 1 2 AND 3

bull MEDIAN CURRENT VALUE bull

bull DETERMINE ULTIMATE CAPACITY FOR THREE DIRECTIONS

bull DETERMINE ANNUAL FAILURE PROBABILITIES

Broadsidemiddot Diagonal End-on

0037 0025 0007

0071 0044 0018

0022 0016 0004

DETERMINE TOTAL ANNUAL PROBABILITY OF FAILURE

Pf - 007

Pf - 013

Pr - 004

bull

EXAMPLE USE OF BIAS FACTORS - PHASE II middot

bull DETERMINE PREDICTIVE PROBABILITY OF FAILURE

Case - 1

Pf _ S S Pf (b b2) fa1a2 (b b2) db db2

g bull -1 (b1 R I S) - 1

92 - (b2 R2 I S) - 1

Case -2

Pf s ~ [bJ f11 (b) db

Pf [bJ p [g lt 0)

g - bull 1 (b1 R1 I S) - 1

1

bull1) l

I

BENEFITS OF ANDREW JIP bull PHASE II

lt

bull IMPROVED LOAD AND RESISTANCE RECIPE

bull IMPROVED CALIBRATION PROCEDURE

bull MULTIPLE BIAS FACTORS

bull PROCEDURE TO INCLUDE EFFECT OF OTHER PLATFORMS

bull PROCEDURE TO USE MULTIPLE BIAS FACTORS

  • Structure Bookmarks
    • ANDREW PHASE II JOINT1 INDUSTRY PROJECT
    • PROGRESS MEETING NUMBER 1
    • ANDREW PHASE II JOINT INDUSTRY PROJECT PROGRESS MEETING NUMBER 1 NOVEMBER 16 AGENDA
    • ANDREW PHASE II UST OF PARTICIPANTS
    • ANDREW PHASE I OBJECTIVES AND RESULTS
    • ANDREW PHASE II PROJECT OBJECTIVES
    • SUBSEQUENT PROJECTS
    • ANDREW PHASE II PROJECT TEAM
    • ANDREW PHASE II UPDATE MODELING middotAND ANALYSIS RECIPE
    • -ANDREW PHASE II CONDUCTOR RESISTANCE
    • ANDREW PHASE Il WA VE FORCE PROFILE
    • ANDREW PHASE II WA VE FORCE PROFILEmiddot
    • ANDREW PHASE II BRACE MODELING
    • ANDREW PHASE II FOUNDATIONS
    • ANDREW PHASE II middot middot JOINT MODELING
    • ANDREW PHASE Il JOINT MODELING OPTIONS DISCUSSION
    • ANDREW PHASE II JOINT MODEJJNG OPTIONS ILLUSTRATIONS
    • ANDREW PHASE II JOINT MODELING PHASE II METHOD
    • ANDREW PHASE II ANDREW STORM CONDmONS UPDATED
    • Platform Selection bullbullbull1111111111111
    • BasisCriteria for Selection bullbullbull1111111111111
    • CALIBRATION
    • WHERE TO INTRODUCE BIAS FACTOR (B
    • PHASE I -KEY CONCLUSIONS
    • bull CALIBRATION ISSUES PHASE II
    • PHASE II WORK SCOPE (CALIBRATION)
    • SOFTWARE
    • SENSITIVITY STUDIES POSTERIOR DISTRIBUTION OF B
    • WEIGHTING APPROACH bull
    • INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON
    • CATEGORIZATION OF ADDITIONAL PLATFORMS
    • UPDATING OF CALIBRATION RESULTS
    • APPLICATIONS OF BIAS FACTORS
    • EXAMPLE USE OF BIAS FACTORS
    • EXAMPLE APPLICATION bull PHASE I PLATFORM ST 151 K
    • EXAMPLE USE OF BIAS FACTORS -PHASE II middot
    • BENEFITS OF ANDREW JIP bull PHASE II
Page 49:  · . ' ANDREW PHASE II PROJECT OBJECTIVES . ' • ·UPDATE MODELING AND ANALYSIS RECIPE· JOINTS ' ' ·BRACES WAVE PROFILE FOUNDATIONS . ' ' CONDUCTORS· • ANALVZE PLATFORMS USING

CALIBRATION

bull PHASEIAPPROACH

bull CALIBRATION ISSUES

bull PHASE II WORK SCOPE

bull SOFlWARE AVAILABILITY

bull EFFECT OF ADDITIONAL PLATFORMS

bull WEIGHTING APPROACH

bull EXAMPLE APPLICATION middot

(l=Ofce Mode] Wind

~

fcurrent l -middot

~SW ~leldl4

ultimate resistance force I first

design load Illgt

i _

deflection

-

bull

4

Historical Hurrlcane Dariiage and Survival of Platforms

Components Required for Calibration (Petrauskas 1992)

PRIOR Distribution

PLATFORM EXPERIENCE

BAYESIAN Survivals and Failures

UPDATING

Waves Winds and Currents bull

POSTERIOR Distribution

- A_middotJA ~ Survival Data Failure Data

Result Result

Load Resistance

middotfl

-

Bayesian Updaing middotGeneral Description (Petrauskas 1992)

L Imiddot

WHERE TO INTRODUCE BIAS FACTOR (B

On Resistance (Rmue - B (RPREDICreD I

middotOn Loads (Smue middot - B (SPRE01cTEo

On Safety Margin (RSmue - B (RSPREDICTED

CALIBRATION METHODOLOGY

gt

r

FIELD OBSERVATIONS

Observed Behavior

survivals Damages

Failures

r

bull

CAPACITY ANALYSIS

Predicted Load and Resistance Predicted Failure Modes

Survival Damage Failure cases

RELIABILITY ANALYSIS

- Probability of Failure ~

Ukellhood Failure

BAYESIAN UPDATING

Prior Distnbution of B Combined Likelihood Function

Posterior Dlstnbution of B

BIAS FACTOR

Mean Value cov bull

Represents Modeling Uncertainties

Prior Distribution of B Likelihood Functionmiddot Observing Success case STISIK and Posterior Distribution of B

2 00

uo

160

140

120

e 100 L $ uoIshy

0 60 ~ ~ 040

020

0 00

0 00 0 50

_ I 00

b

I 50 2 00

-shy lk [bl STISIK)

--shy Pnor-f(b)

--shy Pos1enor r(bl ISIK)

Prior or B Mean bull 10 COV bull 030

Posterlor of Bbull Meanbull 136COV 0 16

2 50

Likelihood Function and Posterior Distribution Example Platform STISIK

- -

~

- Prior and Posterior Distributions or B Failure cases (ST177B STlSlH ST130A T21) Damage cases (T23 T25 ST161A) Success cases (STlSlK ST130Q ST134W WD90A MC311 MC397)

400

350

300

2 so 0-

I

bull bull

bull bull gt

- bullbull ~ bull bull I bull

_ bull bull bullbull I bull bull bullII V -

bull bull

I bull ~ bull

bullbull bullI bull bullI bull bull bull

middot -~

bull__ ~ middotI middotmiddot

L 2 00

e ~ I SO

-I 00

- 0 so

000 gt

0 00 020 0 40 060 0 80 I 00 120 140 160 110 200

b -

-

shy

Prior-r

- - - - Posterior fLlll failure cases

bull bull bull bull bull bull bull Posterior flall damage cases

- bull bull - bull middot Postenor flall success cases

Posterior flall success fulure damage cases

-

Posterior of B Mean= 119 COV 0101

Posterior Distribution or Blas Factor (8) bull Different Categories and All Cases Combined

shy

l

PHASE I - KEY CONCLUSIONS

bull GLOBAL CORRECTION FACTOR

bull 15 TO 20 CONSERVATISM MEAN)

bull UNEXPECTED SURVIVALS HAVE GREATEST IMPACT

bull DAMAGED CASES MORE INFORMATIVE

bull CALIBRATION ISSUES PHASE II

bull

EFFECT OF IMPROVED CAPACITY ANALYSIS RECIPE

bull APPLICABILITY TO MULTIPLE FAILURE MODES

bull NO FOUNDATION DAMAGE CASE INCLUDED

bull IMPROVED REPRESENTATION OF FULL ANDREW DISTRIBUTION

-bull NUMBER OF PLATFORMS CONSIDERED

I 1 I

I t ) bull ~

PHASE II WORK SCOPE (CALIBRATION)

bull

bull ESTABLISH MULTIPLE BIAS FACTORS

Use Nine Base Case Platforms

middot - Capacity Analysis for Revised recipe

Foundation Blas Factor from Foundation JIP

bull DEVELOP WEIGHTING APPROACH

bull UPDATE MULTIPLE BIAS FACTORS

bullEffect of Platforms Not In the Nine Base Cases

bull EXAMPLE APPLICATION OF BIAS FACTORS I

bullbull

PHASE I PHASE II

CAPACITYbull ANALYSISRu bull

---------------------1----middot

Jc1 - iJbullraquoJ dbull

FttlS frt[f~l ratdhJJ

-

LIKELIHOOD FUNCTION

DEVELOPMENT

lk Jb Jfall1N) bull Jb)

111 Ill Isurvtvoq bull 1 - bl

-middot -------------------shy ~---------------------middot-middot

r111ir8 Cb1 111 Cbl BAYESIAN

UPDATING

~cove

R1R2 bull middot---------------------~----

Jblbull P(g cOJ

1 bull JbRI (MBSI I bull 1

~middot

llcJllJfalluro) bull P(gcOJ Ill 111 b21fallun) bull P ((g1 CO) U 1112 c OD

middot--------------------shy ----------------------shy bull

r111111r8 Cbl 111 Ill i-1nro1

deg ~1b2)laquottfrr (b1b21-lnfo) re Ill bull 1-11middot b2) bull lodegamp

middot

i1 covamp i2 cov~i

-

shy

--

Likelihood Function vs b All Success Case STISI K STl30Q STl34W WD90A MC311 MC397

I 0000

0 9000

OIOOO

0 7000

~ 0amp000

ta osoOO

~ 0 4000

0 3000

0 2000

0 1000

bull 0 0000

bull

I

middot

middot bull I

Ibull

-sibull ~ -middot-middot-middot-middot -middot-

----7----+-----~ I

Ibull bull bull I

rbull

--- Jk ISIK[blsuccess)

- bull - bull - lk 130Q[blsuccess) bull

lk 134W[blsuccess)

Jk-WD90A(blsuccc~I

- - - lk MC397[blsucccss)

--ltgt-- lk middotMC31 l[blsucccss)

--- Jk [bl 6 out of6succcsHascs)

05 I 5 2 250

b

Likelihood Functionsmiddot Success Cases - Phbullsc L

shy

bullbull bull bull bull bull bull -------------------~

bull bull bullbull bullbull bullbull bullbull bullbull bullbull bull

bull bullbull

bullbullbullbull

bullbull

bullbullbullbullbullbullbull

bullbullbullbullbullbull

bullbullbullbullbullbullbull

z 0

t z ) LL

c 0 0c i ~ i

J

~ sect ~--------------~ e l5

I

I

SOFTWARE

bull PF PROGRAM

bull CALREL

bull PF +CALREL

bull RELACS

bull

gt i

l

SENSITIVITY STUDIES

POSTERIOR DISTRIBUTION OF B

bull EFFECT OF CHANGE IN DISTRIBUTION OF PRIOR OF B

bull EFFECT OF REMOVAL OF ONE PLATFORM FROM EACH CATEGORY

bull EFFECT OF REMOVAL OF THREE DAMAGE CASES

bull EFFECT OF ADDITIONAL 13 SIMILAR PLATFORMS

-I

Prior and Posterior Distributions of B Variation In COV of PriorofB fromOl to04 -- -

4500

-4000

-

- -

3500 --middot - bull bull - bull bull Prior-r ICOV =02

3000

- - - - Prior-r ICOV =03 -

bull -~ e 2soo -

bull bull bull bull bull bull bull Pnor-r ICOV =o 4t frJ ~ 2000 Posterior rlCOV =0 2

bull bull Postenor rlCOV =0 3 bull bull 1500 bull middot bull - shy -middot- Posterior rlCOV =04

middot~ I 000 - bull

-middotmiddot ~ ~middotmiddot M -~

- middot 0 500

wbullbullbull _ I middot~ middotmiddotmiddotmiddotmiddot middotmiddotmiddotmiddotmiddot - -~ ~ __ middot_ _ bull -shy0 000

0 000 0200 0 400 0 600 0 800 IOOO 1200 1400 1600 1800 2000 -

- - b

-

=shy

Posterior Distribution of Bias Factor (B) bull Effect of Variation In COV of Prior of B

bull

bullPrior and Posterior Distributions of B Evaluating Effect of Removal of One Platform from Each Category from the Base Case of 13 Platforms

400

~ ~ 350

j IA i

3 00 bull I bull I ---Prior-r I bull l

250 1 I 1

I bull bull ---Posterior r lall 13 casesshy- ~E 2 00 bull bull bull bull bull bull bull Posterior r without STISIK

- ~ 1 bull - bull I - bull bull - bull bull Posterior r Iwithout T2I I 50 I

J-tt~i - bull - bull - Posterior r Iwithout T2S

tOO [-

I I ~ ~ I 0 50

0 00

___ r J i

l~T

~--

0 00 0 20 0 40 0 60 080 I 00 120 140 160 110 2 00

b

zshy

Posterior Distribution of Bias Factor (8) bull Effect of Removal of middotPlatform from Any Category

bullPrior and Posterior Distributions or B Evaluating Effect or Removal or 3 Damage Cases from the Base Case or 13 Platrorms

4

35

3

25

e L 2 ~ bulle shy

I 15

bull I

05

0

0 02 04 06 I 2 I 4 16 IS 2

I I

I

I

I Ibull

J -tmiddot L--

bullI

~

I

fl

II ~

I I v I

middot bull bull

bull L

~ ll- I-~lt

OS I

b

I

bull

1middot bullbullbullbull

~ __

bull

- - - - Prior-r

- - - - Postenor flall failure cases

Postenor flall success cases

--- Posterior flall 13 cases

---Posterior fI success failure cases only

IPosterior or B Mean =t11 COV =0121

-

_I

Posterir Distribution or Bias Factor (B) bull Effect or Removal or 3-Damage Cases

-

bullPrior and Posterior Distributions or B Evaluating Effect or Additional 13 Similar Observations as

Used in Andrew JIP The POsterior or B from Andrew-JIP Considered as New Prior or B

- 500

450

400

350

bull bull

bull bull bull bull J bull

bull bull f ~ I

I ~ Ibull bull

bullbull bull

l bull I -bullbull bull bull

I

I bull bull

bull I I~~bull bull bull

bull ~

I bull bull

~ ~_ ~~~

- - New Prior r bull

Posterior 114 new failure cases ~ 300- bull bull bull bull bull bull bull Posterior f 13 new damage casesl 150 ~

- bull bull - bull bull Posterior 116 new success cases i 2 00

Postenor 1113 new success failure -150 damage cases

I 00

0 50 -

0 00 0 00 0 20 0 40 0 60 0 80 I 00 120 I 40 160 110 2 00

Posterior ol B Mean= 120 COV= 0071 b

shy

shy

_rshy

Posterior Distribution of Bias Factor (B) bull Effect of Additional 13-Similar Observations

1

1 I

WEIGHTING APPROACH bull

bull ASSESSMENT OF PLATFORMS DATA

Select Platforms Most Mappable to the Analyzed Platforms

Unlikely SurvivalDamage Cases Identification

Based on Cut-Off Criteria Agemiddot Hs etc

bull INFORMATION REQUIRED FOR COMPARISON

bull CATEGORIZATION OF ADDITIONALmiddot PLATFORMS

bull UPDATING OF CALIBRATION RESULTS

--

OVERVIEW OF PLATFORMS SURVIVEDDAMAGED -~

Operator -

- Platforms

Survived CS)

Platforms

Damaged CD)

Platforms Used In

Phase IJIP -

JI~ ~Bill~IRBDll -

Amoco -

Chevron -

Exxon

Mob II

Phllllps

Shell

Unocal -

30

143

72 22

7

39 25shy -

1

12

bull

1 bull

1

1(S)+1 (D)

3 (S) +3(D)

2 (S)

bull shy

bull

bull

bull

Total Partlclnants Platforms 338 - 15 10C6S+4D) -

Other Operators

310

13

3 (D)

Total Platforms 648 28 13C6S+7D) -

Summary of Andrew JIP bull Phase I Platforms

Plairshy w v- N-ber Lea-le 1 shy car Hind_ o-w N-bull Deplh fl fl IJpe la bull n Yeadon Jadtet lbelledtf liefalit -- _ -OD

IL H - -shy _ICllshy IL IL - - rQClto ~ _

K STUIK 137 1963 8 365 608S4 Cellardeck

-middot STl-shy 170 1964 4 -~ K 39 62269 Cellardeck - --ST134W 137 1911 4 43 608S4 Smvlnlshy- _ - -

WD90A 184 1964 8 K 267Sbull34 83 SOS31 SUboftllar deck SmvMI

MC311 343 1978 8 x S7 62 - MC397 468 1991 4 x shy S33bull6SO 65911 -shySmeSmvlvol -- gt -

r-- T23ST52 63 1969 4 ton- shyKshy 52594 S03 -T25SSl39 62 1969 4 Kshy SOS94 S063 - middot ton-shy__ -

bullbull a ton--shyST161A 118 1964 8 K - 347S 51495 Cellarclecti

r~

ST1771 142 1965 8 K 39 60167 CeDordeck failme

STISIH 137 1964 8 K 395 608S4 CeDardeck Failme

K ST130A 140 1958 a SS 6095 I lblv Pai1uno - T21ST72 61 1969 4 Kshy S4S94 4974 I lhlv Pai1uno -

- -

JE

INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON

bull WATER DEPTH

bull bull YEAR OF INSTALLATIONDESIGN

bull PLATFORM ORIENTATION

bull PLATFORM LATITUDE AND LONGITUDE

bull KEY SKETCHES OF PLATFORM STRUCTURAL FRAMING

Number of Legs Plies Leg Batter Number of Conductors bull Deck Elevations (Including sub-cellar deck If any) Jacket Framing Elevations Brace type In Vertical Frames Sizes of Primary Members (leg pile vertical frames braces primary horizontals) Joint Details (can gap etc) LegPiie Annulus Grouting Status

bull middotPOST ANDREW INSPECTION REPORT Significant Wave Height Hs (ft)

bull INFORMATION ON DAMAGES PRIOR TO ANDREW

bull RESULTS FROM IN-HOUSE STUDIES IF AVAILABLE

bull ~RIGINAL DESIGN CRITERIA INFORMATION IF AVAILABLE

bull SOIL SHEAR STRENGTH PROFILE IF AVAILABLE

Platforms Identified for Evaluating Therr Likelihood of Falling Under Unexpected Survival Category

OPlRATOR

middotADA

BLOClt

llmucnJllE

NAME

YEAll

INSTAUJD

WA1ER

DBPTll

DISTANCE

JlllOM SHOllE

(MILU)

Bl

(M)

AMOCO 111113

STl11

WDI075

A

bull D

St

17

120

1n

44

20

10 I

CllllVllQfl STlllO c 61 110 21 11 STIUO D Q 1110 21 11

STl134 p Q 137 29 11

ST0151 G 61 137 11

ST0151 I 3 121 11

ST0151 J Q 140 11

STl151 It 137 11

ST0151 PJl001 Q UT 11

ST0151 PROD-2 Q UT 11

ST017 A 3 140 35 11

WDIUT c 65 214 34 10

llXXON STI055 II 51 6T 14 I

STl15 A t3 34

WDI073

WD0073

Amiddotshy Q

1111 1111

22

22

I

I

WDI073 c 1n 22 I

- WDOCl93-shy II

65 1110 2T A 3 194 31 MOBU 111121

SS IOl7

A B

501

50

50

31

T

I

T 6

MtlllPllY SS114 B T 50 T T

PlllllIPS SS0149 A 5 55 33 T

SAMllDAN STln A 107 T T

SBllLL WDl103 A 65 223 2T WDl103 B 65 221 2T WD0104 c 65 221 2T WD0104 D 6T 20 2T

UNOCAL SS020I

SS 020I SSl20I

SS 0209

SS0209

II p

PURll

A

B

3

n 61

103

97

97

95

100

34

34

34

34 34

bullbull bullbullbull

- SS0209 D u 95 34 I

- SS 1209 SSl253

SS0253

G A

c

6T

2

100

165

1T5

34

54

54

bullbull I

I I

I I

SOUTH MARCH

~LAND

N

SOUTH PaTOI ENE ISWltD AREA

MMS Defined Limits ror Offshore Structures Requiring Inspections Following Andrew

t---Jbulln3 T25bull I bullTtl

bull1$1Htl1Abull bull1SIK1779bull

~

middot-DaTA AREA~T

CRANO rsLE

AREA bull1300

bullll4W

=shy

bullPath of Hurricane Andrew with Platforms Used In Calibration

Likelihood Function vs b All Success Case STISI K STIJOQ ST134W WD90A MCJll MCJ97

I 0000

0 9000

0 8000

07000

-Bo6000

tiiii 0 5000 -

04000

0 3000

0 2000

01000

0 0000

0 05 25

ri1rt middot-middot-middot-middot-middot-

I 5 2

b

---11 middot151K(blsu~ccsI

- bull - bull - fk 130Q(bfU~lCSS)

lie bull134W(blsuccc]

lk-WD90A[blsultCS]

- - - lk middotMC397(bfUtlCS]

--o-- lk-MCJll(bbuctcsJ

--- ll (bl 6 out of 6 Mlttcs case]

_

( Likelihood Functions middotSuccess Cases - p - 1 c I

CATEGORIZATION OF ADDITIONAL PLATFORMS

I I I

bull CATEGORIES

Unexpected Survivals

Expected Survivals

bull APPROACH

Identify Corresponding Platform(s) Evaluated In the JIP

Determine Max Wave Height at Platform Location

Evaluate Whether Wave Hit the Deck

middot - Evaluate Likelihood of R and S to Differ from the Base Cases

Review Joint Information

Assess Special Features

bull IDENTIFY NUMBER OF PLATFORMS IN EACH CATEGORY

I

UPDATING OF CALIBRATION RESULTS

bull INCLUDE EFFECT OF ADDITIONAL PLATFORMS ON bullBbull

Consider Increased 1Uncertalntles In R and S

Develop bullNeW Ukellhood Functions for Different Categories

middotLet bullxbull Number of Ukely Unexpected Survival Cases and y Number of Ukely Expected Survival Cases

Determine Combined Likelihood Function lk (b Iadditional platforms) middot

bull REVISED LIKELIHOOD FUNCTION

lk(b I all cases) = lk(b I 9 platforms) middot x lk(b Ix platforms) x lk(b I y platforms)

bull ESTABLISH REVISED BIAS FACTOR B

f8 (b) u f8 (b) lk (b I all cases)

II

t t ~

shy

APPLICATIONS OF BIAS FACTORS

ii APPLICABLE TO PARTICULAR LOAD AND RESISTANCE RECIPE

bull USE BY THE API COMMITTEES AND MMS

Criteria Revisions

Recommended Design Guidelines Revisions bull

bullbull REASSESSMENT STUDIES AND DECISION MAKING

middot

bull RISK ASSESSMENT STUDIES BY THE OPERATORS

Individual Platforms

Fleet of Platforms

middot IMR Program Implementation

bullbull NOT TO BE USED IN DESIGN

I I fl J

EXAMPLE USE OF BIAS FACTORS

bull PHASE I EXAMPLE APPLICATION

bull EVALUATION OF ANNUAL PROBABILITY OF FAILURE

Establish Annual Maximum Seastate Data

Determine Coefficients to Define Base Shear

Determine Ultimate Capacity of Platform

Determine Probablllty of Failure Versus bullbbull

-bull DETERMINE UPDATED PROBABILITY OF FAILURE

Pf - P [S gt R]

Pf I P [~ gt Rb I b] f 8 (b) db

Pr - J P [S gt Rb I b] f bull 8 (b) db

bull TOTAL PROBABILITIES OF FAILURE

Total Pf - 1 bull Il [1 middotPf] - l Pf bull

Total Pf - 1 bull Il [1middot Pf] - l Pfmiddot

Total Pr = 1 - n [1middot PfbullJ = l Pfshy

I I I

EXAMPLE APPLICATION bull PHASE I

PLATFORM ST 151 K

bull GULF OF MEXICO

bull WATER DEPTH = 137 FT

bull ANNUAL MAXIMUM HS MEDIAN 175 ft COV 031

bull ASSUMED STORM DURATION 3 hoursmiddot

bull ASSUMED NUMBER OF WAVES 800

bull MEDIAN HS FOR STORM HOURS 1 2 AND 3

bull MEDIAN CURRENT VALUE bull

bull DETERMINE ULTIMATE CAPACITY FOR THREE DIRECTIONS

bull DETERMINE ANNUAL FAILURE PROBABILITIES

Broadsidemiddot Diagonal End-on

0037 0025 0007

0071 0044 0018

0022 0016 0004

DETERMINE TOTAL ANNUAL PROBABILITY OF FAILURE

Pf - 007

Pf - 013

Pr - 004

bull

EXAMPLE USE OF BIAS FACTORS - PHASE II middot

bull DETERMINE PREDICTIVE PROBABILITY OF FAILURE

Case - 1

Pf _ S S Pf (b b2) fa1a2 (b b2) db db2

g bull -1 (b1 R I S) - 1

92 - (b2 R2 I S) - 1

Case -2

Pf s ~ [bJ f11 (b) db

Pf [bJ p [g lt 0)

g - bull 1 (b1 R1 I S) - 1

1

bull1) l

I

BENEFITS OF ANDREW JIP bull PHASE II

lt

bull IMPROVED LOAD AND RESISTANCE RECIPE

bull IMPROVED CALIBRATION PROCEDURE

bull MULTIPLE BIAS FACTORS

bull PROCEDURE TO INCLUDE EFFECT OF OTHER PLATFORMS

bull PROCEDURE TO USE MULTIPLE BIAS FACTORS

  • Structure Bookmarks
    • ANDREW PHASE II JOINT1 INDUSTRY PROJECT
    • PROGRESS MEETING NUMBER 1
    • ANDREW PHASE II JOINT INDUSTRY PROJECT PROGRESS MEETING NUMBER 1 NOVEMBER 16 AGENDA
    • ANDREW PHASE II UST OF PARTICIPANTS
    • ANDREW PHASE I OBJECTIVES AND RESULTS
    • ANDREW PHASE II PROJECT OBJECTIVES
    • SUBSEQUENT PROJECTS
    • ANDREW PHASE II PROJECT TEAM
    • ANDREW PHASE II UPDATE MODELING middotAND ANALYSIS RECIPE
    • -ANDREW PHASE II CONDUCTOR RESISTANCE
    • ANDREW PHASE Il WA VE FORCE PROFILE
    • ANDREW PHASE II WA VE FORCE PROFILEmiddot
    • ANDREW PHASE II BRACE MODELING
    • ANDREW PHASE II FOUNDATIONS
    • ANDREW PHASE II middot middot JOINT MODELING
    • ANDREW PHASE Il JOINT MODELING OPTIONS DISCUSSION
    • ANDREW PHASE II JOINT MODEJJNG OPTIONS ILLUSTRATIONS
    • ANDREW PHASE II JOINT MODELING PHASE II METHOD
    • ANDREW PHASE II ANDREW STORM CONDmONS UPDATED
    • Platform Selection bullbullbull1111111111111
    • BasisCriteria for Selection bullbullbull1111111111111
    • CALIBRATION
    • WHERE TO INTRODUCE BIAS FACTOR (B
    • PHASE I -KEY CONCLUSIONS
    • bull CALIBRATION ISSUES PHASE II
    • PHASE II WORK SCOPE (CALIBRATION)
    • SOFTWARE
    • SENSITIVITY STUDIES POSTERIOR DISTRIBUTION OF B
    • WEIGHTING APPROACH bull
    • INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON
    • CATEGORIZATION OF ADDITIONAL PLATFORMS
    • UPDATING OF CALIBRATION RESULTS
    • APPLICATIONS OF BIAS FACTORS
    • EXAMPLE USE OF BIAS FACTORS
    • EXAMPLE APPLICATION bull PHASE I PLATFORM ST 151 K
    • EXAMPLE USE OF BIAS FACTORS -PHASE II middot
    • BENEFITS OF ANDREW JIP bull PHASE II
Page 50:  · . ' ANDREW PHASE II PROJECT OBJECTIVES . ' • ·UPDATE MODELING AND ANALYSIS RECIPE· JOINTS ' ' ·BRACES WAVE PROFILE FOUNDATIONS . ' ' CONDUCTORS· • ANALVZE PLATFORMS USING

(l=Ofce Mode] Wind

~

fcurrent l -middot

~SW ~leldl4

ultimate resistance force I first

design load Illgt

i _

deflection

-

bull

4

Historical Hurrlcane Dariiage and Survival of Platforms

Components Required for Calibration (Petrauskas 1992)

PRIOR Distribution

PLATFORM EXPERIENCE

BAYESIAN Survivals and Failures

UPDATING

Waves Winds and Currents bull

POSTERIOR Distribution

- A_middotJA ~ Survival Data Failure Data

Result Result

Load Resistance

middotfl

-

Bayesian Updaing middotGeneral Description (Petrauskas 1992)

L Imiddot

WHERE TO INTRODUCE BIAS FACTOR (B

On Resistance (Rmue - B (RPREDICreD I

middotOn Loads (Smue middot - B (SPRE01cTEo

On Safety Margin (RSmue - B (RSPREDICTED

CALIBRATION METHODOLOGY

gt

r

FIELD OBSERVATIONS

Observed Behavior

survivals Damages

Failures

r

bull

CAPACITY ANALYSIS

Predicted Load and Resistance Predicted Failure Modes

Survival Damage Failure cases

RELIABILITY ANALYSIS

- Probability of Failure ~

Ukellhood Failure

BAYESIAN UPDATING

Prior Distnbution of B Combined Likelihood Function

Posterior Dlstnbution of B

BIAS FACTOR

Mean Value cov bull

Represents Modeling Uncertainties

Prior Distribution of B Likelihood Functionmiddot Observing Success case STISIK and Posterior Distribution of B

2 00

uo

160

140

120

e 100 L $ uoIshy

0 60 ~ ~ 040

020

0 00

0 00 0 50

_ I 00

b

I 50 2 00

-shy lk [bl STISIK)

--shy Pnor-f(b)

--shy Pos1enor r(bl ISIK)

Prior or B Mean bull 10 COV bull 030

Posterlor of Bbull Meanbull 136COV 0 16

2 50

Likelihood Function and Posterior Distribution Example Platform STISIK

- -

~

- Prior and Posterior Distributions or B Failure cases (ST177B STlSlH ST130A T21) Damage cases (T23 T25 ST161A) Success cases (STlSlK ST130Q ST134W WD90A MC311 MC397)

400

350

300

2 so 0-

I

bull bull

bull bull gt

- bullbull ~ bull bull I bull

_ bull bull bullbull I bull bull bullII V -

bull bull

I bull ~ bull

bullbull bullI bull bullI bull bull bull

middot -~

bull__ ~ middotI middotmiddot

L 2 00

e ~ I SO

-I 00

- 0 so

000 gt

0 00 020 0 40 060 0 80 I 00 120 140 160 110 200

b -

-

shy

Prior-r

- - - - Posterior fLlll failure cases

bull bull bull bull bull bull bull Posterior flall damage cases

- bull bull - bull middot Postenor flall success cases

Posterior flall success fulure damage cases

-

Posterior of B Mean= 119 COV 0101

Posterior Distribution or Blas Factor (8) bull Different Categories and All Cases Combined

shy

l

PHASE I - KEY CONCLUSIONS

bull GLOBAL CORRECTION FACTOR

bull 15 TO 20 CONSERVATISM MEAN)

bull UNEXPECTED SURVIVALS HAVE GREATEST IMPACT

bull DAMAGED CASES MORE INFORMATIVE

bull CALIBRATION ISSUES PHASE II

bull

EFFECT OF IMPROVED CAPACITY ANALYSIS RECIPE

bull APPLICABILITY TO MULTIPLE FAILURE MODES

bull NO FOUNDATION DAMAGE CASE INCLUDED

bull IMPROVED REPRESENTATION OF FULL ANDREW DISTRIBUTION

-bull NUMBER OF PLATFORMS CONSIDERED

I 1 I

I t ) bull ~

PHASE II WORK SCOPE (CALIBRATION)

bull

bull ESTABLISH MULTIPLE BIAS FACTORS

Use Nine Base Case Platforms

middot - Capacity Analysis for Revised recipe

Foundation Blas Factor from Foundation JIP

bull DEVELOP WEIGHTING APPROACH

bull UPDATE MULTIPLE BIAS FACTORS

bullEffect of Platforms Not In the Nine Base Cases

bull EXAMPLE APPLICATION OF BIAS FACTORS I

bullbull

PHASE I PHASE II

CAPACITYbull ANALYSISRu bull

---------------------1----middot

Jc1 - iJbullraquoJ dbull

FttlS frt[f~l ratdhJJ

-

LIKELIHOOD FUNCTION

DEVELOPMENT

lk Jb Jfall1N) bull Jb)

111 Ill Isurvtvoq bull 1 - bl

-middot -------------------shy ~---------------------middot-middot

r111ir8 Cb1 111 Cbl BAYESIAN

UPDATING

~cove

R1R2 bull middot---------------------~----

Jblbull P(g cOJ

1 bull JbRI (MBSI I bull 1

~middot

llcJllJfalluro) bull P(gcOJ Ill 111 b21fallun) bull P ((g1 CO) U 1112 c OD

middot--------------------shy ----------------------shy bull

r111111r8 Cbl 111 Ill i-1nro1

deg ~1b2)laquottfrr (b1b21-lnfo) re Ill bull 1-11middot b2) bull lodegamp

middot

i1 covamp i2 cov~i

-

shy

--

Likelihood Function vs b All Success Case STISI K STl30Q STl34W WD90A MC311 MC397

I 0000

0 9000

OIOOO

0 7000

~ 0amp000

ta osoOO

~ 0 4000

0 3000

0 2000

0 1000

bull 0 0000

bull

I

middot

middot bull I

Ibull

-sibull ~ -middot-middot-middot-middot -middot-

----7----+-----~ I

Ibull bull bull I

rbull

--- Jk ISIK[blsuccess)

- bull - bull - lk 130Q[blsuccess) bull

lk 134W[blsuccess)

Jk-WD90A(blsuccc~I

- - - lk MC397[blsucccss)

--ltgt-- lk middotMC31 l[blsucccss)

--- Jk [bl 6 out of6succcsHascs)

05 I 5 2 250

b

Likelihood Functionsmiddot Success Cases - Phbullsc L

shy

bullbull bull bull bull bull bull -------------------~

bull bull bullbull bullbull bullbull bullbull bullbull bullbull bull

bull bullbull

bullbullbullbull

bullbull

bullbullbullbullbullbullbull

bullbullbullbullbullbull

bullbullbullbullbullbullbull

z 0

t z ) LL

c 0 0c i ~ i

J

~ sect ~--------------~ e l5

I

I

SOFTWARE

bull PF PROGRAM

bull CALREL

bull PF +CALREL

bull RELACS

bull

gt i

l

SENSITIVITY STUDIES

POSTERIOR DISTRIBUTION OF B

bull EFFECT OF CHANGE IN DISTRIBUTION OF PRIOR OF B

bull EFFECT OF REMOVAL OF ONE PLATFORM FROM EACH CATEGORY

bull EFFECT OF REMOVAL OF THREE DAMAGE CASES

bull EFFECT OF ADDITIONAL 13 SIMILAR PLATFORMS

-I

Prior and Posterior Distributions of B Variation In COV of PriorofB fromOl to04 -- -

4500

-4000

-

- -

3500 --middot - bull bull - bull bull Prior-r ICOV =02

3000

- - - - Prior-r ICOV =03 -

bull -~ e 2soo -

bull bull bull bull bull bull bull Pnor-r ICOV =o 4t frJ ~ 2000 Posterior rlCOV =0 2

bull bull Postenor rlCOV =0 3 bull bull 1500 bull middot bull - shy -middot- Posterior rlCOV =04

middot~ I 000 - bull

-middotmiddot ~ ~middotmiddot M -~

- middot 0 500

wbullbullbull _ I middot~ middotmiddotmiddotmiddotmiddot middotmiddotmiddotmiddotmiddot - -~ ~ __ middot_ _ bull -shy0 000

0 000 0200 0 400 0 600 0 800 IOOO 1200 1400 1600 1800 2000 -

- - b

-

=shy

Posterior Distribution of Bias Factor (B) bull Effect of Variation In COV of Prior of B

bull

bullPrior and Posterior Distributions of B Evaluating Effect of Removal of One Platform from Each Category from the Base Case of 13 Platforms

400

~ ~ 350

j IA i

3 00 bull I bull I ---Prior-r I bull l

250 1 I 1

I bull bull ---Posterior r lall 13 casesshy- ~E 2 00 bull bull bull bull bull bull bull Posterior r without STISIK

- ~ 1 bull - bull I - bull bull - bull bull Posterior r Iwithout T2I I 50 I

J-tt~i - bull - bull - Posterior r Iwithout T2S

tOO [-

I I ~ ~ I 0 50

0 00

___ r J i

l~T

~--

0 00 0 20 0 40 0 60 080 I 00 120 140 160 110 2 00

b

zshy

Posterior Distribution of Bias Factor (8) bull Effect of Removal of middotPlatform from Any Category

bullPrior and Posterior Distributions or B Evaluating Effect or Removal or 3 Damage Cases from the Base Case or 13 Platrorms

4

35

3

25

e L 2 ~ bulle shy

I 15

bull I

05

0

0 02 04 06 I 2 I 4 16 IS 2

I I

I

I

I Ibull

J -tmiddot L--

bullI

~

I

fl

II ~

I I v I

middot bull bull

bull L

~ ll- I-~lt

OS I

b

I

bull

1middot bullbullbullbull

~ __

bull

- - - - Prior-r

- - - - Postenor flall failure cases

Postenor flall success cases

--- Posterior flall 13 cases

---Posterior fI success failure cases only

IPosterior or B Mean =t11 COV =0121

-

_I

Posterir Distribution or Bias Factor (B) bull Effect or Removal or 3-Damage Cases

-

bullPrior and Posterior Distributions or B Evaluating Effect or Additional 13 Similar Observations as

Used in Andrew JIP The POsterior or B from Andrew-JIP Considered as New Prior or B

- 500

450

400

350

bull bull

bull bull bull bull J bull

bull bull f ~ I

I ~ Ibull bull

bullbull bull

l bull I -bullbull bull bull

I

I bull bull

bull I I~~bull bull bull

bull ~

I bull bull

~ ~_ ~~~

- - New Prior r bull

Posterior 114 new failure cases ~ 300- bull bull bull bull bull bull bull Posterior f 13 new damage casesl 150 ~

- bull bull - bull bull Posterior 116 new success cases i 2 00

Postenor 1113 new success failure -150 damage cases

I 00

0 50 -

0 00 0 00 0 20 0 40 0 60 0 80 I 00 120 I 40 160 110 2 00

Posterior ol B Mean= 120 COV= 0071 b

shy

shy

_rshy

Posterior Distribution of Bias Factor (B) bull Effect of Additional 13-Similar Observations

1

1 I

WEIGHTING APPROACH bull

bull ASSESSMENT OF PLATFORMS DATA

Select Platforms Most Mappable to the Analyzed Platforms

Unlikely SurvivalDamage Cases Identification

Based on Cut-Off Criteria Agemiddot Hs etc

bull INFORMATION REQUIRED FOR COMPARISON

bull CATEGORIZATION OF ADDITIONALmiddot PLATFORMS

bull UPDATING OF CALIBRATION RESULTS

--

OVERVIEW OF PLATFORMS SURVIVEDDAMAGED -~

Operator -

- Platforms

Survived CS)

Platforms

Damaged CD)

Platforms Used In

Phase IJIP -

JI~ ~Bill~IRBDll -

Amoco -

Chevron -

Exxon

Mob II

Phllllps

Shell

Unocal -

30

143

72 22

7

39 25shy -

1

12

bull

1 bull

1

1(S)+1 (D)

3 (S) +3(D)

2 (S)

bull shy

bull

bull

bull

Total Partlclnants Platforms 338 - 15 10C6S+4D) -

Other Operators

310

13

3 (D)

Total Platforms 648 28 13C6S+7D) -

Summary of Andrew JIP bull Phase I Platforms

Plairshy w v- N-ber Lea-le 1 shy car Hind_ o-w N-bull Deplh fl fl IJpe la bull n Yeadon Jadtet lbelledtf liefalit -- _ -OD

IL H - -shy _ICllshy IL IL - - rQClto ~ _

K STUIK 137 1963 8 365 608S4 Cellardeck

-middot STl-shy 170 1964 4 -~ K 39 62269 Cellardeck - --ST134W 137 1911 4 43 608S4 Smvlnlshy- _ - -

WD90A 184 1964 8 K 267Sbull34 83 SOS31 SUboftllar deck SmvMI

MC311 343 1978 8 x S7 62 - MC397 468 1991 4 x shy S33bull6SO 65911 -shySmeSmvlvol -- gt -

r-- T23ST52 63 1969 4 ton- shyKshy 52594 S03 -T25SSl39 62 1969 4 Kshy SOS94 S063 - middot ton-shy__ -

bullbull a ton--shyST161A 118 1964 8 K - 347S 51495 Cellarclecti

r~

ST1771 142 1965 8 K 39 60167 CeDordeck failme

STISIH 137 1964 8 K 395 608S4 CeDardeck Failme

K ST130A 140 1958 a SS 6095 I lblv Pai1uno - T21ST72 61 1969 4 Kshy S4S94 4974 I lhlv Pai1uno -

- -

JE

INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON

bull WATER DEPTH

bull bull YEAR OF INSTALLATIONDESIGN

bull PLATFORM ORIENTATION

bull PLATFORM LATITUDE AND LONGITUDE

bull KEY SKETCHES OF PLATFORM STRUCTURAL FRAMING

Number of Legs Plies Leg Batter Number of Conductors bull Deck Elevations (Including sub-cellar deck If any) Jacket Framing Elevations Brace type In Vertical Frames Sizes of Primary Members (leg pile vertical frames braces primary horizontals) Joint Details (can gap etc) LegPiie Annulus Grouting Status

bull middotPOST ANDREW INSPECTION REPORT Significant Wave Height Hs (ft)

bull INFORMATION ON DAMAGES PRIOR TO ANDREW

bull RESULTS FROM IN-HOUSE STUDIES IF AVAILABLE

bull ~RIGINAL DESIGN CRITERIA INFORMATION IF AVAILABLE

bull SOIL SHEAR STRENGTH PROFILE IF AVAILABLE

Platforms Identified for Evaluating Therr Likelihood of Falling Under Unexpected Survival Category

OPlRATOR

middotADA

BLOClt

llmucnJllE

NAME

YEAll

INSTAUJD

WA1ER

DBPTll

DISTANCE

JlllOM SHOllE

(MILU)

Bl

(M)

AMOCO 111113

STl11

WDI075

A

bull D

St

17

120

1n

44

20

10 I

CllllVllQfl STlllO c 61 110 21 11 STIUO D Q 1110 21 11

STl134 p Q 137 29 11

ST0151 G 61 137 11

ST0151 I 3 121 11

ST0151 J Q 140 11

STl151 It 137 11

ST0151 PJl001 Q UT 11

ST0151 PROD-2 Q UT 11

ST017 A 3 140 35 11

WDIUT c 65 214 34 10

llXXON STI055 II 51 6T 14 I

STl15 A t3 34

WDI073

WD0073

Amiddotshy Q

1111 1111

22

22

I

I

WDI073 c 1n 22 I

- WDOCl93-shy II

65 1110 2T A 3 194 31 MOBU 111121

SS IOl7

A B

501

50

50

31

T

I

T 6

MtlllPllY SS114 B T 50 T T

PlllllIPS SS0149 A 5 55 33 T

SAMllDAN STln A 107 T T

SBllLL WDl103 A 65 223 2T WDl103 B 65 221 2T WD0104 c 65 221 2T WD0104 D 6T 20 2T

UNOCAL SS020I

SS 020I SSl20I

SS 0209

SS0209

II p

PURll

A

B

3

n 61

103

97

97

95

100

34

34

34

34 34

bullbull bullbullbull

- SS0209 D u 95 34 I

- SS 1209 SSl253

SS0253

G A

c

6T

2

100

165

1T5

34

54

54

bullbull I

I I

I I

SOUTH MARCH

~LAND

N

SOUTH PaTOI ENE ISWltD AREA

MMS Defined Limits ror Offshore Structures Requiring Inspections Following Andrew

t---Jbulln3 T25bull I bullTtl

bull1$1Htl1Abull bull1SIK1779bull

~

middot-DaTA AREA~T

CRANO rsLE

AREA bull1300

bullll4W

=shy

bullPath of Hurricane Andrew with Platforms Used In Calibration

Likelihood Function vs b All Success Case STISI K STIJOQ ST134W WD90A MCJll MCJ97

I 0000

0 9000

0 8000

07000

-Bo6000

tiiii 0 5000 -

04000

0 3000

0 2000

01000

0 0000

0 05 25

ri1rt middot-middot-middot-middot-middot-

I 5 2

b

---11 middot151K(blsu~ccsI

- bull - bull - fk 130Q(bfU~lCSS)

lie bull134W(blsuccc]

lk-WD90A[blsultCS]

- - - lk middotMC397(bfUtlCS]

--o-- lk-MCJll(bbuctcsJ

--- ll (bl 6 out of 6 Mlttcs case]

_

( Likelihood Functions middotSuccess Cases - p - 1 c I

CATEGORIZATION OF ADDITIONAL PLATFORMS

I I I

bull CATEGORIES

Unexpected Survivals

Expected Survivals

bull APPROACH

Identify Corresponding Platform(s) Evaluated In the JIP

Determine Max Wave Height at Platform Location

Evaluate Whether Wave Hit the Deck

middot - Evaluate Likelihood of R and S to Differ from the Base Cases

Review Joint Information

Assess Special Features

bull IDENTIFY NUMBER OF PLATFORMS IN EACH CATEGORY

I

UPDATING OF CALIBRATION RESULTS

bull INCLUDE EFFECT OF ADDITIONAL PLATFORMS ON bullBbull

Consider Increased 1Uncertalntles In R and S

Develop bullNeW Ukellhood Functions for Different Categories

middotLet bullxbull Number of Ukely Unexpected Survival Cases and y Number of Ukely Expected Survival Cases

Determine Combined Likelihood Function lk (b Iadditional platforms) middot

bull REVISED LIKELIHOOD FUNCTION

lk(b I all cases) = lk(b I 9 platforms) middot x lk(b Ix platforms) x lk(b I y platforms)

bull ESTABLISH REVISED BIAS FACTOR B

f8 (b) u f8 (b) lk (b I all cases)

II

t t ~

shy

APPLICATIONS OF BIAS FACTORS

ii APPLICABLE TO PARTICULAR LOAD AND RESISTANCE RECIPE

bull USE BY THE API COMMITTEES AND MMS

Criteria Revisions

Recommended Design Guidelines Revisions bull

bullbull REASSESSMENT STUDIES AND DECISION MAKING

middot

bull RISK ASSESSMENT STUDIES BY THE OPERATORS

Individual Platforms

Fleet of Platforms

middot IMR Program Implementation

bullbull NOT TO BE USED IN DESIGN

I I fl J

EXAMPLE USE OF BIAS FACTORS

bull PHASE I EXAMPLE APPLICATION

bull EVALUATION OF ANNUAL PROBABILITY OF FAILURE

Establish Annual Maximum Seastate Data

Determine Coefficients to Define Base Shear

Determine Ultimate Capacity of Platform

Determine Probablllty of Failure Versus bullbbull

-bull DETERMINE UPDATED PROBABILITY OF FAILURE

Pf - P [S gt R]

Pf I P [~ gt Rb I b] f 8 (b) db

Pr - J P [S gt Rb I b] f bull 8 (b) db

bull TOTAL PROBABILITIES OF FAILURE

Total Pf - 1 bull Il [1 middotPf] - l Pf bull

Total Pf - 1 bull Il [1middot Pf] - l Pfmiddot

Total Pr = 1 - n [1middot PfbullJ = l Pfshy

I I I

EXAMPLE APPLICATION bull PHASE I

PLATFORM ST 151 K

bull GULF OF MEXICO

bull WATER DEPTH = 137 FT

bull ANNUAL MAXIMUM HS MEDIAN 175 ft COV 031

bull ASSUMED STORM DURATION 3 hoursmiddot

bull ASSUMED NUMBER OF WAVES 800

bull MEDIAN HS FOR STORM HOURS 1 2 AND 3

bull MEDIAN CURRENT VALUE bull

bull DETERMINE ULTIMATE CAPACITY FOR THREE DIRECTIONS

bull DETERMINE ANNUAL FAILURE PROBABILITIES

Broadsidemiddot Diagonal End-on

0037 0025 0007

0071 0044 0018

0022 0016 0004

DETERMINE TOTAL ANNUAL PROBABILITY OF FAILURE

Pf - 007

Pf - 013

Pr - 004

bull

EXAMPLE USE OF BIAS FACTORS - PHASE II middot

bull DETERMINE PREDICTIVE PROBABILITY OF FAILURE

Case - 1

Pf _ S S Pf (b b2) fa1a2 (b b2) db db2

g bull -1 (b1 R I S) - 1

92 - (b2 R2 I S) - 1

Case -2

Pf s ~ [bJ f11 (b) db

Pf [bJ p [g lt 0)

g - bull 1 (b1 R1 I S) - 1

1

bull1) l

I

BENEFITS OF ANDREW JIP bull PHASE II

lt

bull IMPROVED LOAD AND RESISTANCE RECIPE

bull IMPROVED CALIBRATION PROCEDURE

bull MULTIPLE BIAS FACTORS

bull PROCEDURE TO INCLUDE EFFECT OF OTHER PLATFORMS

bull PROCEDURE TO USE MULTIPLE BIAS FACTORS

  • Structure Bookmarks
    • ANDREW PHASE II JOINT1 INDUSTRY PROJECT
    • PROGRESS MEETING NUMBER 1
    • ANDREW PHASE II JOINT INDUSTRY PROJECT PROGRESS MEETING NUMBER 1 NOVEMBER 16 AGENDA
    • ANDREW PHASE II UST OF PARTICIPANTS
    • ANDREW PHASE I OBJECTIVES AND RESULTS
    • ANDREW PHASE II PROJECT OBJECTIVES
    • SUBSEQUENT PROJECTS
    • ANDREW PHASE II PROJECT TEAM
    • ANDREW PHASE II UPDATE MODELING middotAND ANALYSIS RECIPE
    • -ANDREW PHASE II CONDUCTOR RESISTANCE
    • ANDREW PHASE Il WA VE FORCE PROFILE
    • ANDREW PHASE II WA VE FORCE PROFILEmiddot
    • ANDREW PHASE II BRACE MODELING
    • ANDREW PHASE II FOUNDATIONS
    • ANDREW PHASE II middot middot JOINT MODELING
    • ANDREW PHASE Il JOINT MODELING OPTIONS DISCUSSION
    • ANDREW PHASE II JOINT MODEJJNG OPTIONS ILLUSTRATIONS
    • ANDREW PHASE II JOINT MODELING PHASE II METHOD
    • ANDREW PHASE II ANDREW STORM CONDmONS UPDATED
    • Platform Selection bullbullbull1111111111111
    • BasisCriteria for Selection bullbullbull1111111111111
    • CALIBRATION
    • WHERE TO INTRODUCE BIAS FACTOR (B
    • PHASE I -KEY CONCLUSIONS
    • bull CALIBRATION ISSUES PHASE II
    • PHASE II WORK SCOPE (CALIBRATION)
    • SOFTWARE
    • SENSITIVITY STUDIES POSTERIOR DISTRIBUTION OF B
    • WEIGHTING APPROACH bull
    • INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON
    • CATEGORIZATION OF ADDITIONAL PLATFORMS
    • UPDATING OF CALIBRATION RESULTS
    • APPLICATIONS OF BIAS FACTORS
    • EXAMPLE USE OF BIAS FACTORS
    • EXAMPLE APPLICATION bull PHASE I PLATFORM ST 151 K
    • EXAMPLE USE OF BIAS FACTORS -PHASE II middot
    • BENEFITS OF ANDREW JIP bull PHASE II
Page 51:  · . ' ANDREW PHASE II PROJECT OBJECTIVES . ' • ·UPDATE MODELING AND ANALYSIS RECIPE· JOINTS ' ' ·BRACES WAVE PROFILE FOUNDATIONS . ' ' CONDUCTORS· • ANALVZE PLATFORMS USING

PRIOR Distribution

PLATFORM EXPERIENCE

BAYESIAN Survivals and Failures

UPDATING

Waves Winds and Currents bull

POSTERIOR Distribution

- A_middotJA ~ Survival Data Failure Data

Result Result

Load Resistance

middotfl

-

Bayesian Updaing middotGeneral Description (Petrauskas 1992)

L Imiddot

WHERE TO INTRODUCE BIAS FACTOR (B

On Resistance (Rmue - B (RPREDICreD I

middotOn Loads (Smue middot - B (SPRE01cTEo

On Safety Margin (RSmue - B (RSPREDICTED

CALIBRATION METHODOLOGY

gt

r

FIELD OBSERVATIONS

Observed Behavior

survivals Damages

Failures

r

bull

CAPACITY ANALYSIS

Predicted Load and Resistance Predicted Failure Modes

Survival Damage Failure cases

RELIABILITY ANALYSIS

- Probability of Failure ~

Ukellhood Failure

BAYESIAN UPDATING

Prior Distnbution of B Combined Likelihood Function

Posterior Dlstnbution of B

BIAS FACTOR

Mean Value cov bull

Represents Modeling Uncertainties

Prior Distribution of B Likelihood Functionmiddot Observing Success case STISIK and Posterior Distribution of B

2 00

uo

160

140

120

e 100 L $ uoIshy

0 60 ~ ~ 040

020

0 00

0 00 0 50

_ I 00

b

I 50 2 00

-shy lk [bl STISIK)

--shy Pnor-f(b)

--shy Pos1enor r(bl ISIK)

Prior or B Mean bull 10 COV bull 030

Posterlor of Bbull Meanbull 136COV 0 16

2 50

Likelihood Function and Posterior Distribution Example Platform STISIK

- -

~

- Prior and Posterior Distributions or B Failure cases (ST177B STlSlH ST130A T21) Damage cases (T23 T25 ST161A) Success cases (STlSlK ST130Q ST134W WD90A MC311 MC397)

400

350

300

2 so 0-

I

bull bull

bull bull gt

- bullbull ~ bull bull I bull

_ bull bull bullbull I bull bull bullII V -

bull bull

I bull ~ bull

bullbull bullI bull bullI bull bull bull

middot -~

bull__ ~ middotI middotmiddot

L 2 00

e ~ I SO

-I 00

- 0 so

000 gt

0 00 020 0 40 060 0 80 I 00 120 140 160 110 200

b -

-

shy

Prior-r

- - - - Posterior fLlll failure cases

bull bull bull bull bull bull bull Posterior flall damage cases

- bull bull - bull middot Postenor flall success cases

Posterior flall success fulure damage cases

-

Posterior of B Mean= 119 COV 0101

Posterior Distribution or Blas Factor (8) bull Different Categories and All Cases Combined

shy

l

PHASE I - KEY CONCLUSIONS

bull GLOBAL CORRECTION FACTOR

bull 15 TO 20 CONSERVATISM MEAN)

bull UNEXPECTED SURVIVALS HAVE GREATEST IMPACT

bull DAMAGED CASES MORE INFORMATIVE

bull CALIBRATION ISSUES PHASE II

bull

EFFECT OF IMPROVED CAPACITY ANALYSIS RECIPE

bull APPLICABILITY TO MULTIPLE FAILURE MODES

bull NO FOUNDATION DAMAGE CASE INCLUDED

bull IMPROVED REPRESENTATION OF FULL ANDREW DISTRIBUTION

-bull NUMBER OF PLATFORMS CONSIDERED

I 1 I

I t ) bull ~

PHASE II WORK SCOPE (CALIBRATION)

bull

bull ESTABLISH MULTIPLE BIAS FACTORS

Use Nine Base Case Platforms

middot - Capacity Analysis for Revised recipe

Foundation Blas Factor from Foundation JIP

bull DEVELOP WEIGHTING APPROACH

bull UPDATE MULTIPLE BIAS FACTORS

bullEffect of Platforms Not In the Nine Base Cases

bull EXAMPLE APPLICATION OF BIAS FACTORS I

bullbull

PHASE I PHASE II

CAPACITYbull ANALYSISRu bull

---------------------1----middot

Jc1 - iJbullraquoJ dbull

FttlS frt[f~l ratdhJJ

-

LIKELIHOOD FUNCTION

DEVELOPMENT

lk Jb Jfall1N) bull Jb)

111 Ill Isurvtvoq bull 1 - bl

-middot -------------------shy ~---------------------middot-middot

r111ir8 Cb1 111 Cbl BAYESIAN

UPDATING

~cove

R1R2 bull middot---------------------~----

Jblbull P(g cOJ

1 bull JbRI (MBSI I bull 1

~middot

llcJllJfalluro) bull P(gcOJ Ill 111 b21fallun) bull P ((g1 CO) U 1112 c OD

middot--------------------shy ----------------------shy bull

r111111r8 Cbl 111 Ill i-1nro1

deg ~1b2)laquottfrr (b1b21-lnfo) re Ill bull 1-11middot b2) bull lodegamp

middot

i1 covamp i2 cov~i

-

shy

--

Likelihood Function vs b All Success Case STISI K STl30Q STl34W WD90A MC311 MC397

I 0000

0 9000

OIOOO

0 7000

~ 0amp000

ta osoOO

~ 0 4000

0 3000

0 2000

0 1000

bull 0 0000

bull

I

middot

middot bull I

Ibull

-sibull ~ -middot-middot-middot-middot -middot-

----7----+-----~ I

Ibull bull bull I

rbull

--- Jk ISIK[blsuccess)

- bull - bull - lk 130Q[blsuccess) bull

lk 134W[blsuccess)

Jk-WD90A(blsuccc~I

- - - lk MC397[blsucccss)

--ltgt-- lk middotMC31 l[blsucccss)

--- Jk [bl 6 out of6succcsHascs)

05 I 5 2 250

b

Likelihood Functionsmiddot Success Cases - Phbullsc L

shy

bullbull bull bull bull bull bull -------------------~

bull bull bullbull bullbull bullbull bullbull bullbull bullbull bull

bull bullbull

bullbullbullbull

bullbull

bullbullbullbullbullbullbull

bullbullbullbullbullbull

bullbullbullbullbullbullbull

z 0

t z ) LL

c 0 0c i ~ i

J

~ sect ~--------------~ e l5

I

I

SOFTWARE

bull PF PROGRAM

bull CALREL

bull PF +CALREL

bull RELACS

bull

gt i

l

SENSITIVITY STUDIES

POSTERIOR DISTRIBUTION OF B

bull EFFECT OF CHANGE IN DISTRIBUTION OF PRIOR OF B

bull EFFECT OF REMOVAL OF ONE PLATFORM FROM EACH CATEGORY

bull EFFECT OF REMOVAL OF THREE DAMAGE CASES

bull EFFECT OF ADDITIONAL 13 SIMILAR PLATFORMS

-I

Prior and Posterior Distributions of B Variation In COV of PriorofB fromOl to04 -- -

4500

-4000

-

- -

3500 --middot - bull bull - bull bull Prior-r ICOV =02

3000

- - - - Prior-r ICOV =03 -

bull -~ e 2soo -

bull bull bull bull bull bull bull Pnor-r ICOV =o 4t frJ ~ 2000 Posterior rlCOV =0 2

bull bull Postenor rlCOV =0 3 bull bull 1500 bull middot bull - shy -middot- Posterior rlCOV =04

middot~ I 000 - bull

-middotmiddot ~ ~middotmiddot M -~

- middot 0 500

wbullbullbull _ I middot~ middotmiddotmiddotmiddotmiddot middotmiddotmiddotmiddotmiddot - -~ ~ __ middot_ _ bull -shy0 000

0 000 0200 0 400 0 600 0 800 IOOO 1200 1400 1600 1800 2000 -

- - b

-

=shy

Posterior Distribution of Bias Factor (B) bull Effect of Variation In COV of Prior of B

bull

bullPrior and Posterior Distributions of B Evaluating Effect of Removal of One Platform from Each Category from the Base Case of 13 Platforms

400

~ ~ 350

j IA i

3 00 bull I bull I ---Prior-r I bull l

250 1 I 1

I bull bull ---Posterior r lall 13 casesshy- ~E 2 00 bull bull bull bull bull bull bull Posterior r without STISIK

- ~ 1 bull - bull I - bull bull - bull bull Posterior r Iwithout T2I I 50 I

J-tt~i - bull - bull - Posterior r Iwithout T2S

tOO [-

I I ~ ~ I 0 50

0 00

___ r J i

l~T

~--

0 00 0 20 0 40 0 60 080 I 00 120 140 160 110 2 00

b

zshy

Posterior Distribution of Bias Factor (8) bull Effect of Removal of middotPlatform from Any Category

bullPrior and Posterior Distributions or B Evaluating Effect or Removal or 3 Damage Cases from the Base Case or 13 Platrorms

4

35

3

25

e L 2 ~ bulle shy

I 15

bull I

05

0

0 02 04 06 I 2 I 4 16 IS 2

I I

I

I

I Ibull

J -tmiddot L--

bullI

~

I

fl

II ~

I I v I

middot bull bull

bull L

~ ll- I-~lt

OS I

b

I

bull

1middot bullbullbullbull

~ __

bull

- - - - Prior-r

- - - - Postenor flall failure cases

Postenor flall success cases

--- Posterior flall 13 cases

---Posterior fI success failure cases only

IPosterior or B Mean =t11 COV =0121

-

_I

Posterir Distribution or Bias Factor (B) bull Effect or Removal or 3-Damage Cases

-

bullPrior and Posterior Distributions or B Evaluating Effect or Additional 13 Similar Observations as

Used in Andrew JIP The POsterior or B from Andrew-JIP Considered as New Prior or B

- 500

450

400

350

bull bull

bull bull bull bull J bull

bull bull f ~ I

I ~ Ibull bull

bullbull bull

l bull I -bullbull bull bull

I

I bull bull

bull I I~~bull bull bull

bull ~

I bull bull

~ ~_ ~~~

- - New Prior r bull

Posterior 114 new failure cases ~ 300- bull bull bull bull bull bull bull Posterior f 13 new damage casesl 150 ~

- bull bull - bull bull Posterior 116 new success cases i 2 00

Postenor 1113 new success failure -150 damage cases

I 00

0 50 -

0 00 0 00 0 20 0 40 0 60 0 80 I 00 120 I 40 160 110 2 00

Posterior ol B Mean= 120 COV= 0071 b

shy

shy

_rshy

Posterior Distribution of Bias Factor (B) bull Effect of Additional 13-Similar Observations

1

1 I

WEIGHTING APPROACH bull

bull ASSESSMENT OF PLATFORMS DATA

Select Platforms Most Mappable to the Analyzed Platforms

Unlikely SurvivalDamage Cases Identification

Based on Cut-Off Criteria Agemiddot Hs etc

bull INFORMATION REQUIRED FOR COMPARISON

bull CATEGORIZATION OF ADDITIONALmiddot PLATFORMS

bull UPDATING OF CALIBRATION RESULTS

--

OVERVIEW OF PLATFORMS SURVIVEDDAMAGED -~

Operator -

- Platforms

Survived CS)

Platforms

Damaged CD)

Platforms Used In

Phase IJIP -

JI~ ~Bill~IRBDll -

Amoco -

Chevron -

Exxon

Mob II

Phllllps

Shell

Unocal -

30

143

72 22

7

39 25shy -

1

12

bull

1 bull

1

1(S)+1 (D)

3 (S) +3(D)

2 (S)

bull shy

bull

bull

bull

Total Partlclnants Platforms 338 - 15 10C6S+4D) -

Other Operators

310

13

3 (D)

Total Platforms 648 28 13C6S+7D) -

Summary of Andrew JIP bull Phase I Platforms

Plairshy w v- N-ber Lea-le 1 shy car Hind_ o-w N-bull Deplh fl fl IJpe la bull n Yeadon Jadtet lbelledtf liefalit -- _ -OD

IL H - -shy _ICllshy IL IL - - rQClto ~ _

K STUIK 137 1963 8 365 608S4 Cellardeck

-middot STl-shy 170 1964 4 -~ K 39 62269 Cellardeck - --ST134W 137 1911 4 43 608S4 Smvlnlshy- _ - -

WD90A 184 1964 8 K 267Sbull34 83 SOS31 SUboftllar deck SmvMI

MC311 343 1978 8 x S7 62 - MC397 468 1991 4 x shy S33bull6SO 65911 -shySmeSmvlvol -- gt -

r-- T23ST52 63 1969 4 ton- shyKshy 52594 S03 -T25SSl39 62 1969 4 Kshy SOS94 S063 - middot ton-shy__ -

bullbull a ton--shyST161A 118 1964 8 K - 347S 51495 Cellarclecti

r~

ST1771 142 1965 8 K 39 60167 CeDordeck failme

STISIH 137 1964 8 K 395 608S4 CeDardeck Failme

K ST130A 140 1958 a SS 6095 I lblv Pai1uno - T21ST72 61 1969 4 Kshy S4S94 4974 I lhlv Pai1uno -

- -

JE

INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON

bull WATER DEPTH

bull bull YEAR OF INSTALLATIONDESIGN

bull PLATFORM ORIENTATION

bull PLATFORM LATITUDE AND LONGITUDE

bull KEY SKETCHES OF PLATFORM STRUCTURAL FRAMING

Number of Legs Plies Leg Batter Number of Conductors bull Deck Elevations (Including sub-cellar deck If any) Jacket Framing Elevations Brace type In Vertical Frames Sizes of Primary Members (leg pile vertical frames braces primary horizontals) Joint Details (can gap etc) LegPiie Annulus Grouting Status

bull middotPOST ANDREW INSPECTION REPORT Significant Wave Height Hs (ft)

bull INFORMATION ON DAMAGES PRIOR TO ANDREW

bull RESULTS FROM IN-HOUSE STUDIES IF AVAILABLE

bull ~RIGINAL DESIGN CRITERIA INFORMATION IF AVAILABLE

bull SOIL SHEAR STRENGTH PROFILE IF AVAILABLE

Platforms Identified for Evaluating Therr Likelihood of Falling Under Unexpected Survival Category

OPlRATOR

middotADA

BLOClt

llmucnJllE

NAME

YEAll

INSTAUJD

WA1ER

DBPTll

DISTANCE

JlllOM SHOllE

(MILU)

Bl

(M)

AMOCO 111113

STl11

WDI075

A

bull D

St

17

120

1n

44

20

10 I

CllllVllQfl STlllO c 61 110 21 11 STIUO D Q 1110 21 11

STl134 p Q 137 29 11

ST0151 G 61 137 11

ST0151 I 3 121 11

ST0151 J Q 140 11

STl151 It 137 11

ST0151 PJl001 Q UT 11

ST0151 PROD-2 Q UT 11

ST017 A 3 140 35 11

WDIUT c 65 214 34 10

llXXON STI055 II 51 6T 14 I

STl15 A t3 34

WDI073

WD0073

Amiddotshy Q

1111 1111

22

22

I

I

WDI073 c 1n 22 I

- WDOCl93-shy II

65 1110 2T A 3 194 31 MOBU 111121

SS IOl7

A B

501

50

50

31

T

I

T 6

MtlllPllY SS114 B T 50 T T

PlllllIPS SS0149 A 5 55 33 T

SAMllDAN STln A 107 T T

SBllLL WDl103 A 65 223 2T WDl103 B 65 221 2T WD0104 c 65 221 2T WD0104 D 6T 20 2T

UNOCAL SS020I

SS 020I SSl20I

SS 0209

SS0209

II p

PURll

A

B

3

n 61

103

97

97

95

100

34

34

34

34 34

bullbull bullbullbull

- SS0209 D u 95 34 I

- SS 1209 SSl253

SS0253

G A

c

6T

2

100

165

1T5

34

54

54

bullbull I

I I

I I

SOUTH MARCH

~LAND

N

SOUTH PaTOI ENE ISWltD AREA

MMS Defined Limits ror Offshore Structures Requiring Inspections Following Andrew

t---Jbulln3 T25bull I bullTtl

bull1$1Htl1Abull bull1SIK1779bull

~

middot-DaTA AREA~T

CRANO rsLE

AREA bull1300

bullll4W

=shy

bullPath of Hurricane Andrew with Platforms Used In Calibration

Likelihood Function vs b All Success Case STISI K STIJOQ ST134W WD90A MCJll MCJ97

I 0000

0 9000

0 8000

07000

-Bo6000

tiiii 0 5000 -

04000

0 3000

0 2000

01000

0 0000

0 05 25

ri1rt middot-middot-middot-middot-middot-

I 5 2

b

---11 middot151K(blsu~ccsI

- bull - bull - fk 130Q(bfU~lCSS)

lie bull134W(blsuccc]

lk-WD90A[blsultCS]

- - - lk middotMC397(bfUtlCS]

--o-- lk-MCJll(bbuctcsJ

--- ll (bl 6 out of 6 Mlttcs case]

_

( Likelihood Functions middotSuccess Cases - p - 1 c I

CATEGORIZATION OF ADDITIONAL PLATFORMS

I I I

bull CATEGORIES

Unexpected Survivals

Expected Survivals

bull APPROACH

Identify Corresponding Platform(s) Evaluated In the JIP

Determine Max Wave Height at Platform Location

Evaluate Whether Wave Hit the Deck

middot - Evaluate Likelihood of R and S to Differ from the Base Cases

Review Joint Information

Assess Special Features

bull IDENTIFY NUMBER OF PLATFORMS IN EACH CATEGORY

I

UPDATING OF CALIBRATION RESULTS

bull INCLUDE EFFECT OF ADDITIONAL PLATFORMS ON bullBbull

Consider Increased 1Uncertalntles In R and S

Develop bullNeW Ukellhood Functions for Different Categories

middotLet bullxbull Number of Ukely Unexpected Survival Cases and y Number of Ukely Expected Survival Cases

Determine Combined Likelihood Function lk (b Iadditional platforms) middot

bull REVISED LIKELIHOOD FUNCTION

lk(b I all cases) = lk(b I 9 platforms) middot x lk(b Ix platforms) x lk(b I y platforms)

bull ESTABLISH REVISED BIAS FACTOR B

f8 (b) u f8 (b) lk (b I all cases)

II

t t ~

shy

APPLICATIONS OF BIAS FACTORS

ii APPLICABLE TO PARTICULAR LOAD AND RESISTANCE RECIPE

bull USE BY THE API COMMITTEES AND MMS

Criteria Revisions

Recommended Design Guidelines Revisions bull

bullbull REASSESSMENT STUDIES AND DECISION MAKING

middot

bull RISK ASSESSMENT STUDIES BY THE OPERATORS

Individual Platforms

Fleet of Platforms

middot IMR Program Implementation

bullbull NOT TO BE USED IN DESIGN

I I fl J

EXAMPLE USE OF BIAS FACTORS

bull PHASE I EXAMPLE APPLICATION

bull EVALUATION OF ANNUAL PROBABILITY OF FAILURE

Establish Annual Maximum Seastate Data

Determine Coefficients to Define Base Shear

Determine Ultimate Capacity of Platform

Determine Probablllty of Failure Versus bullbbull

-bull DETERMINE UPDATED PROBABILITY OF FAILURE

Pf - P [S gt R]

Pf I P [~ gt Rb I b] f 8 (b) db

Pr - J P [S gt Rb I b] f bull 8 (b) db

bull TOTAL PROBABILITIES OF FAILURE

Total Pf - 1 bull Il [1 middotPf] - l Pf bull

Total Pf - 1 bull Il [1middot Pf] - l Pfmiddot

Total Pr = 1 - n [1middot PfbullJ = l Pfshy

I I I

EXAMPLE APPLICATION bull PHASE I

PLATFORM ST 151 K

bull GULF OF MEXICO

bull WATER DEPTH = 137 FT

bull ANNUAL MAXIMUM HS MEDIAN 175 ft COV 031

bull ASSUMED STORM DURATION 3 hoursmiddot

bull ASSUMED NUMBER OF WAVES 800

bull MEDIAN HS FOR STORM HOURS 1 2 AND 3

bull MEDIAN CURRENT VALUE bull

bull DETERMINE ULTIMATE CAPACITY FOR THREE DIRECTIONS

bull DETERMINE ANNUAL FAILURE PROBABILITIES

Broadsidemiddot Diagonal End-on

0037 0025 0007

0071 0044 0018

0022 0016 0004

DETERMINE TOTAL ANNUAL PROBABILITY OF FAILURE

Pf - 007

Pf - 013

Pr - 004

bull

EXAMPLE USE OF BIAS FACTORS - PHASE II middot

bull DETERMINE PREDICTIVE PROBABILITY OF FAILURE

Case - 1

Pf _ S S Pf (b b2) fa1a2 (b b2) db db2

g bull -1 (b1 R I S) - 1

92 - (b2 R2 I S) - 1

Case -2

Pf s ~ [bJ f11 (b) db

Pf [bJ p [g lt 0)

g - bull 1 (b1 R1 I S) - 1

1

bull1) l

I

BENEFITS OF ANDREW JIP bull PHASE II

lt

bull IMPROVED LOAD AND RESISTANCE RECIPE

bull IMPROVED CALIBRATION PROCEDURE

bull MULTIPLE BIAS FACTORS

bull PROCEDURE TO INCLUDE EFFECT OF OTHER PLATFORMS

bull PROCEDURE TO USE MULTIPLE BIAS FACTORS

  • Structure Bookmarks
    • ANDREW PHASE II JOINT1 INDUSTRY PROJECT
    • PROGRESS MEETING NUMBER 1
    • ANDREW PHASE II JOINT INDUSTRY PROJECT PROGRESS MEETING NUMBER 1 NOVEMBER 16 AGENDA
    • ANDREW PHASE II UST OF PARTICIPANTS
    • ANDREW PHASE I OBJECTIVES AND RESULTS
    • ANDREW PHASE II PROJECT OBJECTIVES
    • SUBSEQUENT PROJECTS
    • ANDREW PHASE II PROJECT TEAM
    • ANDREW PHASE II UPDATE MODELING middotAND ANALYSIS RECIPE
    • -ANDREW PHASE II CONDUCTOR RESISTANCE
    • ANDREW PHASE Il WA VE FORCE PROFILE
    • ANDREW PHASE II WA VE FORCE PROFILEmiddot
    • ANDREW PHASE II BRACE MODELING
    • ANDREW PHASE II FOUNDATIONS
    • ANDREW PHASE II middot middot JOINT MODELING
    • ANDREW PHASE Il JOINT MODELING OPTIONS DISCUSSION
    • ANDREW PHASE II JOINT MODEJJNG OPTIONS ILLUSTRATIONS
    • ANDREW PHASE II JOINT MODELING PHASE II METHOD
    • ANDREW PHASE II ANDREW STORM CONDmONS UPDATED
    • Platform Selection bullbullbull1111111111111
    • BasisCriteria for Selection bullbullbull1111111111111
    • CALIBRATION
    • WHERE TO INTRODUCE BIAS FACTOR (B
    • PHASE I -KEY CONCLUSIONS
    • bull CALIBRATION ISSUES PHASE II
    • PHASE II WORK SCOPE (CALIBRATION)
    • SOFTWARE
    • SENSITIVITY STUDIES POSTERIOR DISTRIBUTION OF B
    • WEIGHTING APPROACH bull
    • INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON
    • CATEGORIZATION OF ADDITIONAL PLATFORMS
    • UPDATING OF CALIBRATION RESULTS
    • APPLICATIONS OF BIAS FACTORS
    • EXAMPLE USE OF BIAS FACTORS
    • EXAMPLE APPLICATION bull PHASE I PLATFORM ST 151 K
    • EXAMPLE USE OF BIAS FACTORS -PHASE II middot
    • BENEFITS OF ANDREW JIP bull PHASE II
Page 52:  · . ' ANDREW PHASE II PROJECT OBJECTIVES . ' • ·UPDATE MODELING AND ANALYSIS RECIPE· JOINTS ' ' ·BRACES WAVE PROFILE FOUNDATIONS . ' ' CONDUCTORS· • ANALVZE PLATFORMS USING

L Imiddot

WHERE TO INTRODUCE BIAS FACTOR (B

On Resistance (Rmue - B (RPREDICreD I

middotOn Loads (Smue middot - B (SPRE01cTEo

On Safety Margin (RSmue - B (RSPREDICTED

CALIBRATION METHODOLOGY

gt

r

FIELD OBSERVATIONS

Observed Behavior

survivals Damages

Failures

r

bull

CAPACITY ANALYSIS

Predicted Load and Resistance Predicted Failure Modes

Survival Damage Failure cases

RELIABILITY ANALYSIS

- Probability of Failure ~

Ukellhood Failure

BAYESIAN UPDATING

Prior Distnbution of B Combined Likelihood Function

Posterior Dlstnbution of B

BIAS FACTOR

Mean Value cov bull

Represents Modeling Uncertainties

Prior Distribution of B Likelihood Functionmiddot Observing Success case STISIK and Posterior Distribution of B

2 00

uo

160

140

120

e 100 L $ uoIshy

0 60 ~ ~ 040

020

0 00

0 00 0 50

_ I 00

b

I 50 2 00

-shy lk [bl STISIK)

--shy Pnor-f(b)

--shy Pos1enor r(bl ISIK)

Prior or B Mean bull 10 COV bull 030

Posterlor of Bbull Meanbull 136COV 0 16

2 50

Likelihood Function and Posterior Distribution Example Platform STISIK

- -

~

- Prior and Posterior Distributions or B Failure cases (ST177B STlSlH ST130A T21) Damage cases (T23 T25 ST161A) Success cases (STlSlK ST130Q ST134W WD90A MC311 MC397)

400

350

300

2 so 0-

I

bull bull

bull bull gt

- bullbull ~ bull bull I bull

_ bull bull bullbull I bull bull bullII V -

bull bull

I bull ~ bull

bullbull bullI bull bullI bull bull bull

middot -~

bull__ ~ middotI middotmiddot

L 2 00

e ~ I SO

-I 00

- 0 so

000 gt

0 00 020 0 40 060 0 80 I 00 120 140 160 110 200

b -

-

shy

Prior-r

- - - - Posterior fLlll failure cases

bull bull bull bull bull bull bull Posterior flall damage cases

- bull bull - bull middot Postenor flall success cases

Posterior flall success fulure damage cases

-

Posterior of B Mean= 119 COV 0101

Posterior Distribution or Blas Factor (8) bull Different Categories and All Cases Combined

shy

l

PHASE I - KEY CONCLUSIONS

bull GLOBAL CORRECTION FACTOR

bull 15 TO 20 CONSERVATISM MEAN)

bull UNEXPECTED SURVIVALS HAVE GREATEST IMPACT

bull DAMAGED CASES MORE INFORMATIVE

bull CALIBRATION ISSUES PHASE II

bull

EFFECT OF IMPROVED CAPACITY ANALYSIS RECIPE

bull APPLICABILITY TO MULTIPLE FAILURE MODES

bull NO FOUNDATION DAMAGE CASE INCLUDED

bull IMPROVED REPRESENTATION OF FULL ANDREW DISTRIBUTION

-bull NUMBER OF PLATFORMS CONSIDERED

I 1 I

I t ) bull ~

PHASE II WORK SCOPE (CALIBRATION)

bull

bull ESTABLISH MULTIPLE BIAS FACTORS

Use Nine Base Case Platforms

middot - Capacity Analysis for Revised recipe

Foundation Blas Factor from Foundation JIP

bull DEVELOP WEIGHTING APPROACH

bull UPDATE MULTIPLE BIAS FACTORS

bullEffect of Platforms Not In the Nine Base Cases

bull EXAMPLE APPLICATION OF BIAS FACTORS I

bullbull

PHASE I PHASE II

CAPACITYbull ANALYSISRu bull

---------------------1----middot

Jc1 - iJbullraquoJ dbull

FttlS frt[f~l ratdhJJ

-

LIKELIHOOD FUNCTION

DEVELOPMENT

lk Jb Jfall1N) bull Jb)

111 Ill Isurvtvoq bull 1 - bl

-middot -------------------shy ~---------------------middot-middot

r111ir8 Cb1 111 Cbl BAYESIAN

UPDATING

~cove

R1R2 bull middot---------------------~----

Jblbull P(g cOJ

1 bull JbRI (MBSI I bull 1

~middot

llcJllJfalluro) bull P(gcOJ Ill 111 b21fallun) bull P ((g1 CO) U 1112 c OD

middot--------------------shy ----------------------shy bull

r111111r8 Cbl 111 Ill i-1nro1

deg ~1b2)laquottfrr (b1b21-lnfo) re Ill bull 1-11middot b2) bull lodegamp

middot

i1 covamp i2 cov~i

-

shy

--

Likelihood Function vs b All Success Case STISI K STl30Q STl34W WD90A MC311 MC397

I 0000

0 9000

OIOOO

0 7000

~ 0amp000

ta osoOO

~ 0 4000

0 3000

0 2000

0 1000

bull 0 0000

bull

I

middot

middot bull I

Ibull

-sibull ~ -middot-middot-middot-middot -middot-

----7----+-----~ I

Ibull bull bull I

rbull

--- Jk ISIK[blsuccess)

- bull - bull - lk 130Q[blsuccess) bull

lk 134W[blsuccess)

Jk-WD90A(blsuccc~I

- - - lk MC397[blsucccss)

--ltgt-- lk middotMC31 l[blsucccss)

--- Jk [bl 6 out of6succcsHascs)

05 I 5 2 250

b

Likelihood Functionsmiddot Success Cases - Phbullsc L

shy

bullbull bull bull bull bull bull -------------------~

bull bull bullbull bullbull bullbull bullbull bullbull bullbull bull

bull bullbull

bullbullbullbull

bullbull

bullbullbullbullbullbullbull

bullbullbullbullbullbull

bullbullbullbullbullbullbull

z 0

t z ) LL

c 0 0c i ~ i

J

~ sect ~--------------~ e l5

I

I

SOFTWARE

bull PF PROGRAM

bull CALREL

bull PF +CALREL

bull RELACS

bull

gt i

l

SENSITIVITY STUDIES

POSTERIOR DISTRIBUTION OF B

bull EFFECT OF CHANGE IN DISTRIBUTION OF PRIOR OF B

bull EFFECT OF REMOVAL OF ONE PLATFORM FROM EACH CATEGORY

bull EFFECT OF REMOVAL OF THREE DAMAGE CASES

bull EFFECT OF ADDITIONAL 13 SIMILAR PLATFORMS

-I

Prior and Posterior Distributions of B Variation In COV of PriorofB fromOl to04 -- -

4500

-4000

-

- -

3500 --middot - bull bull - bull bull Prior-r ICOV =02

3000

- - - - Prior-r ICOV =03 -

bull -~ e 2soo -

bull bull bull bull bull bull bull Pnor-r ICOV =o 4t frJ ~ 2000 Posterior rlCOV =0 2

bull bull Postenor rlCOV =0 3 bull bull 1500 bull middot bull - shy -middot- Posterior rlCOV =04

middot~ I 000 - bull

-middotmiddot ~ ~middotmiddot M -~

- middot 0 500

wbullbullbull _ I middot~ middotmiddotmiddotmiddotmiddot middotmiddotmiddotmiddotmiddot - -~ ~ __ middot_ _ bull -shy0 000

0 000 0200 0 400 0 600 0 800 IOOO 1200 1400 1600 1800 2000 -

- - b

-

=shy

Posterior Distribution of Bias Factor (B) bull Effect of Variation In COV of Prior of B

bull

bullPrior and Posterior Distributions of B Evaluating Effect of Removal of One Platform from Each Category from the Base Case of 13 Platforms

400

~ ~ 350

j IA i

3 00 bull I bull I ---Prior-r I bull l

250 1 I 1

I bull bull ---Posterior r lall 13 casesshy- ~E 2 00 bull bull bull bull bull bull bull Posterior r without STISIK

- ~ 1 bull - bull I - bull bull - bull bull Posterior r Iwithout T2I I 50 I

J-tt~i - bull - bull - Posterior r Iwithout T2S

tOO [-

I I ~ ~ I 0 50

0 00

___ r J i

l~T

~--

0 00 0 20 0 40 0 60 080 I 00 120 140 160 110 2 00

b

zshy

Posterior Distribution of Bias Factor (8) bull Effect of Removal of middotPlatform from Any Category

bullPrior and Posterior Distributions or B Evaluating Effect or Removal or 3 Damage Cases from the Base Case or 13 Platrorms

4

35

3

25

e L 2 ~ bulle shy

I 15

bull I

05

0

0 02 04 06 I 2 I 4 16 IS 2

I I

I

I

I Ibull

J -tmiddot L--

bullI

~

I

fl

II ~

I I v I

middot bull bull

bull L

~ ll- I-~lt

OS I

b

I

bull

1middot bullbullbullbull

~ __

bull

- - - - Prior-r

- - - - Postenor flall failure cases

Postenor flall success cases

--- Posterior flall 13 cases

---Posterior fI success failure cases only

IPosterior or B Mean =t11 COV =0121

-

_I

Posterir Distribution or Bias Factor (B) bull Effect or Removal or 3-Damage Cases

-

bullPrior and Posterior Distributions or B Evaluating Effect or Additional 13 Similar Observations as

Used in Andrew JIP The POsterior or B from Andrew-JIP Considered as New Prior or B

- 500

450

400

350

bull bull

bull bull bull bull J bull

bull bull f ~ I

I ~ Ibull bull

bullbull bull

l bull I -bullbull bull bull

I

I bull bull

bull I I~~bull bull bull

bull ~

I bull bull

~ ~_ ~~~

- - New Prior r bull

Posterior 114 new failure cases ~ 300- bull bull bull bull bull bull bull Posterior f 13 new damage casesl 150 ~

- bull bull - bull bull Posterior 116 new success cases i 2 00

Postenor 1113 new success failure -150 damage cases

I 00

0 50 -

0 00 0 00 0 20 0 40 0 60 0 80 I 00 120 I 40 160 110 2 00

Posterior ol B Mean= 120 COV= 0071 b

shy

shy

_rshy

Posterior Distribution of Bias Factor (B) bull Effect of Additional 13-Similar Observations

1

1 I

WEIGHTING APPROACH bull

bull ASSESSMENT OF PLATFORMS DATA

Select Platforms Most Mappable to the Analyzed Platforms

Unlikely SurvivalDamage Cases Identification

Based on Cut-Off Criteria Agemiddot Hs etc

bull INFORMATION REQUIRED FOR COMPARISON

bull CATEGORIZATION OF ADDITIONALmiddot PLATFORMS

bull UPDATING OF CALIBRATION RESULTS

--

OVERVIEW OF PLATFORMS SURVIVEDDAMAGED -~

Operator -

- Platforms

Survived CS)

Platforms

Damaged CD)

Platforms Used In

Phase IJIP -

JI~ ~Bill~IRBDll -

Amoco -

Chevron -

Exxon

Mob II

Phllllps

Shell

Unocal -

30

143

72 22

7

39 25shy -

1

12

bull

1 bull

1

1(S)+1 (D)

3 (S) +3(D)

2 (S)

bull shy

bull

bull

bull

Total Partlclnants Platforms 338 - 15 10C6S+4D) -

Other Operators

310

13

3 (D)

Total Platforms 648 28 13C6S+7D) -

Summary of Andrew JIP bull Phase I Platforms

Plairshy w v- N-ber Lea-le 1 shy car Hind_ o-w N-bull Deplh fl fl IJpe la bull n Yeadon Jadtet lbelledtf liefalit -- _ -OD

IL H - -shy _ICllshy IL IL - - rQClto ~ _

K STUIK 137 1963 8 365 608S4 Cellardeck

-middot STl-shy 170 1964 4 -~ K 39 62269 Cellardeck - --ST134W 137 1911 4 43 608S4 Smvlnlshy- _ - -

WD90A 184 1964 8 K 267Sbull34 83 SOS31 SUboftllar deck SmvMI

MC311 343 1978 8 x S7 62 - MC397 468 1991 4 x shy S33bull6SO 65911 -shySmeSmvlvol -- gt -

r-- T23ST52 63 1969 4 ton- shyKshy 52594 S03 -T25SSl39 62 1969 4 Kshy SOS94 S063 - middot ton-shy__ -

bullbull a ton--shyST161A 118 1964 8 K - 347S 51495 Cellarclecti

r~

ST1771 142 1965 8 K 39 60167 CeDordeck failme

STISIH 137 1964 8 K 395 608S4 CeDardeck Failme

K ST130A 140 1958 a SS 6095 I lblv Pai1uno - T21ST72 61 1969 4 Kshy S4S94 4974 I lhlv Pai1uno -

- -

JE

INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON

bull WATER DEPTH

bull bull YEAR OF INSTALLATIONDESIGN

bull PLATFORM ORIENTATION

bull PLATFORM LATITUDE AND LONGITUDE

bull KEY SKETCHES OF PLATFORM STRUCTURAL FRAMING

Number of Legs Plies Leg Batter Number of Conductors bull Deck Elevations (Including sub-cellar deck If any) Jacket Framing Elevations Brace type In Vertical Frames Sizes of Primary Members (leg pile vertical frames braces primary horizontals) Joint Details (can gap etc) LegPiie Annulus Grouting Status

bull middotPOST ANDREW INSPECTION REPORT Significant Wave Height Hs (ft)

bull INFORMATION ON DAMAGES PRIOR TO ANDREW

bull RESULTS FROM IN-HOUSE STUDIES IF AVAILABLE

bull ~RIGINAL DESIGN CRITERIA INFORMATION IF AVAILABLE

bull SOIL SHEAR STRENGTH PROFILE IF AVAILABLE

Platforms Identified for Evaluating Therr Likelihood of Falling Under Unexpected Survival Category

OPlRATOR

middotADA

BLOClt

llmucnJllE

NAME

YEAll

INSTAUJD

WA1ER

DBPTll

DISTANCE

JlllOM SHOllE

(MILU)

Bl

(M)

AMOCO 111113

STl11

WDI075

A

bull D

St

17

120

1n

44

20

10 I

CllllVllQfl STlllO c 61 110 21 11 STIUO D Q 1110 21 11

STl134 p Q 137 29 11

ST0151 G 61 137 11

ST0151 I 3 121 11

ST0151 J Q 140 11

STl151 It 137 11

ST0151 PJl001 Q UT 11

ST0151 PROD-2 Q UT 11

ST017 A 3 140 35 11

WDIUT c 65 214 34 10

llXXON STI055 II 51 6T 14 I

STl15 A t3 34

WDI073

WD0073

Amiddotshy Q

1111 1111

22

22

I

I

WDI073 c 1n 22 I

- WDOCl93-shy II

65 1110 2T A 3 194 31 MOBU 111121

SS IOl7

A B

501

50

50

31

T

I

T 6

MtlllPllY SS114 B T 50 T T

PlllllIPS SS0149 A 5 55 33 T

SAMllDAN STln A 107 T T

SBllLL WDl103 A 65 223 2T WDl103 B 65 221 2T WD0104 c 65 221 2T WD0104 D 6T 20 2T

UNOCAL SS020I

SS 020I SSl20I

SS 0209

SS0209

II p

PURll

A

B

3

n 61

103

97

97

95

100

34

34

34

34 34

bullbull bullbullbull

- SS0209 D u 95 34 I

- SS 1209 SSl253

SS0253

G A

c

6T

2

100

165

1T5

34

54

54

bullbull I

I I

I I

SOUTH MARCH

~LAND

N

SOUTH PaTOI ENE ISWltD AREA

MMS Defined Limits ror Offshore Structures Requiring Inspections Following Andrew

t---Jbulln3 T25bull I bullTtl

bull1$1Htl1Abull bull1SIK1779bull

~

middot-DaTA AREA~T

CRANO rsLE

AREA bull1300

bullll4W

=shy

bullPath of Hurricane Andrew with Platforms Used In Calibration

Likelihood Function vs b All Success Case STISI K STIJOQ ST134W WD90A MCJll MCJ97

I 0000

0 9000

0 8000

07000

-Bo6000

tiiii 0 5000 -

04000

0 3000

0 2000

01000

0 0000

0 05 25

ri1rt middot-middot-middot-middot-middot-

I 5 2

b

---11 middot151K(blsu~ccsI

- bull - bull - fk 130Q(bfU~lCSS)

lie bull134W(blsuccc]

lk-WD90A[blsultCS]

- - - lk middotMC397(bfUtlCS]

--o-- lk-MCJll(bbuctcsJ

--- ll (bl 6 out of 6 Mlttcs case]

_

( Likelihood Functions middotSuccess Cases - p - 1 c I

CATEGORIZATION OF ADDITIONAL PLATFORMS

I I I

bull CATEGORIES

Unexpected Survivals

Expected Survivals

bull APPROACH

Identify Corresponding Platform(s) Evaluated In the JIP

Determine Max Wave Height at Platform Location

Evaluate Whether Wave Hit the Deck

middot - Evaluate Likelihood of R and S to Differ from the Base Cases

Review Joint Information

Assess Special Features

bull IDENTIFY NUMBER OF PLATFORMS IN EACH CATEGORY

I

UPDATING OF CALIBRATION RESULTS

bull INCLUDE EFFECT OF ADDITIONAL PLATFORMS ON bullBbull

Consider Increased 1Uncertalntles In R and S

Develop bullNeW Ukellhood Functions for Different Categories

middotLet bullxbull Number of Ukely Unexpected Survival Cases and y Number of Ukely Expected Survival Cases

Determine Combined Likelihood Function lk (b Iadditional platforms) middot

bull REVISED LIKELIHOOD FUNCTION

lk(b I all cases) = lk(b I 9 platforms) middot x lk(b Ix platforms) x lk(b I y platforms)

bull ESTABLISH REVISED BIAS FACTOR B

f8 (b) u f8 (b) lk (b I all cases)

II

t t ~

shy

APPLICATIONS OF BIAS FACTORS

ii APPLICABLE TO PARTICULAR LOAD AND RESISTANCE RECIPE

bull USE BY THE API COMMITTEES AND MMS

Criteria Revisions

Recommended Design Guidelines Revisions bull

bullbull REASSESSMENT STUDIES AND DECISION MAKING

middot

bull RISK ASSESSMENT STUDIES BY THE OPERATORS

Individual Platforms

Fleet of Platforms

middot IMR Program Implementation

bullbull NOT TO BE USED IN DESIGN

I I fl J

EXAMPLE USE OF BIAS FACTORS

bull PHASE I EXAMPLE APPLICATION

bull EVALUATION OF ANNUAL PROBABILITY OF FAILURE

Establish Annual Maximum Seastate Data

Determine Coefficients to Define Base Shear

Determine Ultimate Capacity of Platform

Determine Probablllty of Failure Versus bullbbull

-bull DETERMINE UPDATED PROBABILITY OF FAILURE

Pf - P [S gt R]

Pf I P [~ gt Rb I b] f 8 (b) db

Pr - J P [S gt Rb I b] f bull 8 (b) db

bull TOTAL PROBABILITIES OF FAILURE

Total Pf - 1 bull Il [1 middotPf] - l Pf bull

Total Pf - 1 bull Il [1middot Pf] - l Pfmiddot

Total Pr = 1 - n [1middot PfbullJ = l Pfshy

I I I

EXAMPLE APPLICATION bull PHASE I

PLATFORM ST 151 K

bull GULF OF MEXICO

bull WATER DEPTH = 137 FT

bull ANNUAL MAXIMUM HS MEDIAN 175 ft COV 031

bull ASSUMED STORM DURATION 3 hoursmiddot

bull ASSUMED NUMBER OF WAVES 800

bull MEDIAN HS FOR STORM HOURS 1 2 AND 3

bull MEDIAN CURRENT VALUE bull

bull DETERMINE ULTIMATE CAPACITY FOR THREE DIRECTIONS

bull DETERMINE ANNUAL FAILURE PROBABILITIES

Broadsidemiddot Diagonal End-on

0037 0025 0007

0071 0044 0018

0022 0016 0004

DETERMINE TOTAL ANNUAL PROBABILITY OF FAILURE

Pf - 007

Pf - 013

Pr - 004

bull

EXAMPLE USE OF BIAS FACTORS - PHASE II middot

bull DETERMINE PREDICTIVE PROBABILITY OF FAILURE

Case - 1

Pf _ S S Pf (b b2) fa1a2 (b b2) db db2

g bull -1 (b1 R I S) - 1

92 - (b2 R2 I S) - 1

Case -2

Pf s ~ [bJ f11 (b) db

Pf [bJ p [g lt 0)

g - bull 1 (b1 R1 I S) - 1

1

bull1) l

I

BENEFITS OF ANDREW JIP bull PHASE II

lt

bull IMPROVED LOAD AND RESISTANCE RECIPE

bull IMPROVED CALIBRATION PROCEDURE

bull MULTIPLE BIAS FACTORS

bull PROCEDURE TO INCLUDE EFFECT OF OTHER PLATFORMS

bull PROCEDURE TO USE MULTIPLE BIAS FACTORS

  • Structure Bookmarks
    • ANDREW PHASE II JOINT1 INDUSTRY PROJECT
    • PROGRESS MEETING NUMBER 1
    • ANDREW PHASE II JOINT INDUSTRY PROJECT PROGRESS MEETING NUMBER 1 NOVEMBER 16 AGENDA
    • ANDREW PHASE II UST OF PARTICIPANTS
    • ANDREW PHASE I OBJECTIVES AND RESULTS
    • ANDREW PHASE II PROJECT OBJECTIVES
    • SUBSEQUENT PROJECTS
    • ANDREW PHASE II PROJECT TEAM
    • ANDREW PHASE II UPDATE MODELING middotAND ANALYSIS RECIPE
    • -ANDREW PHASE II CONDUCTOR RESISTANCE
    • ANDREW PHASE Il WA VE FORCE PROFILE
    • ANDREW PHASE II WA VE FORCE PROFILEmiddot
    • ANDREW PHASE II BRACE MODELING
    • ANDREW PHASE II FOUNDATIONS
    • ANDREW PHASE II middot middot JOINT MODELING
    • ANDREW PHASE Il JOINT MODELING OPTIONS DISCUSSION
    • ANDREW PHASE II JOINT MODEJJNG OPTIONS ILLUSTRATIONS
    • ANDREW PHASE II JOINT MODELING PHASE II METHOD
    • ANDREW PHASE II ANDREW STORM CONDmONS UPDATED
    • Platform Selection bullbullbull1111111111111
    • BasisCriteria for Selection bullbullbull1111111111111
    • CALIBRATION
    • WHERE TO INTRODUCE BIAS FACTOR (B
    • PHASE I -KEY CONCLUSIONS
    • bull CALIBRATION ISSUES PHASE II
    • PHASE II WORK SCOPE (CALIBRATION)
    • SOFTWARE
    • SENSITIVITY STUDIES POSTERIOR DISTRIBUTION OF B
    • WEIGHTING APPROACH bull
    • INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON
    • CATEGORIZATION OF ADDITIONAL PLATFORMS
    • UPDATING OF CALIBRATION RESULTS
    • APPLICATIONS OF BIAS FACTORS
    • EXAMPLE USE OF BIAS FACTORS
    • EXAMPLE APPLICATION bull PHASE I PLATFORM ST 151 K
    • EXAMPLE USE OF BIAS FACTORS -PHASE II middot
    • BENEFITS OF ANDREW JIP bull PHASE II
Page 53:  · . ' ANDREW PHASE II PROJECT OBJECTIVES . ' • ·UPDATE MODELING AND ANALYSIS RECIPE· JOINTS ' ' ·BRACES WAVE PROFILE FOUNDATIONS . ' ' CONDUCTORS· • ANALVZE PLATFORMS USING

CALIBRATION METHODOLOGY

gt

r

FIELD OBSERVATIONS

Observed Behavior

survivals Damages

Failures

r

bull

CAPACITY ANALYSIS

Predicted Load and Resistance Predicted Failure Modes

Survival Damage Failure cases

RELIABILITY ANALYSIS

- Probability of Failure ~

Ukellhood Failure

BAYESIAN UPDATING

Prior Distnbution of B Combined Likelihood Function

Posterior Dlstnbution of B

BIAS FACTOR

Mean Value cov bull

Represents Modeling Uncertainties

Prior Distribution of B Likelihood Functionmiddot Observing Success case STISIK and Posterior Distribution of B

2 00

uo

160

140

120

e 100 L $ uoIshy

0 60 ~ ~ 040

020

0 00

0 00 0 50

_ I 00

b

I 50 2 00

-shy lk [bl STISIK)

--shy Pnor-f(b)

--shy Pos1enor r(bl ISIK)

Prior or B Mean bull 10 COV bull 030

Posterlor of Bbull Meanbull 136COV 0 16

2 50

Likelihood Function and Posterior Distribution Example Platform STISIK

- -

~

- Prior and Posterior Distributions or B Failure cases (ST177B STlSlH ST130A T21) Damage cases (T23 T25 ST161A) Success cases (STlSlK ST130Q ST134W WD90A MC311 MC397)

400

350

300

2 so 0-

I

bull bull

bull bull gt

- bullbull ~ bull bull I bull

_ bull bull bullbull I bull bull bullII V -

bull bull

I bull ~ bull

bullbull bullI bull bullI bull bull bull

middot -~

bull__ ~ middotI middotmiddot

L 2 00

e ~ I SO

-I 00

- 0 so

000 gt

0 00 020 0 40 060 0 80 I 00 120 140 160 110 200

b -

-

shy

Prior-r

- - - - Posterior fLlll failure cases

bull bull bull bull bull bull bull Posterior flall damage cases

- bull bull - bull middot Postenor flall success cases

Posterior flall success fulure damage cases

-

Posterior of B Mean= 119 COV 0101

Posterior Distribution or Blas Factor (8) bull Different Categories and All Cases Combined

shy

l

PHASE I - KEY CONCLUSIONS

bull GLOBAL CORRECTION FACTOR

bull 15 TO 20 CONSERVATISM MEAN)

bull UNEXPECTED SURVIVALS HAVE GREATEST IMPACT

bull DAMAGED CASES MORE INFORMATIVE

bull CALIBRATION ISSUES PHASE II

bull

EFFECT OF IMPROVED CAPACITY ANALYSIS RECIPE

bull APPLICABILITY TO MULTIPLE FAILURE MODES

bull NO FOUNDATION DAMAGE CASE INCLUDED

bull IMPROVED REPRESENTATION OF FULL ANDREW DISTRIBUTION

-bull NUMBER OF PLATFORMS CONSIDERED

I 1 I

I t ) bull ~

PHASE II WORK SCOPE (CALIBRATION)

bull

bull ESTABLISH MULTIPLE BIAS FACTORS

Use Nine Base Case Platforms

middot - Capacity Analysis for Revised recipe

Foundation Blas Factor from Foundation JIP

bull DEVELOP WEIGHTING APPROACH

bull UPDATE MULTIPLE BIAS FACTORS

bullEffect of Platforms Not In the Nine Base Cases

bull EXAMPLE APPLICATION OF BIAS FACTORS I

bullbull

PHASE I PHASE II

CAPACITYbull ANALYSISRu bull

---------------------1----middot

Jc1 - iJbullraquoJ dbull

FttlS frt[f~l ratdhJJ

-

LIKELIHOOD FUNCTION

DEVELOPMENT

lk Jb Jfall1N) bull Jb)

111 Ill Isurvtvoq bull 1 - bl

-middot -------------------shy ~---------------------middot-middot

r111ir8 Cb1 111 Cbl BAYESIAN

UPDATING

~cove

R1R2 bull middot---------------------~----

Jblbull P(g cOJ

1 bull JbRI (MBSI I bull 1

~middot

llcJllJfalluro) bull P(gcOJ Ill 111 b21fallun) bull P ((g1 CO) U 1112 c OD

middot--------------------shy ----------------------shy bull

r111111r8 Cbl 111 Ill i-1nro1

deg ~1b2)laquottfrr (b1b21-lnfo) re Ill bull 1-11middot b2) bull lodegamp

middot

i1 covamp i2 cov~i

-

shy

--

Likelihood Function vs b All Success Case STISI K STl30Q STl34W WD90A MC311 MC397

I 0000

0 9000

OIOOO

0 7000

~ 0amp000

ta osoOO

~ 0 4000

0 3000

0 2000

0 1000

bull 0 0000

bull

I

middot

middot bull I

Ibull

-sibull ~ -middot-middot-middot-middot -middot-

----7----+-----~ I

Ibull bull bull I

rbull

--- Jk ISIK[blsuccess)

- bull - bull - lk 130Q[blsuccess) bull

lk 134W[blsuccess)

Jk-WD90A(blsuccc~I

- - - lk MC397[blsucccss)

--ltgt-- lk middotMC31 l[blsucccss)

--- Jk [bl 6 out of6succcsHascs)

05 I 5 2 250

b

Likelihood Functionsmiddot Success Cases - Phbullsc L

shy

bullbull bull bull bull bull bull -------------------~

bull bull bullbull bullbull bullbull bullbull bullbull bullbull bull

bull bullbull

bullbullbullbull

bullbull

bullbullbullbullbullbullbull

bullbullbullbullbullbull

bullbullbullbullbullbullbull

z 0

t z ) LL

c 0 0c i ~ i

J

~ sect ~--------------~ e l5

I

I

SOFTWARE

bull PF PROGRAM

bull CALREL

bull PF +CALREL

bull RELACS

bull

gt i

l

SENSITIVITY STUDIES

POSTERIOR DISTRIBUTION OF B

bull EFFECT OF CHANGE IN DISTRIBUTION OF PRIOR OF B

bull EFFECT OF REMOVAL OF ONE PLATFORM FROM EACH CATEGORY

bull EFFECT OF REMOVAL OF THREE DAMAGE CASES

bull EFFECT OF ADDITIONAL 13 SIMILAR PLATFORMS

-I

Prior and Posterior Distributions of B Variation In COV of PriorofB fromOl to04 -- -

4500

-4000

-

- -

3500 --middot - bull bull - bull bull Prior-r ICOV =02

3000

- - - - Prior-r ICOV =03 -

bull -~ e 2soo -

bull bull bull bull bull bull bull Pnor-r ICOV =o 4t frJ ~ 2000 Posterior rlCOV =0 2

bull bull Postenor rlCOV =0 3 bull bull 1500 bull middot bull - shy -middot- Posterior rlCOV =04

middot~ I 000 - bull

-middotmiddot ~ ~middotmiddot M -~

- middot 0 500

wbullbullbull _ I middot~ middotmiddotmiddotmiddotmiddot middotmiddotmiddotmiddotmiddot - -~ ~ __ middot_ _ bull -shy0 000

0 000 0200 0 400 0 600 0 800 IOOO 1200 1400 1600 1800 2000 -

- - b

-

=shy

Posterior Distribution of Bias Factor (B) bull Effect of Variation In COV of Prior of B

bull

bullPrior and Posterior Distributions of B Evaluating Effect of Removal of One Platform from Each Category from the Base Case of 13 Platforms

400

~ ~ 350

j IA i

3 00 bull I bull I ---Prior-r I bull l

250 1 I 1

I bull bull ---Posterior r lall 13 casesshy- ~E 2 00 bull bull bull bull bull bull bull Posterior r without STISIK

- ~ 1 bull - bull I - bull bull - bull bull Posterior r Iwithout T2I I 50 I

J-tt~i - bull - bull - Posterior r Iwithout T2S

tOO [-

I I ~ ~ I 0 50

0 00

___ r J i

l~T

~--

0 00 0 20 0 40 0 60 080 I 00 120 140 160 110 2 00

b

zshy

Posterior Distribution of Bias Factor (8) bull Effect of Removal of middotPlatform from Any Category

bullPrior and Posterior Distributions or B Evaluating Effect or Removal or 3 Damage Cases from the Base Case or 13 Platrorms

4

35

3

25

e L 2 ~ bulle shy

I 15

bull I

05

0

0 02 04 06 I 2 I 4 16 IS 2

I I

I

I

I Ibull

J -tmiddot L--

bullI

~

I

fl

II ~

I I v I

middot bull bull

bull L

~ ll- I-~lt

OS I

b

I

bull

1middot bullbullbullbull

~ __

bull

- - - - Prior-r

- - - - Postenor flall failure cases

Postenor flall success cases

--- Posterior flall 13 cases

---Posterior fI success failure cases only

IPosterior or B Mean =t11 COV =0121

-

_I

Posterir Distribution or Bias Factor (B) bull Effect or Removal or 3-Damage Cases

-

bullPrior and Posterior Distributions or B Evaluating Effect or Additional 13 Similar Observations as

Used in Andrew JIP The POsterior or B from Andrew-JIP Considered as New Prior or B

- 500

450

400

350

bull bull

bull bull bull bull J bull

bull bull f ~ I

I ~ Ibull bull

bullbull bull

l bull I -bullbull bull bull

I

I bull bull

bull I I~~bull bull bull

bull ~

I bull bull

~ ~_ ~~~

- - New Prior r bull

Posterior 114 new failure cases ~ 300- bull bull bull bull bull bull bull Posterior f 13 new damage casesl 150 ~

- bull bull - bull bull Posterior 116 new success cases i 2 00

Postenor 1113 new success failure -150 damage cases

I 00

0 50 -

0 00 0 00 0 20 0 40 0 60 0 80 I 00 120 I 40 160 110 2 00

Posterior ol B Mean= 120 COV= 0071 b

shy

shy

_rshy

Posterior Distribution of Bias Factor (B) bull Effect of Additional 13-Similar Observations

1

1 I

WEIGHTING APPROACH bull

bull ASSESSMENT OF PLATFORMS DATA

Select Platforms Most Mappable to the Analyzed Platforms

Unlikely SurvivalDamage Cases Identification

Based on Cut-Off Criteria Agemiddot Hs etc

bull INFORMATION REQUIRED FOR COMPARISON

bull CATEGORIZATION OF ADDITIONALmiddot PLATFORMS

bull UPDATING OF CALIBRATION RESULTS

--

OVERVIEW OF PLATFORMS SURVIVEDDAMAGED -~

Operator -

- Platforms

Survived CS)

Platforms

Damaged CD)

Platforms Used In

Phase IJIP -

JI~ ~Bill~IRBDll -

Amoco -

Chevron -

Exxon

Mob II

Phllllps

Shell

Unocal -

30

143

72 22

7

39 25shy -

1

12

bull

1 bull

1

1(S)+1 (D)

3 (S) +3(D)

2 (S)

bull shy

bull

bull

bull

Total Partlclnants Platforms 338 - 15 10C6S+4D) -

Other Operators

310

13

3 (D)

Total Platforms 648 28 13C6S+7D) -

Summary of Andrew JIP bull Phase I Platforms

Plairshy w v- N-ber Lea-le 1 shy car Hind_ o-w N-bull Deplh fl fl IJpe la bull n Yeadon Jadtet lbelledtf liefalit -- _ -OD

IL H - -shy _ICllshy IL IL - - rQClto ~ _

K STUIK 137 1963 8 365 608S4 Cellardeck

-middot STl-shy 170 1964 4 -~ K 39 62269 Cellardeck - --ST134W 137 1911 4 43 608S4 Smvlnlshy- _ - -

WD90A 184 1964 8 K 267Sbull34 83 SOS31 SUboftllar deck SmvMI

MC311 343 1978 8 x S7 62 - MC397 468 1991 4 x shy S33bull6SO 65911 -shySmeSmvlvol -- gt -

r-- T23ST52 63 1969 4 ton- shyKshy 52594 S03 -T25SSl39 62 1969 4 Kshy SOS94 S063 - middot ton-shy__ -

bullbull a ton--shyST161A 118 1964 8 K - 347S 51495 Cellarclecti

r~

ST1771 142 1965 8 K 39 60167 CeDordeck failme

STISIH 137 1964 8 K 395 608S4 CeDardeck Failme

K ST130A 140 1958 a SS 6095 I lblv Pai1uno - T21ST72 61 1969 4 Kshy S4S94 4974 I lhlv Pai1uno -

- -

JE

INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON

bull WATER DEPTH

bull bull YEAR OF INSTALLATIONDESIGN

bull PLATFORM ORIENTATION

bull PLATFORM LATITUDE AND LONGITUDE

bull KEY SKETCHES OF PLATFORM STRUCTURAL FRAMING

Number of Legs Plies Leg Batter Number of Conductors bull Deck Elevations (Including sub-cellar deck If any) Jacket Framing Elevations Brace type In Vertical Frames Sizes of Primary Members (leg pile vertical frames braces primary horizontals) Joint Details (can gap etc) LegPiie Annulus Grouting Status

bull middotPOST ANDREW INSPECTION REPORT Significant Wave Height Hs (ft)

bull INFORMATION ON DAMAGES PRIOR TO ANDREW

bull RESULTS FROM IN-HOUSE STUDIES IF AVAILABLE

bull ~RIGINAL DESIGN CRITERIA INFORMATION IF AVAILABLE

bull SOIL SHEAR STRENGTH PROFILE IF AVAILABLE

Platforms Identified for Evaluating Therr Likelihood of Falling Under Unexpected Survival Category

OPlRATOR

middotADA

BLOClt

llmucnJllE

NAME

YEAll

INSTAUJD

WA1ER

DBPTll

DISTANCE

JlllOM SHOllE

(MILU)

Bl

(M)

AMOCO 111113

STl11

WDI075

A

bull D

St

17

120

1n

44

20

10 I

CllllVllQfl STlllO c 61 110 21 11 STIUO D Q 1110 21 11

STl134 p Q 137 29 11

ST0151 G 61 137 11

ST0151 I 3 121 11

ST0151 J Q 140 11

STl151 It 137 11

ST0151 PJl001 Q UT 11

ST0151 PROD-2 Q UT 11

ST017 A 3 140 35 11

WDIUT c 65 214 34 10

llXXON STI055 II 51 6T 14 I

STl15 A t3 34

WDI073

WD0073

Amiddotshy Q

1111 1111

22

22

I

I

WDI073 c 1n 22 I

- WDOCl93-shy II

65 1110 2T A 3 194 31 MOBU 111121

SS IOl7

A B

501

50

50

31

T

I

T 6

MtlllPllY SS114 B T 50 T T

PlllllIPS SS0149 A 5 55 33 T

SAMllDAN STln A 107 T T

SBllLL WDl103 A 65 223 2T WDl103 B 65 221 2T WD0104 c 65 221 2T WD0104 D 6T 20 2T

UNOCAL SS020I

SS 020I SSl20I

SS 0209

SS0209

II p

PURll

A

B

3

n 61

103

97

97

95

100

34

34

34

34 34

bullbull bullbullbull

- SS0209 D u 95 34 I

- SS 1209 SSl253

SS0253

G A

c

6T

2

100

165

1T5

34

54

54

bullbull I

I I

I I

SOUTH MARCH

~LAND

N

SOUTH PaTOI ENE ISWltD AREA

MMS Defined Limits ror Offshore Structures Requiring Inspections Following Andrew

t---Jbulln3 T25bull I bullTtl

bull1$1Htl1Abull bull1SIK1779bull

~

middot-DaTA AREA~T

CRANO rsLE

AREA bull1300

bullll4W

=shy

bullPath of Hurricane Andrew with Platforms Used In Calibration

Likelihood Function vs b All Success Case STISI K STIJOQ ST134W WD90A MCJll MCJ97

I 0000

0 9000

0 8000

07000

-Bo6000

tiiii 0 5000 -

04000

0 3000

0 2000

01000

0 0000

0 05 25

ri1rt middot-middot-middot-middot-middot-

I 5 2

b

---11 middot151K(blsu~ccsI

- bull - bull - fk 130Q(bfU~lCSS)

lie bull134W(blsuccc]

lk-WD90A[blsultCS]

- - - lk middotMC397(bfUtlCS]

--o-- lk-MCJll(bbuctcsJ

--- ll (bl 6 out of 6 Mlttcs case]

_

( Likelihood Functions middotSuccess Cases - p - 1 c I

CATEGORIZATION OF ADDITIONAL PLATFORMS

I I I

bull CATEGORIES

Unexpected Survivals

Expected Survivals

bull APPROACH

Identify Corresponding Platform(s) Evaluated In the JIP

Determine Max Wave Height at Platform Location

Evaluate Whether Wave Hit the Deck

middot - Evaluate Likelihood of R and S to Differ from the Base Cases

Review Joint Information

Assess Special Features

bull IDENTIFY NUMBER OF PLATFORMS IN EACH CATEGORY

I

UPDATING OF CALIBRATION RESULTS

bull INCLUDE EFFECT OF ADDITIONAL PLATFORMS ON bullBbull

Consider Increased 1Uncertalntles In R and S

Develop bullNeW Ukellhood Functions for Different Categories

middotLet bullxbull Number of Ukely Unexpected Survival Cases and y Number of Ukely Expected Survival Cases

Determine Combined Likelihood Function lk (b Iadditional platforms) middot

bull REVISED LIKELIHOOD FUNCTION

lk(b I all cases) = lk(b I 9 platforms) middot x lk(b Ix platforms) x lk(b I y platforms)

bull ESTABLISH REVISED BIAS FACTOR B

f8 (b) u f8 (b) lk (b I all cases)

II

t t ~

shy

APPLICATIONS OF BIAS FACTORS

ii APPLICABLE TO PARTICULAR LOAD AND RESISTANCE RECIPE

bull USE BY THE API COMMITTEES AND MMS

Criteria Revisions

Recommended Design Guidelines Revisions bull

bullbull REASSESSMENT STUDIES AND DECISION MAKING

middot

bull RISK ASSESSMENT STUDIES BY THE OPERATORS

Individual Platforms

Fleet of Platforms

middot IMR Program Implementation

bullbull NOT TO BE USED IN DESIGN

I I fl J

EXAMPLE USE OF BIAS FACTORS

bull PHASE I EXAMPLE APPLICATION

bull EVALUATION OF ANNUAL PROBABILITY OF FAILURE

Establish Annual Maximum Seastate Data

Determine Coefficients to Define Base Shear

Determine Ultimate Capacity of Platform

Determine Probablllty of Failure Versus bullbbull

-bull DETERMINE UPDATED PROBABILITY OF FAILURE

Pf - P [S gt R]

Pf I P [~ gt Rb I b] f 8 (b) db

Pr - J P [S gt Rb I b] f bull 8 (b) db

bull TOTAL PROBABILITIES OF FAILURE

Total Pf - 1 bull Il [1 middotPf] - l Pf bull

Total Pf - 1 bull Il [1middot Pf] - l Pfmiddot

Total Pr = 1 - n [1middot PfbullJ = l Pfshy

I I I

EXAMPLE APPLICATION bull PHASE I

PLATFORM ST 151 K

bull GULF OF MEXICO

bull WATER DEPTH = 137 FT

bull ANNUAL MAXIMUM HS MEDIAN 175 ft COV 031

bull ASSUMED STORM DURATION 3 hoursmiddot

bull ASSUMED NUMBER OF WAVES 800

bull MEDIAN HS FOR STORM HOURS 1 2 AND 3

bull MEDIAN CURRENT VALUE bull

bull DETERMINE ULTIMATE CAPACITY FOR THREE DIRECTIONS

bull DETERMINE ANNUAL FAILURE PROBABILITIES

Broadsidemiddot Diagonal End-on

0037 0025 0007

0071 0044 0018

0022 0016 0004

DETERMINE TOTAL ANNUAL PROBABILITY OF FAILURE

Pf - 007

Pf - 013

Pr - 004

bull

EXAMPLE USE OF BIAS FACTORS - PHASE II middot

bull DETERMINE PREDICTIVE PROBABILITY OF FAILURE

Case - 1

Pf _ S S Pf (b b2) fa1a2 (b b2) db db2

g bull -1 (b1 R I S) - 1

92 - (b2 R2 I S) - 1

Case -2

Pf s ~ [bJ f11 (b) db

Pf [bJ p [g lt 0)

g - bull 1 (b1 R1 I S) - 1

1

bull1) l

I

BENEFITS OF ANDREW JIP bull PHASE II

lt

bull IMPROVED LOAD AND RESISTANCE RECIPE

bull IMPROVED CALIBRATION PROCEDURE

bull MULTIPLE BIAS FACTORS

bull PROCEDURE TO INCLUDE EFFECT OF OTHER PLATFORMS

bull PROCEDURE TO USE MULTIPLE BIAS FACTORS

  • Structure Bookmarks
    • ANDREW PHASE II JOINT1 INDUSTRY PROJECT
    • PROGRESS MEETING NUMBER 1
    • ANDREW PHASE II JOINT INDUSTRY PROJECT PROGRESS MEETING NUMBER 1 NOVEMBER 16 AGENDA
    • ANDREW PHASE II UST OF PARTICIPANTS
    • ANDREW PHASE I OBJECTIVES AND RESULTS
    • ANDREW PHASE II PROJECT OBJECTIVES
    • SUBSEQUENT PROJECTS
    • ANDREW PHASE II PROJECT TEAM
    • ANDREW PHASE II UPDATE MODELING middotAND ANALYSIS RECIPE
    • -ANDREW PHASE II CONDUCTOR RESISTANCE
    • ANDREW PHASE Il WA VE FORCE PROFILE
    • ANDREW PHASE II WA VE FORCE PROFILEmiddot
    • ANDREW PHASE II BRACE MODELING
    • ANDREW PHASE II FOUNDATIONS
    • ANDREW PHASE II middot middot JOINT MODELING
    • ANDREW PHASE Il JOINT MODELING OPTIONS DISCUSSION
    • ANDREW PHASE II JOINT MODEJJNG OPTIONS ILLUSTRATIONS
    • ANDREW PHASE II JOINT MODELING PHASE II METHOD
    • ANDREW PHASE II ANDREW STORM CONDmONS UPDATED
    • Platform Selection bullbullbull1111111111111
    • BasisCriteria for Selection bullbullbull1111111111111
    • CALIBRATION
    • WHERE TO INTRODUCE BIAS FACTOR (B
    • PHASE I -KEY CONCLUSIONS
    • bull CALIBRATION ISSUES PHASE II
    • PHASE II WORK SCOPE (CALIBRATION)
    • SOFTWARE
    • SENSITIVITY STUDIES POSTERIOR DISTRIBUTION OF B
    • WEIGHTING APPROACH bull
    • INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON
    • CATEGORIZATION OF ADDITIONAL PLATFORMS
    • UPDATING OF CALIBRATION RESULTS
    • APPLICATIONS OF BIAS FACTORS
    • EXAMPLE USE OF BIAS FACTORS
    • EXAMPLE APPLICATION bull PHASE I PLATFORM ST 151 K
    • EXAMPLE USE OF BIAS FACTORS -PHASE II middot
    • BENEFITS OF ANDREW JIP bull PHASE II
Page 54:  · . ' ANDREW PHASE II PROJECT OBJECTIVES . ' • ·UPDATE MODELING AND ANALYSIS RECIPE· JOINTS ' ' ·BRACES WAVE PROFILE FOUNDATIONS . ' ' CONDUCTORS· • ANALVZE PLATFORMS USING

Prior Distribution of B Likelihood Functionmiddot Observing Success case STISIK and Posterior Distribution of B

2 00

uo

160

140

120

e 100 L $ uoIshy

0 60 ~ ~ 040

020

0 00

0 00 0 50

_ I 00

b

I 50 2 00

-shy lk [bl STISIK)

--shy Pnor-f(b)

--shy Pos1enor r(bl ISIK)

Prior or B Mean bull 10 COV bull 030

Posterlor of Bbull Meanbull 136COV 0 16

2 50

Likelihood Function and Posterior Distribution Example Platform STISIK

- -

~

- Prior and Posterior Distributions or B Failure cases (ST177B STlSlH ST130A T21) Damage cases (T23 T25 ST161A) Success cases (STlSlK ST130Q ST134W WD90A MC311 MC397)

400

350

300

2 so 0-

I

bull bull

bull bull gt

- bullbull ~ bull bull I bull

_ bull bull bullbull I bull bull bullII V -

bull bull

I bull ~ bull

bullbull bullI bull bullI bull bull bull

middot -~

bull__ ~ middotI middotmiddot

L 2 00

e ~ I SO

-I 00

- 0 so

000 gt

0 00 020 0 40 060 0 80 I 00 120 140 160 110 200

b -

-

shy

Prior-r

- - - - Posterior fLlll failure cases

bull bull bull bull bull bull bull Posterior flall damage cases

- bull bull - bull middot Postenor flall success cases

Posterior flall success fulure damage cases

-

Posterior of B Mean= 119 COV 0101

Posterior Distribution or Blas Factor (8) bull Different Categories and All Cases Combined

shy

l

PHASE I - KEY CONCLUSIONS

bull GLOBAL CORRECTION FACTOR

bull 15 TO 20 CONSERVATISM MEAN)

bull UNEXPECTED SURVIVALS HAVE GREATEST IMPACT

bull DAMAGED CASES MORE INFORMATIVE

bull CALIBRATION ISSUES PHASE II

bull

EFFECT OF IMPROVED CAPACITY ANALYSIS RECIPE

bull APPLICABILITY TO MULTIPLE FAILURE MODES

bull NO FOUNDATION DAMAGE CASE INCLUDED

bull IMPROVED REPRESENTATION OF FULL ANDREW DISTRIBUTION

-bull NUMBER OF PLATFORMS CONSIDERED

I 1 I

I t ) bull ~

PHASE II WORK SCOPE (CALIBRATION)

bull

bull ESTABLISH MULTIPLE BIAS FACTORS

Use Nine Base Case Platforms

middot - Capacity Analysis for Revised recipe

Foundation Blas Factor from Foundation JIP

bull DEVELOP WEIGHTING APPROACH

bull UPDATE MULTIPLE BIAS FACTORS

bullEffect of Platforms Not In the Nine Base Cases

bull EXAMPLE APPLICATION OF BIAS FACTORS I

bullbull

PHASE I PHASE II

CAPACITYbull ANALYSISRu bull

---------------------1----middot

Jc1 - iJbullraquoJ dbull

FttlS frt[f~l ratdhJJ

-

LIKELIHOOD FUNCTION

DEVELOPMENT

lk Jb Jfall1N) bull Jb)

111 Ill Isurvtvoq bull 1 - bl

-middot -------------------shy ~---------------------middot-middot

r111ir8 Cb1 111 Cbl BAYESIAN

UPDATING

~cove

R1R2 bull middot---------------------~----

Jblbull P(g cOJ

1 bull JbRI (MBSI I bull 1

~middot

llcJllJfalluro) bull P(gcOJ Ill 111 b21fallun) bull P ((g1 CO) U 1112 c OD

middot--------------------shy ----------------------shy bull

r111111r8 Cbl 111 Ill i-1nro1

deg ~1b2)laquottfrr (b1b21-lnfo) re Ill bull 1-11middot b2) bull lodegamp

middot

i1 covamp i2 cov~i

-

shy

--

Likelihood Function vs b All Success Case STISI K STl30Q STl34W WD90A MC311 MC397

I 0000

0 9000

OIOOO

0 7000

~ 0amp000

ta osoOO

~ 0 4000

0 3000

0 2000

0 1000

bull 0 0000

bull

I

middot

middot bull I

Ibull

-sibull ~ -middot-middot-middot-middot -middot-

----7----+-----~ I

Ibull bull bull I

rbull

--- Jk ISIK[blsuccess)

- bull - bull - lk 130Q[blsuccess) bull

lk 134W[blsuccess)

Jk-WD90A(blsuccc~I

- - - lk MC397[blsucccss)

--ltgt-- lk middotMC31 l[blsucccss)

--- Jk [bl 6 out of6succcsHascs)

05 I 5 2 250

b

Likelihood Functionsmiddot Success Cases - Phbullsc L

shy

bullbull bull bull bull bull bull -------------------~

bull bull bullbull bullbull bullbull bullbull bullbull bullbull bull

bull bullbull

bullbullbullbull

bullbull

bullbullbullbullbullbullbull

bullbullbullbullbullbull

bullbullbullbullbullbullbull

z 0

t z ) LL

c 0 0c i ~ i

J

~ sect ~--------------~ e l5

I

I

SOFTWARE

bull PF PROGRAM

bull CALREL

bull PF +CALREL

bull RELACS

bull

gt i

l

SENSITIVITY STUDIES

POSTERIOR DISTRIBUTION OF B

bull EFFECT OF CHANGE IN DISTRIBUTION OF PRIOR OF B

bull EFFECT OF REMOVAL OF ONE PLATFORM FROM EACH CATEGORY

bull EFFECT OF REMOVAL OF THREE DAMAGE CASES

bull EFFECT OF ADDITIONAL 13 SIMILAR PLATFORMS

-I

Prior and Posterior Distributions of B Variation In COV of PriorofB fromOl to04 -- -

4500

-4000

-

- -

3500 --middot - bull bull - bull bull Prior-r ICOV =02

3000

- - - - Prior-r ICOV =03 -

bull -~ e 2soo -

bull bull bull bull bull bull bull Pnor-r ICOV =o 4t frJ ~ 2000 Posterior rlCOV =0 2

bull bull Postenor rlCOV =0 3 bull bull 1500 bull middot bull - shy -middot- Posterior rlCOV =04

middot~ I 000 - bull

-middotmiddot ~ ~middotmiddot M -~

- middot 0 500

wbullbullbull _ I middot~ middotmiddotmiddotmiddotmiddot middotmiddotmiddotmiddotmiddot - -~ ~ __ middot_ _ bull -shy0 000

0 000 0200 0 400 0 600 0 800 IOOO 1200 1400 1600 1800 2000 -

- - b

-

=shy

Posterior Distribution of Bias Factor (B) bull Effect of Variation In COV of Prior of B

bull

bullPrior and Posterior Distributions of B Evaluating Effect of Removal of One Platform from Each Category from the Base Case of 13 Platforms

400

~ ~ 350

j IA i

3 00 bull I bull I ---Prior-r I bull l

250 1 I 1

I bull bull ---Posterior r lall 13 casesshy- ~E 2 00 bull bull bull bull bull bull bull Posterior r without STISIK

- ~ 1 bull - bull I - bull bull - bull bull Posterior r Iwithout T2I I 50 I

J-tt~i - bull - bull - Posterior r Iwithout T2S

tOO [-

I I ~ ~ I 0 50

0 00

___ r J i

l~T

~--

0 00 0 20 0 40 0 60 080 I 00 120 140 160 110 2 00

b

zshy

Posterior Distribution of Bias Factor (8) bull Effect of Removal of middotPlatform from Any Category

bullPrior and Posterior Distributions or B Evaluating Effect or Removal or 3 Damage Cases from the Base Case or 13 Platrorms

4

35

3

25

e L 2 ~ bulle shy

I 15

bull I

05

0

0 02 04 06 I 2 I 4 16 IS 2

I I

I

I

I Ibull

J -tmiddot L--

bullI

~

I

fl

II ~

I I v I

middot bull bull

bull L

~ ll- I-~lt

OS I

b

I

bull

1middot bullbullbullbull

~ __

bull

- - - - Prior-r

- - - - Postenor flall failure cases

Postenor flall success cases

--- Posterior flall 13 cases

---Posterior fI success failure cases only

IPosterior or B Mean =t11 COV =0121

-

_I

Posterir Distribution or Bias Factor (B) bull Effect or Removal or 3-Damage Cases

-

bullPrior and Posterior Distributions or B Evaluating Effect or Additional 13 Similar Observations as

Used in Andrew JIP The POsterior or B from Andrew-JIP Considered as New Prior or B

- 500

450

400

350

bull bull

bull bull bull bull J bull

bull bull f ~ I

I ~ Ibull bull

bullbull bull

l bull I -bullbull bull bull

I

I bull bull

bull I I~~bull bull bull

bull ~

I bull bull

~ ~_ ~~~

- - New Prior r bull

Posterior 114 new failure cases ~ 300- bull bull bull bull bull bull bull Posterior f 13 new damage casesl 150 ~

- bull bull - bull bull Posterior 116 new success cases i 2 00

Postenor 1113 new success failure -150 damage cases

I 00

0 50 -

0 00 0 00 0 20 0 40 0 60 0 80 I 00 120 I 40 160 110 2 00

Posterior ol B Mean= 120 COV= 0071 b

shy

shy

_rshy

Posterior Distribution of Bias Factor (B) bull Effect of Additional 13-Similar Observations

1

1 I

WEIGHTING APPROACH bull

bull ASSESSMENT OF PLATFORMS DATA

Select Platforms Most Mappable to the Analyzed Platforms

Unlikely SurvivalDamage Cases Identification

Based on Cut-Off Criteria Agemiddot Hs etc

bull INFORMATION REQUIRED FOR COMPARISON

bull CATEGORIZATION OF ADDITIONALmiddot PLATFORMS

bull UPDATING OF CALIBRATION RESULTS

--

OVERVIEW OF PLATFORMS SURVIVEDDAMAGED -~

Operator -

- Platforms

Survived CS)

Platforms

Damaged CD)

Platforms Used In

Phase IJIP -

JI~ ~Bill~IRBDll -

Amoco -

Chevron -

Exxon

Mob II

Phllllps

Shell

Unocal -

30

143

72 22

7

39 25shy -

1

12

bull

1 bull

1

1(S)+1 (D)

3 (S) +3(D)

2 (S)

bull shy

bull

bull

bull

Total Partlclnants Platforms 338 - 15 10C6S+4D) -

Other Operators

310

13

3 (D)

Total Platforms 648 28 13C6S+7D) -

Summary of Andrew JIP bull Phase I Platforms

Plairshy w v- N-ber Lea-le 1 shy car Hind_ o-w N-bull Deplh fl fl IJpe la bull n Yeadon Jadtet lbelledtf liefalit -- _ -OD

IL H - -shy _ICllshy IL IL - - rQClto ~ _

K STUIK 137 1963 8 365 608S4 Cellardeck

-middot STl-shy 170 1964 4 -~ K 39 62269 Cellardeck - --ST134W 137 1911 4 43 608S4 Smvlnlshy- _ - -

WD90A 184 1964 8 K 267Sbull34 83 SOS31 SUboftllar deck SmvMI

MC311 343 1978 8 x S7 62 - MC397 468 1991 4 x shy S33bull6SO 65911 -shySmeSmvlvol -- gt -

r-- T23ST52 63 1969 4 ton- shyKshy 52594 S03 -T25SSl39 62 1969 4 Kshy SOS94 S063 - middot ton-shy__ -

bullbull a ton--shyST161A 118 1964 8 K - 347S 51495 Cellarclecti

r~

ST1771 142 1965 8 K 39 60167 CeDordeck failme

STISIH 137 1964 8 K 395 608S4 CeDardeck Failme

K ST130A 140 1958 a SS 6095 I lblv Pai1uno - T21ST72 61 1969 4 Kshy S4S94 4974 I lhlv Pai1uno -

- -

JE

INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON

bull WATER DEPTH

bull bull YEAR OF INSTALLATIONDESIGN

bull PLATFORM ORIENTATION

bull PLATFORM LATITUDE AND LONGITUDE

bull KEY SKETCHES OF PLATFORM STRUCTURAL FRAMING

Number of Legs Plies Leg Batter Number of Conductors bull Deck Elevations (Including sub-cellar deck If any) Jacket Framing Elevations Brace type In Vertical Frames Sizes of Primary Members (leg pile vertical frames braces primary horizontals) Joint Details (can gap etc) LegPiie Annulus Grouting Status

bull middotPOST ANDREW INSPECTION REPORT Significant Wave Height Hs (ft)

bull INFORMATION ON DAMAGES PRIOR TO ANDREW

bull RESULTS FROM IN-HOUSE STUDIES IF AVAILABLE

bull ~RIGINAL DESIGN CRITERIA INFORMATION IF AVAILABLE

bull SOIL SHEAR STRENGTH PROFILE IF AVAILABLE

Platforms Identified for Evaluating Therr Likelihood of Falling Under Unexpected Survival Category

OPlRATOR

middotADA

BLOClt

llmucnJllE

NAME

YEAll

INSTAUJD

WA1ER

DBPTll

DISTANCE

JlllOM SHOllE

(MILU)

Bl

(M)

AMOCO 111113

STl11

WDI075

A

bull D

St

17

120

1n

44

20

10 I

CllllVllQfl STlllO c 61 110 21 11 STIUO D Q 1110 21 11

STl134 p Q 137 29 11

ST0151 G 61 137 11

ST0151 I 3 121 11

ST0151 J Q 140 11

STl151 It 137 11

ST0151 PJl001 Q UT 11

ST0151 PROD-2 Q UT 11

ST017 A 3 140 35 11

WDIUT c 65 214 34 10

llXXON STI055 II 51 6T 14 I

STl15 A t3 34

WDI073

WD0073

Amiddotshy Q

1111 1111

22

22

I

I

WDI073 c 1n 22 I

- WDOCl93-shy II

65 1110 2T A 3 194 31 MOBU 111121

SS IOl7

A B

501

50

50

31

T

I

T 6

MtlllPllY SS114 B T 50 T T

PlllllIPS SS0149 A 5 55 33 T

SAMllDAN STln A 107 T T

SBllLL WDl103 A 65 223 2T WDl103 B 65 221 2T WD0104 c 65 221 2T WD0104 D 6T 20 2T

UNOCAL SS020I

SS 020I SSl20I

SS 0209

SS0209

II p

PURll

A

B

3

n 61

103

97

97

95

100

34

34

34

34 34

bullbull bullbullbull

- SS0209 D u 95 34 I

- SS 1209 SSl253

SS0253

G A

c

6T

2

100

165

1T5

34

54

54

bullbull I

I I

I I

SOUTH MARCH

~LAND

N

SOUTH PaTOI ENE ISWltD AREA

MMS Defined Limits ror Offshore Structures Requiring Inspections Following Andrew

t---Jbulln3 T25bull I bullTtl

bull1$1Htl1Abull bull1SIK1779bull

~

middot-DaTA AREA~T

CRANO rsLE

AREA bull1300

bullll4W

=shy

bullPath of Hurricane Andrew with Platforms Used In Calibration

Likelihood Function vs b All Success Case STISI K STIJOQ ST134W WD90A MCJll MCJ97

I 0000

0 9000

0 8000

07000

-Bo6000

tiiii 0 5000 -

04000

0 3000

0 2000

01000

0 0000

0 05 25

ri1rt middot-middot-middot-middot-middot-

I 5 2

b

---11 middot151K(blsu~ccsI

- bull - bull - fk 130Q(bfU~lCSS)

lie bull134W(blsuccc]

lk-WD90A[blsultCS]

- - - lk middotMC397(bfUtlCS]

--o-- lk-MCJll(bbuctcsJ

--- ll (bl 6 out of 6 Mlttcs case]

_

( Likelihood Functions middotSuccess Cases - p - 1 c I

CATEGORIZATION OF ADDITIONAL PLATFORMS

I I I

bull CATEGORIES

Unexpected Survivals

Expected Survivals

bull APPROACH

Identify Corresponding Platform(s) Evaluated In the JIP

Determine Max Wave Height at Platform Location

Evaluate Whether Wave Hit the Deck

middot - Evaluate Likelihood of R and S to Differ from the Base Cases

Review Joint Information

Assess Special Features

bull IDENTIFY NUMBER OF PLATFORMS IN EACH CATEGORY

I

UPDATING OF CALIBRATION RESULTS

bull INCLUDE EFFECT OF ADDITIONAL PLATFORMS ON bullBbull

Consider Increased 1Uncertalntles In R and S

Develop bullNeW Ukellhood Functions for Different Categories

middotLet bullxbull Number of Ukely Unexpected Survival Cases and y Number of Ukely Expected Survival Cases

Determine Combined Likelihood Function lk (b Iadditional platforms) middot

bull REVISED LIKELIHOOD FUNCTION

lk(b I all cases) = lk(b I 9 platforms) middot x lk(b Ix platforms) x lk(b I y platforms)

bull ESTABLISH REVISED BIAS FACTOR B

f8 (b) u f8 (b) lk (b I all cases)

II

t t ~

shy

APPLICATIONS OF BIAS FACTORS

ii APPLICABLE TO PARTICULAR LOAD AND RESISTANCE RECIPE

bull USE BY THE API COMMITTEES AND MMS

Criteria Revisions

Recommended Design Guidelines Revisions bull

bullbull REASSESSMENT STUDIES AND DECISION MAKING

middot

bull RISK ASSESSMENT STUDIES BY THE OPERATORS

Individual Platforms

Fleet of Platforms

middot IMR Program Implementation

bullbull NOT TO BE USED IN DESIGN

I I fl J

EXAMPLE USE OF BIAS FACTORS

bull PHASE I EXAMPLE APPLICATION

bull EVALUATION OF ANNUAL PROBABILITY OF FAILURE

Establish Annual Maximum Seastate Data

Determine Coefficients to Define Base Shear

Determine Ultimate Capacity of Platform

Determine Probablllty of Failure Versus bullbbull

-bull DETERMINE UPDATED PROBABILITY OF FAILURE

Pf - P [S gt R]

Pf I P [~ gt Rb I b] f 8 (b) db

Pr - J P [S gt Rb I b] f bull 8 (b) db

bull TOTAL PROBABILITIES OF FAILURE

Total Pf - 1 bull Il [1 middotPf] - l Pf bull

Total Pf - 1 bull Il [1middot Pf] - l Pfmiddot

Total Pr = 1 - n [1middot PfbullJ = l Pfshy

I I I

EXAMPLE APPLICATION bull PHASE I

PLATFORM ST 151 K

bull GULF OF MEXICO

bull WATER DEPTH = 137 FT

bull ANNUAL MAXIMUM HS MEDIAN 175 ft COV 031

bull ASSUMED STORM DURATION 3 hoursmiddot

bull ASSUMED NUMBER OF WAVES 800

bull MEDIAN HS FOR STORM HOURS 1 2 AND 3

bull MEDIAN CURRENT VALUE bull

bull DETERMINE ULTIMATE CAPACITY FOR THREE DIRECTIONS

bull DETERMINE ANNUAL FAILURE PROBABILITIES

Broadsidemiddot Diagonal End-on

0037 0025 0007

0071 0044 0018

0022 0016 0004

DETERMINE TOTAL ANNUAL PROBABILITY OF FAILURE

Pf - 007

Pf - 013

Pr - 004

bull

EXAMPLE USE OF BIAS FACTORS - PHASE II middot

bull DETERMINE PREDICTIVE PROBABILITY OF FAILURE

Case - 1

Pf _ S S Pf (b b2) fa1a2 (b b2) db db2

g bull -1 (b1 R I S) - 1

92 - (b2 R2 I S) - 1

Case -2

Pf s ~ [bJ f11 (b) db

Pf [bJ p [g lt 0)

g - bull 1 (b1 R1 I S) - 1

1

bull1) l

I

BENEFITS OF ANDREW JIP bull PHASE II

lt

bull IMPROVED LOAD AND RESISTANCE RECIPE

bull IMPROVED CALIBRATION PROCEDURE

bull MULTIPLE BIAS FACTORS

bull PROCEDURE TO INCLUDE EFFECT OF OTHER PLATFORMS

bull PROCEDURE TO USE MULTIPLE BIAS FACTORS

  • Structure Bookmarks
    • ANDREW PHASE II JOINT1 INDUSTRY PROJECT
    • PROGRESS MEETING NUMBER 1
    • ANDREW PHASE II JOINT INDUSTRY PROJECT PROGRESS MEETING NUMBER 1 NOVEMBER 16 AGENDA
    • ANDREW PHASE II UST OF PARTICIPANTS
    • ANDREW PHASE I OBJECTIVES AND RESULTS
    • ANDREW PHASE II PROJECT OBJECTIVES
    • SUBSEQUENT PROJECTS
    • ANDREW PHASE II PROJECT TEAM
    • ANDREW PHASE II UPDATE MODELING middotAND ANALYSIS RECIPE
    • -ANDREW PHASE II CONDUCTOR RESISTANCE
    • ANDREW PHASE Il WA VE FORCE PROFILE
    • ANDREW PHASE II WA VE FORCE PROFILEmiddot
    • ANDREW PHASE II BRACE MODELING
    • ANDREW PHASE II FOUNDATIONS
    • ANDREW PHASE II middot middot JOINT MODELING
    • ANDREW PHASE Il JOINT MODELING OPTIONS DISCUSSION
    • ANDREW PHASE II JOINT MODEJJNG OPTIONS ILLUSTRATIONS
    • ANDREW PHASE II JOINT MODELING PHASE II METHOD
    • ANDREW PHASE II ANDREW STORM CONDmONS UPDATED
    • Platform Selection bullbullbull1111111111111
    • BasisCriteria for Selection bullbullbull1111111111111
    • CALIBRATION
    • WHERE TO INTRODUCE BIAS FACTOR (B
    • PHASE I -KEY CONCLUSIONS
    • bull CALIBRATION ISSUES PHASE II
    • PHASE II WORK SCOPE (CALIBRATION)
    • SOFTWARE
    • SENSITIVITY STUDIES POSTERIOR DISTRIBUTION OF B
    • WEIGHTING APPROACH bull
    • INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON
    • CATEGORIZATION OF ADDITIONAL PLATFORMS
    • UPDATING OF CALIBRATION RESULTS
    • APPLICATIONS OF BIAS FACTORS
    • EXAMPLE USE OF BIAS FACTORS
    • EXAMPLE APPLICATION bull PHASE I PLATFORM ST 151 K
    • EXAMPLE USE OF BIAS FACTORS -PHASE II middot
    • BENEFITS OF ANDREW JIP bull PHASE II
Page 55:  · . ' ANDREW PHASE II PROJECT OBJECTIVES . ' • ·UPDATE MODELING AND ANALYSIS RECIPE· JOINTS ' ' ·BRACES WAVE PROFILE FOUNDATIONS . ' ' CONDUCTORS· • ANALVZE PLATFORMS USING

- -

~

- Prior and Posterior Distributions or B Failure cases (ST177B STlSlH ST130A T21) Damage cases (T23 T25 ST161A) Success cases (STlSlK ST130Q ST134W WD90A MC311 MC397)

400

350

300

2 so 0-

I

bull bull

bull bull gt

- bullbull ~ bull bull I bull

_ bull bull bullbull I bull bull bullII V -

bull bull

I bull ~ bull

bullbull bullI bull bullI bull bull bull

middot -~

bull__ ~ middotI middotmiddot

L 2 00

e ~ I SO

-I 00

- 0 so

000 gt

0 00 020 0 40 060 0 80 I 00 120 140 160 110 200

b -

-

shy

Prior-r

- - - - Posterior fLlll failure cases

bull bull bull bull bull bull bull Posterior flall damage cases

- bull bull - bull middot Postenor flall success cases

Posterior flall success fulure damage cases

-

Posterior of B Mean= 119 COV 0101

Posterior Distribution or Blas Factor (8) bull Different Categories and All Cases Combined

shy

l

PHASE I - KEY CONCLUSIONS

bull GLOBAL CORRECTION FACTOR

bull 15 TO 20 CONSERVATISM MEAN)

bull UNEXPECTED SURVIVALS HAVE GREATEST IMPACT

bull DAMAGED CASES MORE INFORMATIVE

bull CALIBRATION ISSUES PHASE II

bull

EFFECT OF IMPROVED CAPACITY ANALYSIS RECIPE

bull APPLICABILITY TO MULTIPLE FAILURE MODES

bull NO FOUNDATION DAMAGE CASE INCLUDED

bull IMPROVED REPRESENTATION OF FULL ANDREW DISTRIBUTION

-bull NUMBER OF PLATFORMS CONSIDERED

I 1 I

I t ) bull ~

PHASE II WORK SCOPE (CALIBRATION)

bull

bull ESTABLISH MULTIPLE BIAS FACTORS

Use Nine Base Case Platforms

middot - Capacity Analysis for Revised recipe

Foundation Blas Factor from Foundation JIP

bull DEVELOP WEIGHTING APPROACH

bull UPDATE MULTIPLE BIAS FACTORS

bullEffect of Platforms Not In the Nine Base Cases

bull EXAMPLE APPLICATION OF BIAS FACTORS I

bullbull

PHASE I PHASE II

CAPACITYbull ANALYSISRu bull

---------------------1----middot

Jc1 - iJbullraquoJ dbull

FttlS frt[f~l ratdhJJ

-

LIKELIHOOD FUNCTION

DEVELOPMENT

lk Jb Jfall1N) bull Jb)

111 Ill Isurvtvoq bull 1 - bl

-middot -------------------shy ~---------------------middot-middot

r111ir8 Cb1 111 Cbl BAYESIAN

UPDATING

~cove

R1R2 bull middot---------------------~----

Jblbull P(g cOJ

1 bull JbRI (MBSI I bull 1

~middot

llcJllJfalluro) bull P(gcOJ Ill 111 b21fallun) bull P ((g1 CO) U 1112 c OD

middot--------------------shy ----------------------shy bull

r111111r8 Cbl 111 Ill i-1nro1

deg ~1b2)laquottfrr (b1b21-lnfo) re Ill bull 1-11middot b2) bull lodegamp

middot

i1 covamp i2 cov~i

-

shy

--

Likelihood Function vs b All Success Case STISI K STl30Q STl34W WD90A MC311 MC397

I 0000

0 9000

OIOOO

0 7000

~ 0amp000

ta osoOO

~ 0 4000

0 3000

0 2000

0 1000

bull 0 0000

bull

I

middot

middot bull I

Ibull

-sibull ~ -middot-middot-middot-middot -middot-

----7----+-----~ I

Ibull bull bull I

rbull

--- Jk ISIK[blsuccess)

- bull - bull - lk 130Q[blsuccess) bull

lk 134W[blsuccess)

Jk-WD90A(blsuccc~I

- - - lk MC397[blsucccss)

--ltgt-- lk middotMC31 l[blsucccss)

--- Jk [bl 6 out of6succcsHascs)

05 I 5 2 250

b

Likelihood Functionsmiddot Success Cases - Phbullsc L

shy

bullbull bull bull bull bull bull -------------------~

bull bull bullbull bullbull bullbull bullbull bullbull bullbull bull

bull bullbull

bullbullbullbull

bullbull

bullbullbullbullbullbullbull

bullbullbullbullbullbull

bullbullbullbullbullbullbull

z 0

t z ) LL

c 0 0c i ~ i

J

~ sect ~--------------~ e l5

I

I

SOFTWARE

bull PF PROGRAM

bull CALREL

bull PF +CALREL

bull RELACS

bull

gt i

l

SENSITIVITY STUDIES

POSTERIOR DISTRIBUTION OF B

bull EFFECT OF CHANGE IN DISTRIBUTION OF PRIOR OF B

bull EFFECT OF REMOVAL OF ONE PLATFORM FROM EACH CATEGORY

bull EFFECT OF REMOVAL OF THREE DAMAGE CASES

bull EFFECT OF ADDITIONAL 13 SIMILAR PLATFORMS

-I

Prior and Posterior Distributions of B Variation In COV of PriorofB fromOl to04 -- -

4500

-4000

-

- -

3500 --middot - bull bull - bull bull Prior-r ICOV =02

3000

- - - - Prior-r ICOV =03 -

bull -~ e 2soo -

bull bull bull bull bull bull bull Pnor-r ICOV =o 4t frJ ~ 2000 Posterior rlCOV =0 2

bull bull Postenor rlCOV =0 3 bull bull 1500 bull middot bull - shy -middot- Posterior rlCOV =04

middot~ I 000 - bull

-middotmiddot ~ ~middotmiddot M -~

- middot 0 500

wbullbullbull _ I middot~ middotmiddotmiddotmiddotmiddot middotmiddotmiddotmiddotmiddot - -~ ~ __ middot_ _ bull -shy0 000

0 000 0200 0 400 0 600 0 800 IOOO 1200 1400 1600 1800 2000 -

- - b

-

=shy

Posterior Distribution of Bias Factor (B) bull Effect of Variation In COV of Prior of B

bull

bullPrior and Posterior Distributions of B Evaluating Effect of Removal of One Platform from Each Category from the Base Case of 13 Platforms

400

~ ~ 350

j IA i

3 00 bull I bull I ---Prior-r I bull l

250 1 I 1

I bull bull ---Posterior r lall 13 casesshy- ~E 2 00 bull bull bull bull bull bull bull Posterior r without STISIK

- ~ 1 bull - bull I - bull bull - bull bull Posterior r Iwithout T2I I 50 I

J-tt~i - bull - bull - Posterior r Iwithout T2S

tOO [-

I I ~ ~ I 0 50

0 00

___ r J i

l~T

~--

0 00 0 20 0 40 0 60 080 I 00 120 140 160 110 2 00

b

zshy

Posterior Distribution of Bias Factor (8) bull Effect of Removal of middotPlatform from Any Category

bullPrior and Posterior Distributions or B Evaluating Effect or Removal or 3 Damage Cases from the Base Case or 13 Platrorms

4

35

3

25

e L 2 ~ bulle shy

I 15

bull I

05

0

0 02 04 06 I 2 I 4 16 IS 2

I I

I

I

I Ibull

J -tmiddot L--

bullI

~

I

fl

II ~

I I v I

middot bull bull

bull L

~ ll- I-~lt

OS I

b

I

bull

1middot bullbullbullbull

~ __

bull

- - - - Prior-r

- - - - Postenor flall failure cases

Postenor flall success cases

--- Posterior flall 13 cases

---Posterior fI success failure cases only

IPosterior or B Mean =t11 COV =0121

-

_I

Posterir Distribution or Bias Factor (B) bull Effect or Removal or 3-Damage Cases

-

bullPrior and Posterior Distributions or B Evaluating Effect or Additional 13 Similar Observations as

Used in Andrew JIP The POsterior or B from Andrew-JIP Considered as New Prior or B

- 500

450

400

350

bull bull

bull bull bull bull J bull

bull bull f ~ I

I ~ Ibull bull

bullbull bull

l bull I -bullbull bull bull

I

I bull bull

bull I I~~bull bull bull

bull ~

I bull bull

~ ~_ ~~~

- - New Prior r bull

Posterior 114 new failure cases ~ 300- bull bull bull bull bull bull bull Posterior f 13 new damage casesl 150 ~

- bull bull - bull bull Posterior 116 new success cases i 2 00

Postenor 1113 new success failure -150 damage cases

I 00

0 50 -

0 00 0 00 0 20 0 40 0 60 0 80 I 00 120 I 40 160 110 2 00

Posterior ol B Mean= 120 COV= 0071 b

shy

shy

_rshy

Posterior Distribution of Bias Factor (B) bull Effect of Additional 13-Similar Observations

1

1 I

WEIGHTING APPROACH bull

bull ASSESSMENT OF PLATFORMS DATA

Select Platforms Most Mappable to the Analyzed Platforms

Unlikely SurvivalDamage Cases Identification

Based on Cut-Off Criteria Agemiddot Hs etc

bull INFORMATION REQUIRED FOR COMPARISON

bull CATEGORIZATION OF ADDITIONALmiddot PLATFORMS

bull UPDATING OF CALIBRATION RESULTS

--

OVERVIEW OF PLATFORMS SURVIVEDDAMAGED -~

Operator -

- Platforms

Survived CS)

Platforms

Damaged CD)

Platforms Used In

Phase IJIP -

JI~ ~Bill~IRBDll -

Amoco -

Chevron -

Exxon

Mob II

Phllllps

Shell

Unocal -

30

143

72 22

7

39 25shy -

1

12

bull

1 bull

1

1(S)+1 (D)

3 (S) +3(D)

2 (S)

bull shy

bull

bull

bull

Total Partlclnants Platforms 338 - 15 10C6S+4D) -

Other Operators

310

13

3 (D)

Total Platforms 648 28 13C6S+7D) -

Summary of Andrew JIP bull Phase I Platforms

Plairshy w v- N-ber Lea-le 1 shy car Hind_ o-w N-bull Deplh fl fl IJpe la bull n Yeadon Jadtet lbelledtf liefalit -- _ -OD

IL H - -shy _ICllshy IL IL - - rQClto ~ _

K STUIK 137 1963 8 365 608S4 Cellardeck

-middot STl-shy 170 1964 4 -~ K 39 62269 Cellardeck - --ST134W 137 1911 4 43 608S4 Smvlnlshy- _ - -

WD90A 184 1964 8 K 267Sbull34 83 SOS31 SUboftllar deck SmvMI

MC311 343 1978 8 x S7 62 - MC397 468 1991 4 x shy S33bull6SO 65911 -shySmeSmvlvol -- gt -

r-- T23ST52 63 1969 4 ton- shyKshy 52594 S03 -T25SSl39 62 1969 4 Kshy SOS94 S063 - middot ton-shy__ -

bullbull a ton--shyST161A 118 1964 8 K - 347S 51495 Cellarclecti

r~

ST1771 142 1965 8 K 39 60167 CeDordeck failme

STISIH 137 1964 8 K 395 608S4 CeDardeck Failme

K ST130A 140 1958 a SS 6095 I lblv Pai1uno - T21ST72 61 1969 4 Kshy S4S94 4974 I lhlv Pai1uno -

- -

JE

INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON

bull WATER DEPTH

bull bull YEAR OF INSTALLATIONDESIGN

bull PLATFORM ORIENTATION

bull PLATFORM LATITUDE AND LONGITUDE

bull KEY SKETCHES OF PLATFORM STRUCTURAL FRAMING

Number of Legs Plies Leg Batter Number of Conductors bull Deck Elevations (Including sub-cellar deck If any) Jacket Framing Elevations Brace type In Vertical Frames Sizes of Primary Members (leg pile vertical frames braces primary horizontals) Joint Details (can gap etc) LegPiie Annulus Grouting Status

bull middotPOST ANDREW INSPECTION REPORT Significant Wave Height Hs (ft)

bull INFORMATION ON DAMAGES PRIOR TO ANDREW

bull RESULTS FROM IN-HOUSE STUDIES IF AVAILABLE

bull ~RIGINAL DESIGN CRITERIA INFORMATION IF AVAILABLE

bull SOIL SHEAR STRENGTH PROFILE IF AVAILABLE

Platforms Identified for Evaluating Therr Likelihood of Falling Under Unexpected Survival Category

OPlRATOR

middotADA

BLOClt

llmucnJllE

NAME

YEAll

INSTAUJD

WA1ER

DBPTll

DISTANCE

JlllOM SHOllE

(MILU)

Bl

(M)

AMOCO 111113

STl11

WDI075

A

bull D

St

17

120

1n

44

20

10 I

CllllVllQfl STlllO c 61 110 21 11 STIUO D Q 1110 21 11

STl134 p Q 137 29 11

ST0151 G 61 137 11

ST0151 I 3 121 11

ST0151 J Q 140 11

STl151 It 137 11

ST0151 PJl001 Q UT 11

ST0151 PROD-2 Q UT 11

ST017 A 3 140 35 11

WDIUT c 65 214 34 10

llXXON STI055 II 51 6T 14 I

STl15 A t3 34

WDI073

WD0073

Amiddotshy Q

1111 1111

22

22

I

I

WDI073 c 1n 22 I

- WDOCl93-shy II

65 1110 2T A 3 194 31 MOBU 111121

SS IOl7

A B

501

50

50

31

T

I

T 6

MtlllPllY SS114 B T 50 T T

PlllllIPS SS0149 A 5 55 33 T

SAMllDAN STln A 107 T T

SBllLL WDl103 A 65 223 2T WDl103 B 65 221 2T WD0104 c 65 221 2T WD0104 D 6T 20 2T

UNOCAL SS020I

SS 020I SSl20I

SS 0209

SS0209

II p

PURll

A

B

3

n 61

103

97

97

95

100

34

34

34

34 34

bullbull bullbullbull

- SS0209 D u 95 34 I

- SS 1209 SSl253

SS0253

G A

c

6T

2

100

165

1T5

34

54

54

bullbull I

I I

I I

SOUTH MARCH

~LAND

N

SOUTH PaTOI ENE ISWltD AREA

MMS Defined Limits ror Offshore Structures Requiring Inspections Following Andrew

t---Jbulln3 T25bull I bullTtl

bull1$1Htl1Abull bull1SIK1779bull

~

middot-DaTA AREA~T

CRANO rsLE

AREA bull1300

bullll4W

=shy

bullPath of Hurricane Andrew with Platforms Used In Calibration

Likelihood Function vs b All Success Case STISI K STIJOQ ST134W WD90A MCJll MCJ97

I 0000

0 9000

0 8000

07000

-Bo6000

tiiii 0 5000 -

04000

0 3000

0 2000

01000

0 0000

0 05 25

ri1rt middot-middot-middot-middot-middot-

I 5 2

b

---11 middot151K(blsu~ccsI

- bull - bull - fk 130Q(bfU~lCSS)

lie bull134W(blsuccc]

lk-WD90A[blsultCS]

- - - lk middotMC397(bfUtlCS]

--o-- lk-MCJll(bbuctcsJ

--- ll (bl 6 out of 6 Mlttcs case]

_

( Likelihood Functions middotSuccess Cases - p - 1 c I

CATEGORIZATION OF ADDITIONAL PLATFORMS

I I I

bull CATEGORIES

Unexpected Survivals

Expected Survivals

bull APPROACH

Identify Corresponding Platform(s) Evaluated In the JIP

Determine Max Wave Height at Platform Location

Evaluate Whether Wave Hit the Deck

middot - Evaluate Likelihood of R and S to Differ from the Base Cases

Review Joint Information

Assess Special Features

bull IDENTIFY NUMBER OF PLATFORMS IN EACH CATEGORY

I

UPDATING OF CALIBRATION RESULTS

bull INCLUDE EFFECT OF ADDITIONAL PLATFORMS ON bullBbull

Consider Increased 1Uncertalntles In R and S

Develop bullNeW Ukellhood Functions for Different Categories

middotLet bullxbull Number of Ukely Unexpected Survival Cases and y Number of Ukely Expected Survival Cases

Determine Combined Likelihood Function lk (b Iadditional platforms) middot

bull REVISED LIKELIHOOD FUNCTION

lk(b I all cases) = lk(b I 9 platforms) middot x lk(b Ix platforms) x lk(b I y platforms)

bull ESTABLISH REVISED BIAS FACTOR B

f8 (b) u f8 (b) lk (b I all cases)

II

t t ~

shy

APPLICATIONS OF BIAS FACTORS

ii APPLICABLE TO PARTICULAR LOAD AND RESISTANCE RECIPE

bull USE BY THE API COMMITTEES AND MMS

Criteria Revisions

Recommended Design Guidelines Revisions bull

bullbull REASSESSMENT STUDIES AND DECISION MAKING

middot

bull RISK ASSESSMENT STUDIES BY THE OPERATORS

Individual Platforms

Fleet of Platforms

middot IMR Program Implementation

bullbull NOT TO BE USED IN DESIGN

I I fl J

EXAMPLE USE OF BIAS FACTORS

bull PHASE I EXAMPLE APPLICATION

bull EVALUATION OF ANNUAL PROBABILITY OF FAILURE

Establish Annual Maximum Seastate Data

Determine Coefficients to Define Base Shear

Determine Ultimate Capacity of Platform

Determine Probablllty of Failure Versus bullbbull

-bull DETERMINE UPDATED PROBABILITY OF FAILURE

Pf - P [S gt R]

Pf I P [~ gt Rb I b] f 8 (b) db

Pr - J P [S gt Rb I b] f bull 8 (b) db

bull TOTAL PROBABILITIES OF FAILURE

Total Pf - 1 bull Il [1 middotPf] - l Pf bull

Total Pf - 1 bull Il [1middot Pf] - l Pfmiddot

Total Pr = 1 - n [1middot PfbullJ = l Pfshy

I I I

EXAMPLE APPLICATION bull PHASE I

PLATFORM ST 151 K

bull GULF OF MEXICO

bull WATER DEPTH = 137 FT

bull ANNUAL MAXIMUM HS MEDIAN 175 ft COV 031

bull ASSUMED STORM DURATION 3 hoursmiddot

bull ASSUMED NUMBER OF WAVES 800

bull MEDIAN HS FOR STORM HOURS 1 2 AND 3

bull MEDIAN CURRENT VALUE bull

bull DETERMINE ULTIMATE CAPACITY FOR THREE DIRECTIONS

bull DETERMINE ANNUAL FAILURE PROBABILITIES

Broadsidemiddot Diagonal End-on

0037 0025 0007

0071 0044 0018

0022 0016 0004

DETERMINE TOTAL ANNUAL PROBABILITY OF FAILURE

Pf - 007

Pf - 013

Pr - 004

bull

EXAMPLE USE OF BIAS FACTORS - PHASE II middot

bull DETERMINE PREDICTIVE PROBABILITY OF FAILURE

Case - 1

Pf _ S S Pf (b b2) fa1a2 (b b2) db db2

g bull -1 (b1 R I S) - 1

92 - (b2 R2 I S) - 1

Case -2

Pf s ~ [bJ f11 (b) db

Pf [bJ p [g lt 0)

g - bull 1 (b1 R1 I S) - 1

1

bull1) l

I

BENEFITS OF ANDREW JIP bull PHASE II

lt

bull IMPROVED LOAD AND RESISTANCE RECIPE

bull IMPROVED CALIBRATION PROCEDURE

bull MULTIPLE BIAS FACTORS

bull PROCEDURE TO INCLUDE EFFECT OF OTHER PLATFORMS

bull PROCEDURE TO USE MULTIPLE BIAS FACTORS

  • Structure Bookmarks
    • ANDREW PHASE II JOINT1 INDUSTRY PROJECT
    • PROGRESS MEETING NUMBER 1
    • ANDREW PHASE II JOINT INDUSTRY PROJECT PROGRESS MEETING NUMBER 1 NOVEMBER 16 AGENDA
    • ANDREW PHASE II UST OF PARTICIPANTS
    • ANDREW PHASE I OBJECTIVES AND RESULTS
    • ANDREW PHASE II PROJECT OBJECTIVES
    • SUBSEQUENT PROJECTS
    • ANDREW PHASE II PROJECT TEAM
    • ANDREW PHASE II UPDATE MODELING middotAND ANALYSIS RECIPE
    • -ANDREW PHASE II CONDUCTOR RESISTANCE
    • ANDREW PHASE Il WA VE FORCE PROFILE
    • ANDREW PHASE II WA VE FORCE PROFILEmiddot
    • ANDREW PHASE II BRACE MODELING
    • ANDREW PHASE II FOUNDATIONS
    • ANDREW PHASE II middot middot JOINT MODELING
    • ANDREW PHASE Il JOINT MODELING OPTIONS DISCUSSION
    • ANDREW PHASE II JOINT MODEJJNG OPTIONS ILLUSTRATIONS
    • ANDREW PHASE II JOINT MODELING PHASE II METHOD
    • ANDREW PHASE II ANDREW STORM CONDmONS UPDATED
    • Platform Selection bullbullbull1111111111111
    • BasisCriteria for Selection bullbullbull1111111111111
    • CALIBRATION
    • WHERE TO INTRODUCE BIAS FACTOR (B
    • PHASE I -KEY CONCLUSIONS
    • bull CALIBRATION ISSUES PHASE II
    • PHASE II WORK SCOPE (CALIBRATION)
    • SOFTWARE
    • SENSITIVITY STUDIES POSTERIOR DISTRIBUTION OF B
    • WEIGHTING APPROACH bull
    • INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON
    • CATEGORIZATION OF ADDITIONAL PLATFORMS
    • UPDATING OF CALIBRATION RESULTS
    • APPLICATIONS OF BIAS FACTORS
    • EXAMPLE USE OF BIAS FACTORS
    • EXAMPLE APPLICATION bull PHASE I PLATFORM ST 151 K
    • EXAMPLE USE OF BIAS FACTORS -PHASE II middot
    • BENEFITS OF ANDREW JIP bull PHASE II
Page 56:  · . ' ANDREW PHASE II PROJECT OBJECTIVES . ' • ·UPDATE MODELING AND ANALYSIS RECIPE· JOINTS ' ' ·BRACES WAVE PROFILE FOUNDATIONS . ' ' CONDUCTORS· • ANALVZE PLATFORMS USING

l

PHASE I - KEY CONCLUSIONS

bull GLOBAL CORRECTION FACTOR

bull 15 TO 20 CONSERVATISM MEAN)

bull UNEXPECTED SURVIVALS HAVE GREATEST IMPACT

bull DAMAGED CASES MORE INFORMATIVE

bull CALIBRATION ISSUES PHASE II

bull

EFFECT OF IMPROVED CAPACITY ANALYSIS RECIPE

bull APPLICABILITY TO MULTIPLE FAILURE MODES

bull NO FOUNDATION DAMAGE CASE INCLUDED

bull IMPROVED REPRESENTATION OF FULL ANDREW DISTRIBUTION

-bull NUMBER OF PLATFORMS CONSIDERED

I 1 I

I t ) bull ~

PHASE II WORK SCOPE (CALIBRATION)

bull

bull ESTABLISH MULTIPLE BIAS FACTORS

Use Nine Base Case Platforms

middot - Capacity Analysis for Revised recipe

Foundation Blas Factor from Foundation JIP

bull DEVELOP WEIGHTING APPROACH

bull UPDATE MULTIPLE BIAS FACTORS

bullEffect of Platforms Not In the Nine Base Cases

bull EXAMPLE APPLICATION OF BIAS FACTORS I

bullbull

PHASE I PHASE II

CAPACITYbull ANALYSISRu bull

---------------------1----middot

Jc1 - iJbullraquoJ dbull

FttlS frt[f~l ratdhJJ

-

LIKELIHOOD FUNCTION

DEVELOPMENT

lk Jb Jfall1N) bull Jb)

111 Ill Isurvtvoq bull 1 - bl

-middot -------------------shy ~---------------------middot-middot

r111ir8 Cb1 111 Cbl BAYESIAN

UPDATING

~cove

R1R2 bull middot---------------------~----

Jblbull P(g cOJ

1 bull JbRI (MBSI I bull 1

~middot

llcJllJfalluro) bull P(gcOJ Ill 111 b21fallun) bull P ((g1 CO) U 1112 c OD

middot--------------------shy ----------------------shy bull

r111111r8 Cbl 111 Ill i-1nro1

deg ~1b2)laquottfrr (b1b21-lnfo) re Ill bull 1-11middot b2) bull lodegamp

middot

i1 covamp i2 cov~i

-

shy

--

Likelihood Function vs b All Success Case STISI K STl30Q STl34W WD90A MC311 MC397

I 0000

0 9000

OIOOO

0 7000

~ 0amp000

ta osoOO

~ 0 4000

0 3000

0 2000

0 1000

bull 0 0000

bull

I

middot

middot bull I

Ibull

-sibull ~ -middot-middot-middot-middot -middot-

----7----+-----~ I

Ibull bull bull I

rbull

--- Jk ISIK[blsuccess)

- bull - bull - lk 130Q[blsuccess) bull

lk 134W[blsuccess)

Jk-WD90A(blsuccc~I

- - - lk MC397[blsucccss)

--ltgt-- lk middotMC31 l[blsucccss)

--- Jk [bl 6 out of6succcsHascs)

05 I 5 2 250

b

Likelihood Functionsmiddot Success Cases - Phbullsc L

shy

bullbull bull bull bull bull bull -------------------~

bull bull bullbull bullbull bullbull bullbull bullbull bullbull bull

bull bullbull

bullbullbullbull

bullbull

bullbullbullbullbullbullbull

bullbullbullbullbullbull

bullbullbullbullbullbullbull

z 0

t z ) LL

c 0 0c i ~ i

J

~ sect ~--------------~ e l5

I

I

SOFTWARE

bull PF PROGRAM

bull CALREL

bull PF +CALREL

bull RELACS

bull

gt i

l

SENSITIVITY STUDIES

POSTERIOR DISTRIBUTION OF B

bull EFFECT OF CHANGE IN DISTRIBUTION OF PRIOR OF B

bull EFFECT OF REMOVAL OF ONE PLATFORM FROM EACH CATEGORY

bull EFFECT OF REMOVAL OF THREE DAMAGE CASES

bull EFFECT OF ADDITIONAL 13 SIMILAR PLATFORMS

-I

Prior and Posterior Distributions of B Variation In COV of PriorofB fromOl to04 -- -

4500

-4000

-

- -

3500 --middot - bull bull - bull bull Prior-r ICOV =02

3000

- - - - Prior-r ICOV =03 -

bull -~ e 2soo -

bull bull bull bull bull bull bull Pnor-r ICOV =o 4t frJ ~ 2000 Posterior rlCOV =0 2

bull bull Postenor rlCOV =0 3 bull bull 1500 bull middot bull - shy -middot- Posterior rlCOV =04

middot~ I 000 - bull

-middotmiddot ~ ~middotmiddot M -~

- middot 0 500

wbullbullbull _ I middot~ middotmiddotmiddotmiddotmiddot middotmiddotmiddotmiddotmiddot - -~ ~ __ middot_ _ bull -shy0 000

0 000 0200 0 400 0 600 0 800 IOOO 1200 1400 1600 1800 2000 -

- - b

-

=shy

Posterior Distribution of Bias Factor (B) bull Effect of Variation In COV of Prior of B

bull

bullPrior and Posterior Distributions of B Evaluating Effect of Removal of One Platform from Each Category from the Base Case of 13 Platforms

400

~ ~ 350

j IA i

3 00 bull I bull I ---Prior-r I bull l

250 1 I 1

I bull bull ---Posterior r lall 13 casesshy- ~E 2 00 bull bull bull bull bull bull bull Posterior r without STISIK

- ~ 1 bull - bull I - bull bull - bull bull Posterior r Iwithout T2I I 50 I

J-tt~i - bull - bull - Posterior r Iwithout T2S

tOO [-

I I ~ ~ I 0 50

0 00

___ r J i

l~T

~--

0 00 0 20 0 40 0 60 080 I 00 120 140 160 110 2 00

b

zshy

Posterior Distribution of Bias Factor (8) bull Effect of Removal of middotPlatform from Any Category

bullPrior and Posterior Distributions or B Evaluating Effect or Removal or 3 Damage Cases from the Base Case or 13 Platrorms

4

35

3

25

e L 2 ~ bulle shy

I 15

bull I

05

0

0 02 04 06 I 2 I 4 16 IS 2

I I

I

I

I Ibull

J -tmiddot L--

bullI

~

I

fl

II ~

I I v I

middot bull bull

bull L

~ ll- I-~lt

OS I

b

I

bull

1middot bullbullbullbull

~ __

bull

- - - - Prior-r

- - - - Postenor flall failure cases

Postenor flall success cases

--- Posterior flall 13 cases

---Posterior fI success failure cases only

IPosterior or B Mean =t11 COV =0121

-

_I

Posterir Distribution or Bias Factor (B) bull Effect or Removal or 3-Damage Cases

-

bullPrior and Posterior Distributions or B Evaluating Effect or Additional 13 Similar Observations as

Used in Andrew JIP The POsterior or B from Andrew-JIP Considered as New Prior or B

- 500

450

400

350

bull bull

bull bull bull bull J bull

bull bull f ~ I

I ~ Ibull bull

bullbull bull

l bull I -bullbull bull bull

I

I bull bull

bull I I~~bull bull bull

bull ~

I bull bull

~ ~_ ~~~

- - New Prior r bull

Posterior 114 new failure cases ~ 300- bull bull bull bull bull bull bull Posterior f 13 new damage casesl 150 ~

- bull bull - bull bull Posterior 116 new success cases i 2 00

Postenor 1113 new success failure -150 damage cases

I 00

0 50 -

0 00 0 00 0 20 0 40 0 60 0 80 I 00 120 I 40 160 110 2 00

Posterior ol B Mean= 120 COV= 0071 b

shy

shy

_rshy

Posterior Distribution of Bias Factor (B) bull Effect of Additional 13-Similar Observations

1

1 I

WEIGHTING APPROACH bull

bull ASSESSMENT OF PLATFORMS DATA

Select Platforms Most Mappable to the Analyzed Platforms

Unlikely SurvivalDamage Cases Identification

Based on Cut-Off Criteria Agemiddot Hs etc

bull INFORMATION REQUIRED FOR COMPARISON

bull CATEGORIZATION OF ADDITIONALmiddot PLATFORMS

bull UPDATING OF CALIBRATION RESULTS

--

OVERVIEW OF PLATFORMS SURVIVEDDAMAGED -~

Operator -

- Platforms

Survived CS)

Platforms

Damaged CD)

Platforms Used In

Phase IJIP -

JI~ ~Bill~IRBDll -

Amoco -

Chevron -

Exxon

Mob II

Phllllps

Shell

Unocal -

30

143

72 22

7

39 25shy -

1

12

bull

1 bull

1

1(S)+1 (D)

3 (S) +3(D)

2 (S)

bull shy

bull

bull

bull

Total Partlclnants Platforms 338 - 15 10C6S+4D) -

Other Operators

310

13

3 (D)

Total Platforms 648 28 13C6S+7D) -

Summary of Andrew JIP bull Phase I Platforms

Plairshy w v- N-ber Lea-le 1 shy car Hind_ o-w N-bull Deplh fl fl IJpe la bull n Yeadon Jadtet lbelledtf liefalit -- _ -OD

IL H - -shy _ICllshy IL IL - - rQClto ~ _

K STUIK 137 1963 8 365 608S4 Cellardeck

-middot STl-shy 170 1964 4 -~ K 39 62269 Cellardeck - --ST134W 137 1911 4 43 608S4 Smvlnlshy- _ - -

WD90A 184 1964 8 K 267Sbull34 83 SOS31 SUboftllar deck SmvMI

MC311 343 1978 8 x S7 62 - MC397 468 1991 4 x shy S33bull6SO 65911 -shySmeSmvlvol -- gt -

r-- T23ST52 63 1969 4 ton- shyKshy 52594 S03 -T25SSl39 62 1969 4 Kshy SOS94 S063 - middot ton-shy__ -

bullbull a ton--shyST161A 118 1964 8 K - 347S 51495 Cellarclecti

r~

ST1771 142 1965 8 K 39 60167 CeDordeck failme

STISIH 137 1964 8 K 395 608S4 CeDardeck Failme

K ST130A 140 1958 a SS 6095 I lblv Pai1uno - T21ST72 61 1969 4 Kshy S4S94 4974 I lhlv Pai1uno -

- -

JE

INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON

bull WATER DEPTH

bull bull YEAR OF INSTALLATIONDESIGN

bull PLATFORM ORIENTATION

bull PLATFORM LATITUDE AND LONGITUDE

bull KEY SKETCHES OF PLATFORM STRUCTURAL FRAMING

Number of Legs Plies Leg Batter Number of Conductors bull Deck Elevations (Including sub-cellar deck If any) Jacket Framing Elevations Brace type In Vertical Frames Sizes of Primary Members (leg pile vertical frames braces primary horizontals) Joint Details (can gap etc) LegPiie Annulus Grouting Status

bull middotPOST ANDREW INSPECTION REPORT Significant Wave Height Hs (ft)

bull INFORMATION ON DAMAGES PRIOR TO ANDREW

bull RESULTS FROM IN-HOUSE STUDIES IF AVAILABLE

bull ~RIGINAL DESIGN CRITERIA INFORMATION IF AVAILABLE

bull SOIL SHEAR STRENGTH PROFILE IF AVAILABLE

Platforms Identified for Evaluating Therr Likelihood of Falling Under Unexpected Survival Category

OPlRATOR

middotADA

BLOClt

llmucnJllE

NAME

YEAll

INSTAUJD

WA1ER

DBPTll

DISTANCE

JlllOM SHOllE

(MILU)

Bl

(M)

AMOCO 111113

STl11

WDI075

A

bull D

St

17

120

1n

44

20

10 I

CllllVllQfl STlllO c 61 110 21 11 STIUO D Q 1110 21 11

STl134 p Q 137 29 11

ST0151 G 61 137 11

ST0151 I 3 121 11

ST0151 J Q 140 11

STl151 It 137 11

ST0151 PJl001 Q UT 11

ST0151 PROD-2 Q UT 11

ST017 A 3 140 35 11

WDIUT c 65 214 34 10

llXXON STI055 II 51 6T 14 I

STl15 A t3 34

WDI073

WD0073

Amiddotshy Q

1111 1111

22

22

I

I

WDI073 c 1n 22 I

- WDOCl93-shy II

65 1110 2T A 3 194 31 MOBU 111121

SS IOl7

A B

501

50

50

31

T

I

T 6

MtlllPllY SS114 B T 50 T T

PlllllIPS SS0149 A 5 55 33 T

SAMllDAN STln A 107 T T

SBllLL WDl103 A 65 223 2T WDl103 B 65 221 2T WD0104 c 65 221 2T WD0104 D 6T 20 2T

UNOCAL SS020I

SS 020I SSl20I

SS 0209

SS0209

II p

PURll

A

B

3

n 61

103

97

97

95

100

34

34

34

34 34

bullbull bullbullbull

- SS0209 D u 95 34 I

- SS 1209 SSl253

SS0253

G A

c

6T

2

100

165

1T5

34

54

54

bullbull I

I I

I I

SOUTH MARCH

~LAND

N

SOUTH PaTOI ENE ISWltD AREA

MMS Defined Limits ror Offshore Structures Requiring Inspections Following Andrew

t---Jbulln3 T25bull I bullTtl

bull1$1Htl1Abull bull1SIK1779bull

~

middot-DaTA AREA~T

CRANO rsLE

AREA bull1300

bullll4W

=shy

bullPath of Hurricane Andrew with Platforms Used In Calibration

Likelihood Function vs b All Success Case STISI K STIJOQ ST134W WD90A MCJll MCJ97

I 0000

0 9000

0 8000

07000

-Bo6000

tiiii 0 5000 -

04000

0 3000

0 2000

01000

0 0000

0 05 25

ri1rt middot-middot-middot-middot-middot-

I 5 2

b

---11 middot151K(blsu~ccsI

- bull - bull - fk 130Q(bfU~lCSS)

lie bull134W(blsuccc]

lk-WD90A[blsultCS]

- - - lk middotMC397(bfUtlCS]

--o-- lk-MCJll(bbuctcsJ

--- ll (bl 6 out of 6 Mlttcs case]

_

( Likelihood Functions middotSuccess Cases - p - 1 c I

CATEGORIZATION OF ADDITIONAL PLATFORMS

I I I

bull CATEGORIES

Unexpected Survivals

Expected Survivals

bull APPROACH

Identify Corresponding Platform(s) Evaluated In the JIP

Determine Max Wave Height at Platform Location

Evaluate Whether Wave Hit the Deck

middot - Evaluate Likelihood of R and S to Differ from the Base Cases

Review Joint Information

Assess Special Features

bull IDENTIFY NUMBER OF PLATFORMS IN EACH CATEGORY

I

UPDATING OF CALIBRATION RESULTS

bull INCLUDE EFFECT OF ADDITIONAL PLATFORMS ON bullBbull

Consider Increased 1Uncertalntles In R and S

Develop bullNeW Ukellhood Functions for Different Categories

middotLet bullxbull Number of Ukely Unexpected Survival Cases and y Number of Ukely Expected Survival Cases

Determine Combined Likelihood Function lk (b Iadditional platforms) middot

bull REVISED LIKELIHOOD FUNCTION

lk(b I all cases) = lk(b I 9 platforms) middot x lk(b Ix platforms) x lk(b I y platforms)

bull ESTABLISH REVISED BIAS FACTOR B

f8 (b) u f8 (b) lk (b I all cases)

II

t t ~

shy

APPLICATIONS OF BIAS FACTORS

ii APPLICABLE TO PARTICULAR LOAD AND RESISTANCE RECIPE

bull USE BY THE API COMMITTEES AND MMS

Criteria Revisions

Recommended Design Guidelines Revisions bull

bullbull REASSESSMENT STUDIES AND DECISION MAKING

middot

bull RISK ASSESSMENT STUDIES BY THE OPERATORS

Individual Platforms

Fleet of Platforms

middot IMR Program Implementation

bullbull NOT TO BE USED IN DESIGN

I I fl J

EXAMPLE USE OF BIAS FACTORS

bull PHASE I EXAMPLE APPLICATION

bull EVALUATION OF ANNUAL PROBABILITY OF FAILURE

Establish Annual Maximum Seastate Data

Determine Coefficients to Define Base Shear

Determine Ultimate Capacity of Platform

Determine Probablllty of Failure Versus bullbbull

-bull DETERMINE UPDATED PROBABILITY OF FAILURE

Pf - P [S gt R]

Pf I P [~ gt Rb I b] f 8 (b) db

Pr - J P [S gt Rb I b] f bull 8 (b) db

bull TOTAL PROBABILITIES OF FAILURE

Total Pf - 1 bull Il [1 middotPf] - l Pf bull

Total Pf - 1 bull Il [1middot Pf] - l Pfmiddot

Total Pr = 1 - n [1middot PfbullJ = l Pfshy

I I I

EXAMPLE APPLICATION bull PHASE I

PLATFORM ST 151 K

bull GULF OF MEXICO

bull WATER DEPTH = 137 FT

bull ANNUAL MAXIMUM HS MEDIAN 175 ft COV 031

bull ASSUMED STORM DURATION 3 hoursmiddot

bull ASSUMED NUMBER OF WAVES 800

bull MEDIAN HS FOR STORM HOURS 1 2 AND 3

bull MEDIAN CURRENT VALUE bull

bull DETERMINE ULTIMATE CAPACITY FOR THREE DIRECTIONS

bull DETERMINE ANNUAL FAILURE PROBABILITIES

Broadsidemiddot Diagonal End-on

0037 0025 0007

0071 0044 0018

0022 0016 0004

DETERMINE TOTAL ANNUAL PROBABILITY OF FAILURE

Pf - 007

Pf - 013

Pr - 004

bull

EXAMPLE USE OF BIAS FACTORS - PHASE II middot

bull DETERMINE PREDICTIVE PROBABILITY OF FAILURE

Case - 1

Pf _ S S Pf (b b2) fa1a2 (b b2) db db2

g bull -1 (b1 R I S) - 1

92 - (b2 R2 I S) - 1

Case -2

Pf s ~ [bJ f11 (b) db

Pf [bJ p [g lt 0)

g - bull 1 (b1 R1 I S) - 1

1

bull1) l

I

BENEFITS OF ANDREW JIP bull PHASE II

lt

bull IMPROVED LOAD AND RESISTANCE RECIPE

bull IMPROVED CALIBRATION PROCEDURE

bull MULTIPLE BIAS FACTORS

bull PROCEDURE TO INCLUDE EFFECT OF OTHER PLATFORMS

bull PROCEDURE TO USE MULTIPLE BIAS FACTORS

  • Structure Bookmarks
    • ANDREW PHASE II JOINT1 INDUSTRY PROJECT
    • PROGRESS MEETING NUMBER 1
    • ANDREW PHASE II JOINT INDUSTRY PROJECT PROGRESS MEETING NUMBER 1 NOVEMBER 16 AGENDA
    • ANDREW PHASE II UST OF PARTICIPANTS
    • ANDREW PHASE I OBJECTIVES AND RESULTS
    • ANDREW PHASE II PROJECT OBJECTIVES
    • SUBSEQUENT PROJECTS
    • ANDREW PHASE II PROJECT TEAM
    • ANDREW PHASE II UPDATE MODELING middotAND ANALYSIS RECIPE
    • -ANDREW PHASE II CONDUCTOR RESISTANCE
    • ANDREW PHASE Il WA VE FORCE PROFILE
    • ANDREW PHASE II WA VE FORCE PROFILEmiddot
    • ANDREW PHASE II BRACE MODELING
    • ANDREW PHASE II FOUNDATIONS
    • ANDREW PHASE II middot middot JOINT MODELING
    • ANDREW PHASE Il JOINT MODELING OPTIONS DISCUSSION
    • ANDREW PHASE II JOINT MODEJJNG OPTIONS ILLUSTRATIONS
    • ANDREW PHASE II JOINT MODELING PHASE II METHOD
    • ANDREW PHASE II ANDREW STORM CONDmONS UPDATED
    • Platform Selection bullbullbull1111111111111
    • BasisCriteria for Selection bullbullbull1111111111111
    • CALIBRATION
    • WHERE TO INTRODUCE BIAS FACTOR (B
    • PHASE I -KEY CONCLUSIONS
    • bull CALIBRATION ISSUES PHASE II
    • PHASE II WORK SCOPE (CALIBRATION)
    • SOFTWARE
    • SENSITIVITY STUDIES POSTERIOR DISTRIBUTION OF B
    • WEIGHTING APPROACH bull
    • INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON
    • CATEGORIZATION OF ADDITIONAL PLATFORMS
    • UPDATING OF CALIBRATION RESULTS
    • APPLICATIONS OF BIAS FACTORS
    • EXAMPLE USE OF BIAS FACTORS
    • EXAMPLE APPLICATION bull PHASE I PLATFORM ST 151 K
    • EXAMPLE USE OF BIAS FACTORS -PHASE II middot
    • BENEFITS OF ANDREW JIP bull PHASE II
Page 57:  · . ' ANDREW PHASE II PROJECT OBJECTIVES . ' • ·UPDATE MODELING AND ANALYSIS RECIPE· JOINTS ' ' ·BRACES WAVE PROFILE FOUNDATIONS . ' ' CONDUCTORS· • ANALVZE PLATFORMS USING

bull CALIBRATION ISSUES PHASE II

bull

EFFECT OF IMPROVED CAPACITY ANALYSIS RECIPE

bull APPLICABILITY TO MULTIPLE FAILURE MODES

bull NO FOUNDATION DAMAGE CASE INCLUDED

bull IMPROVED REPRESENTATION OF FULL ANDREW DISTRIBUTION

-bull NUMBER OF PLATFORMS CONSIDERED

I 1 I

I t ) bull ~

PHASE II WORK SCOPE (CALIBRATION)

bull

bull ESTABLISH MULTIPLE BIAS FACTORS

Use Nine Base Case Platforms

middot - Capacity Analysis for Revised recipe

Foundation Blas Factor from Foundation JIP

bull DEVELOP WEIGHTING APPROACH

bull UPDATE MULTIPLE BIAS FACTORS

bullEffect of Platforms Not In the Nine Base Cases

bull EXAMPLE APPLICATION OF BIAS FACTORS I

bullbull

PHASE I PHASE II

CAPACITYbull ANALYSISRu bull

---------------------1----middot

Jc1 - iJbullraquoJ dbull

FttlS frt[f~l ratdhJJ

-

LIKELIHOOD FUNCTION

DEVELOPMENT

lk Jb Jfall1N) bull Jb)

111 Ill Isurvtvoq bull 1 - bl

-middot -------------------shy ~---------------------middot-middot

r111ir8 Cb1 111 Cbl BAYESIAN

UPDATING

~cove

R1R2 bull middot---------------------~----

Jblbull P(g cOJ

1 bull JbRI (MBSI I bull 1

~middot

llcJllJfalluro) bull P(gcOJ Ill 111 b21fallun) bull P ((g1 CO) U 1112 c OD

middot--------------------shy ----------------------shy bull

r111111r8 Cbl 111 Ill i-1nro1

deg ~1b2)laquottfrr (b1b21-lnfo) re Ill bull 1-11middot b2) bull lodegamp

middot

i1 covamp i2 cov~i

-

shy

--

Likelihood Function vs b All Success Case STISI K STl30Q STl34W WD90A MC311 MC397

I 0000

0 9000

OIOOO

0 7000

~ 0amp000

ta osoOO

~ 0 4000

0 3000

0 2000

0 1000

bull 0 0000

bull

I

middot

middot bull I

Ibull

-sibull ~ -middot-middot-middot-middot -middot-

----7----+-----~ I

Ibull bull bull I

rbull

--- Jk ISIK[blsuccess)

- bull - bull - lk 130Q[blsuccess) bull

lk 134W[blsuccess)

Jk-WD90A(blsuccc~I

- - - lk MC397[blsucccss)

--ltgt-- lk middotMC31 l[blsucccss)

--- Jk [bl 6 out of6succcsHascs)

05 I 5 2 250

b

Likelihood Functionsmiddot Success Cases - Phbullsc L

shy

bullbull bull bull bull bull bull -------------------~

bull bull bullbull bullbull bullbull bullbull bullbull bullbull bull

bull bullbull

bullbullbullbull

bullbull

bullbullbullbullbullbullbull

bullbullbullbullbullbull

bullbullbullbullbullbullbull

z 0

t z ) LL

c 0 0c i ~ i

J

~ sect ~--------------~ e l5

I

I

SOFTWARE

bull PF PROGRAM

bull CALREL

bull PF +CALREL

bull RELACS

bull

gt i

l

SENSITIVITY STUDIES

POSTERIOR DISTRIBUTION OF B

bull EFFECT OF CHANGE IN DISTRIBUTION OF PRIOR OF B

bull EFFECT OF REMOVAL OF ONE PLATFORM FROM EACH CATEGORY

bull EFFECT OF REMOVAL OF THREE DAMAGE CASES

bull EFFECT OF ADDITIONAL 13 SIMILAR PLATFORMS

-I

Prior and Posterior Distributions of B Variation In COV of PriorofB fromOl to04 -- -

4500

-4000

-

- -

3500 --middot - bull bull - bull bull Prior-r ICOV =02

3000

- - - - Prior-r ICOV =03 -

bull -~ e 2soo -

bull bull bull bull bull bull bull Pnor-r ICOV =o 4t frJ ~ 2000 Posterior rlCOV =0 2

bull bull Postenor rlCOV =0 3 bull bull 1500 bull middot bull - shy -middot- Posterior rlCOV =04

middot~ I 000 - bull

-middotmiddot ~ ~middotmiddot M -~

- middot 0 500

wbullbullbull _ I middot~ middotmiddotmiddotmiddotmiddot middotmiddotmiddotmiddotmiddot - -~ ~ __ middot_ _ bull -shy0 000

0 000 0200 0 400 0 600 0 800 IOOO 1200 1400 1600 1800 2000 -

- - b

-

=shy

Posterior Distribution of Bias Factor (B) bull Effect of Variation In COV of Prior of B

bull

bullPrior and Posterior Distributions of B Evaluating Effect of Removal of One Platform from Each Category from the Base Case of 13 Platforms

400

~ ~ 350

j IA i

3 00 bull I bull I ---Prior-r I bull l

250 1 I 1

I bull bull ---Posterior r lall 13 casesshy- ~E 2 00 bull bull bull bull bull bull bull Posterior r without STISIK

- ~ 1 bull - bull I - bull bull - bull bull Posterior r Iwithout T2I I 50 I

J-tt~i - bull - bull - Posterior r Iwithout T2S

tOO [-

I I ~ ~ I 0 50

0 00

___ r J i

l~T

~--

0 00 0 20 0 40 0 60 080 I 00 120 140 160 110 2 00

b

zshy

Posterior Distribution of Bias Factor (8) bull Effect of Removal of middotPlatform from Any Category

bullPrior and Posterior Distributions or B Evaluating Effect or Removal or 3 Damage Cases from the Base Case or 13 Platrorms

4

35

3

25

e L 2 ~ bulle shy

I 15

bull I

05

0

0 02 04 06 I 2 I 4 16 IS 2

I I

I

I

I Ibull

J -tmiddot L--

bullI

~

I

fl

II ~

I I v I

middot bull bull

bull L

~ ll- I-~lt

OS I

b

I

bull

1middot bullbullbullbull

~ __

bull

- - - - Prior-r

- - - - Postenor flall failure cases

Postenor flall success cases

--- Posterior flall 13 cases

---Posterior fI success failure cases only

IPosterior or B Mean =t11 COV =0121

-

_I

Posterir Distribution or Bias Factor (B) bull Effect or Removal or 3-Damage Cases

-

bullPrior and Posterior Distributions or B Evaluating Effect or Additional 13 Similar Observations as

Used in Andrew JIP The POsterior or B from Andrew-JIP Considered as New Prior or B

- 500

450

400

350

bull bull

bull bull bull bull J bull

bull bull f ~ I

I ~ Ibull bull

bullbull bull

l bull I -bullbull bull bull

I

I bull bull

bull I I~~bull bull bull

bull ~

I bull bull

~ ~_ ~~~

- - New Prior r bull

Posterior 114 new failure cases ~ 300- bull bull bull bull bull bull bull Posterior f 13 new damage casesl 150 ~

- bull bull - bull bull Posterior 116 new success cases i 2 00

Postenor 1113 new success failure -150 damage cases

I 00

0 50 -

0 00 0 00 0 20 0 40 0 60 0 80 I 00 120 I 40 160 110 2 00

Posterior ol B Mean= 120 COV= 0071 b

shy

shy

_rshy

Posterior Distribution of Bias Factor (B) bull Effect of Additional 13-Similar Observations

1

1 I

WEIGHTING APPROACH bull

bull ASSESSMENT OF PLATFORMS DATA

Select Platforms Most Mappable to the Analyzed Platforms

Unlikely SurvivalDamage Cases Identification

Based on Cut-Off Criteria Agemiddot Hs etc

bull INFORMATION REQUIRED FOR COMPARISON

bull CATEGORIZATION OF ADDITIONALmiddot PLATFORMS

bull UPDATING OF CALIBRATION RESULTS

--

OVERVIEW OF PLATFORMS SURVIVEDDAMAGED -~

Operator -

- Platforms

Survived CS)

Platforms

Damaged CD)

Platforms Used In

Phase IJIP -

JI~ ~Bill~IRBDll -

Amoco -

Chevron -

Exxon

Mob II

Phllllps

Shell

Unocal -

30

143

72 22

7

39 25shy -

1

12

bull

1 bull

1

1(S)+1 (D)

3 (S) +3(D)

2 (S)

bull shy

bull

bull

bull

Total Partlclnants Platforms 338 - 15 10C6S+4D) -

Other Operators

310

13

3 (D)

Total Platforms 648 28 13C6S+7D) -

Summary of Andrew JIP bull Phase I Platforms

Plairshy w v- N-ber Lea-le 1 shy car Hind_ o-w N-bull Deplh fl fl IJpe la bull n Yeadon Jadtet lbelledtf liefalit -- _ -OD

IL H - -shy _ICllshy IL IL - - rQClto ~ _

K STUIK 137 1963 8 365 608S4 Cellardeck

-middot STl-shy 170 1964 4 -~ K 39 62269 Cellardeck - --ST134W 137 1911 4 43 608S4 Smvlnlshy- _ - -

WD90A 184 1964 8 K 267Sbull34 83 SOS31 SUboftllar deck SmvMI

MC311 343 1978 8 x S7 62 - MC397 468 1991 4 x shy S33bull6SO 65911 -shySmeSmvlvol -- gt -

r-- T23ST52 63 1969 4 ton- shyKshy 52594 S03 -T25SSl39 62 1969 4 Kshy SOS94 S063 - middot ton-shy__ -

bullbull a ton--shyST161A 118 1964 8 K - 347S 51495 Cellarclecti

r~

ST1771 142 1965 8 K 39 60167 CeDordeck failme

STISIH 137 1964 8 K 395 608S4 CeDardeck Failme

K ST130A 140 1958 a SS 6095 I lblv Pai1uno - T21ST72 61 1969 4 Kshy S4S94 4974 I lhlv Pai1uno -

- -

JE

INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON

bull WATER DEPTH

bull bull YEAR OF INSTALLATIONDESIGN

bull PLATFORM ORIENTATION

bull PLATFORM LATITUDE AND LONGITUDE

bull KEY SKETCHES OF PLATFORM STRUCTURAL FRAMING

Number of Legs Plies Leg Batter Number of Conductors bull Deck Elevations (Including sub-cellar deck If any) Jacket Framing Elevations Brace type In Vertical Frames Sizes of Primary Members (leg pile vertical frames braces primary horizontals) Joint Details (can gap etc) LegPiie Annulus Grouting Status

bull middotPOST ANDREW INSPECTION REPORT Significant Wave Height Hs (ft)

bull INFORMATION ON DAMAGES PRIOR TO ANDREW

bull RESULTS FROM IN-HOUSE STUDIES IF AVAILABLE

bull ~RIGINAL DESIGN CRITERIA INFORMATION IF AVAILABLE

bull SOIL SHEAR STRENGTH PROFILE IF AVAILABLE

Platforms Identified for Evaluating Therr Likelihood of Falling Under Unexpected Survival Category

OPlRATOR

middotADA

BLOClt

llmucnJllE

NAME

YEAll

INSTAUJD

WA1ER

DBPTll

DISTANCE

JlllOM SHOllE

(MILU)

Bl

(M)

AMOCO 111113

STl11

WDI075

A

bull D

St

17

120

1n

44

20

10 I

CllllVllQfl STlllO c 61 110 21 11 STIUO D Q 1110 21 11

STl134 p Q 137 29 11

ST0151 G 61 137 11

ST0151 I 3 121 11

ST0151 J Q 140 11

STl151 It 137 11

ST0151 PJl001 Q UT 11

ST0151 PROD-2 Q UT 11

ST017 A 3 140 35 11

WDIUT c 65 214 34 10

llXXON STI055 II 51 6T 14 I

STl15 A t3 34

WDI073

WD0073

Amiddotshy Q

1111 1111

22

22

I

I

WDI073 c 1n 22 I

- WDOCl93-shy II

65 1110 2T A 3 194 31 MOBU 111121

SS IOl7

A B

501

50

50

31

T

I

T 6

MtlllPllY SS114 B T 50 T T

PlllllIPS SS0149 A 5 55 33 T

SAMllDAN STln A 107 T T

SBllLL WDl103 A 65 223 2T WDl103 B 65 221 2T WD0104 c 65 221 2T WD0104 D 6T 20 2T

UNOCAL SS020I

SS 020I SSl20I

SS 0209

SS0209

II p

PURll

A

B

3

n 61

103

97

97

95

100

34

34

34

34 34

bullbull bullbullbull

- SS0209 D u 95 34 I

- SS 1209 SSl253

SS0253

G A

c

6T

2

100

165

1T5

34

54

54

bullbull I

I I

I I

SOUTH MARCH

~LAND

N

SOUTH PaTOI ENE ISWltD AREA

MMS Defined Limits ror Offshore Structures Requiring Inspections Following Andrew

t---Jbulln3 T25bull I bullTtl

bull1$1Htl1Abull bull1SIK1779bull

~

middot-DaTA AREA~T

CRANO rsLE

AREA bull1300

bullll4W

=shy

bullPath of Hurricane Andrew with Platforms Used In Calibration

Likelihood Function vs b All Success Case STISI K STIJOQ ST134W WD90A MCJll MCJ97

I 0000

0 9000

0 8000

07000

-Bo6000

tiiii 0 5000 -

04000

0 3000

0 2000

01000

0 0000

0 05 25

ri1rt middot-middot-middot-middot-middot-

I 5 2

b

---11 middot151K(blsu~ccsI

- bull - bull - fk 130Q(bfU~lCSS)

lie bull134W(blsuccc]

lk-WD90A[blsultCS]

- - - lk middotMC397(bfUtlCS]

--o-- lk-MCJll(bbuctcsJ

--- ll (bl 6 out of 6 Mlttcs case]

_

( Likelihood Functions middotSuccess Cases - p - 1 c I

CATEGORIZATION OF ADDITIONAL PLATFORMS

I I I

bull CATEGORIES

Unexpected Survivals

Expected Survivals

bull APPROACH

Identify Corresponding Platform(s) Evaluated In the JIP

Determine Max Wave Height at Platform Location

Evaluate Whether Wave Hit the Deck

middot - Evaluate Likelihood of R and S to Differ from the Base Cases

Review Joint Information

Assess Special Features

bull IDENTIFY NUMBER OF PLATFORMS IN EACH CATEGORY

I

UPDATING OF CALIBRATION RESULTS

bull INCLUDE EFFECT OF ADDITIONAL PLATFORMS ON bullBbull

Consider Increased 1Uncertalntles In R and S

Develop bullNeW Ukellhood Functions for Different Categories

middotLet bullxbull Number of Ukely Unexpected Survival Cases and y Number of Ukely Expected Survival Cases

Determine Combined Likelihood Function lk (b Iadditional platforms) middot

bull REVISED LIKELIHOOD FUNCTION

lk(b I all cases) = lk(b I 9 platforms) middot x lk(b Ix platforms) x lk(b I y platforms)

bull ESTABLISH REVISED BIAS FACTOR B

f8 (b) u f8 (b) lk (b I all cases)

II

t t ~

shy

APPLICATIONS OF BIAS FACTORS

ii APPLICABLE TO PARTICULAR LOAD AND RESISTANCE RECIPE

bull USE BY THE API COMMITTEES AND MMS

Criteria Revisions

Recommended Design Guidelines Revisions bull

bullbull REASSESSMENT STUDIES AND DECISION MAKING

middot

bull RISK ASSESSMENT STUDIES BY THE OPERATORS

Individual Platforms

Fleet of Platforms

middot IMR Program Implementation

bullbull NOT TO BE USED IN DESIGN

I I fl J

EXAMPLE USE OF BIAS FACTORS

bull PHASE I EXAMPLE APPLICATION

bull EVALUATION OF ANNUAL PROBABILITY OF FAILURE

Establish Annual Maximum Seastate Data

Determine Coefficients to Define Base Shear

Determine Ultimate Capacity of Platform

Determine Probablllty of Failure Versus bullbbull

-bull DETERMINE UPDATED PROBABILITY OF FAILURE

Pf - P [S gt R]

Pf I P [~ gt Rb I b] f 8 (b) db

Pr - J P [S gt Rb I b] f bull 8 (b) db

bull TOTAL PROBABILITIES OF FAILURE

Total Pf - 1 bull Il [1 middotPf] - l Pf bull

Total Pf - 1 bull Il [1middot Pf] - l Pfmiddot

Total Pr = 1 - n [1middot PfbullJ = l Pfshy

I I I

EXAMPLE APPLICATION bull PHASE I

PLATFORM ST 151 K

bull GULF OF MEXICO

bull WATER DEPTH = 137 FT

bull ANNUAL MAXIMUM HS MEDIAN 175 ft COV 031

bull ASSUMED STORM DURATION 3 hoursmiddot

bull ASSUMED NUMBER OF WAVES 800

bull MEDIAN HS FOR STORM HOURS 1 2 AND 3

bull MEDIAN CURRENT VALUE bull

bull DETERMINE ULTIMATE CAPACITY FOR THREE DIRECTIONS

bull DETERMINE ANNUAL FAILURE PROBABILITIES

Broadsidemiddot Diagonal End-on

0037 0025 0007

0071 0044 0018

0022 0016 0004

DETERMINE TOTAL ANNUAL PROBABILITY OF FAILURE

Pf - 007

Pf - 013

Pr - 004

bull

EXAMPLE USE OF BIAS FACTORS - PHASE II middot

bull DETERMINE PREDICTIVE PROBABILITY OF FAILURE

Case - 1

Pf _ S S Pf (b b2) fa1a2 (b b2) db db2

g bull -1 (b1 R I S) - 1

92 - (b2 R2 I S) - 1

Case -2

Pf s ~ [bJ f11 (b) db

Pf [bJ p [g lt 0)

g - bull 1 (b1 R1 I S) - 1

1

bull1) l

I

BENEFITS OF ANDREW JIP bull PHASE II

lt

bull IMPROVED LOAD AND RESISTANCE RECIPE

bull IMPROVED CALIBRATION PROCEDURE

bull MULTIPLE BIAS FACTORS

bull PROCEDURE TO INCLUDE EFFECT OF OTHER PLATFORMS

bull PROCEDURE TO USE MULTIPLE BIAS FACTORS

  • Structure Bookmarks
    • ANDREW PHASE II JOINT1 INDUSTRY PROJECT
    • PROGRESS MEETING NUMBER 1
    • ANDREW PHASE II JOINT INDUSTRY PROJECT PROGRESS MEETING NUMBER 1 NOVEMBER 16 AGENDA
    • ANDREW PHASE II UST OF PARTICIPANTS
    • ANDREW PHASE I OBJECTIVES AND RESULTS
    • ANDREW PHASE II PROJECT OBJECTIVES
    • SUBSEQUENT PROJECTS
    • ANDREW PHASE II PROJECT TEAM
    • ANDREW PHASE II UPDATE MODELING middotAND ANALYSIS RECIPE
    • -ANDREW PHASE II CONDUCTOR RESISTANCE
    • ANDREW PHASE Il WA VE FORCE PROFILE
    • ANDREW PHASE II WA VE FORCE PROFILEmiddot
    • ANDREW PHASE II BRACE MODELING
    • ANDREW PHASE II FOUNDATIONS
    • ANDREW PHASE II middot middot JOINT MODELING
    • ANDREW PHASE Il JOINT MODELING OPTIONS DISCUSSION
    • ANDREW PHASE II JOINT MODEJJNG OPTIONS ILLUSTRATIONS
    • ANDREW PHASE II JOINT MODELING PHASE II METHOD
    • ANDREW PHASE II ANDREW STORM CONDmONS UPDATED
    • Platform Selection bullbullbull1111111111111
    • BasisCriteria for Selection bullbullbull1111111111111
    • CALIBRATION
    • WHERE TO INTRODUCE BIAS FACTOR (B
    • PHASE I -KEY CONCLUSIONS
    • bull CALIBRATION ISSUES PHASE II
    • PHASE II WORK SCOPE (CALIBRATION)
    • SOFTWARE
    • SENSITIVITY STUDIES POSTERIOR DISTRIBUTION OF B
    • WEIGHTING APPROACH bull
    • INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON
    • CATEGORIZATION OF ADDITIONAL PLATFORMS
    • UPDATING OF CALIBRATION RESULTS
    • APPLICATIONS OF BIAS FACTORS
    • EXAMPLE USE OF BIAS FACTORS
    • EXAMPLE APPLICATION bull PHASE I PLATFORM ST 151 K
    • EXAMPLE USE OF BIAS FACTORS -PHASE II middot
    • BENEFITS OF ANDREW JIP bull PHASE II
Page 58:  · . ' ANDREW PHASE II PROJECT OBJECTIVES . ' • ·UPDATE MODELING AND ANALYSIS RECIPE· JOINTS ' ' ·BRACES WAVE PROFILE FOUNDATIONS . ' ' CONDUCTORS· • ANALVZE PLATFORMS USING

I 1 I

I t ) bull ~

PHASE II WORK SCOPE (CALIBRATION)

bull

bull ESTABLISH MULTIPLE BIAS FACTORS

Use Nine Base Case Platforms

middot - Capacity Analysis for Revised recipe

Foundation Blas Factor from Foundation JIP

bull DEVELOP WEIGHTING APPROACH

bull UPDATE MULTIPLE BIAS FACTORS

bullEffect of Platforms Not In the Nine Base Cases

bull EXAMPLE APPLICATION OF BIAS FACTORS I

bullbull

PHASE I PHASE II

CAPACITYbull ANALYSISRu bull

---------------------1----middot

Jc1 - iJbullraquoJ dbull

FttlS frt[f~l ratdhJJ

-

LIKELIHOOD FUNCTION

DEVELOPMENT

lk Jb Jfall1N) bull Jb)

111 Ill Isurvtvoq bull 1 - bl

-middot -------------------shy ~---------------------middot-middot

r111ir8 Cb1 111 Cbl BAYESIAN

UPDATING

~cove

R1R2 bull middot---------------------~----

Jblbull P(g cOJ

1 bull JbRI (MBSI I bull 1

~middot

llcJllJfalluro) bull P(gcOJ Ill 111 b21fallun) bull P ((g1 CO) U 1112 c OD

middot--------------------shy ----------------------shy bull

r111111r8 Cbl 111 Ill i-1nro1

deg ~1b2)laquottfrr (b1b21-lnfo) re Ill bull 1-11middot b2) bull lodegamp

middot

i1 covamp i2 cov~i

-

shy

--

Likelihood Function vs b All Success Case STISI K STl30Q STl34W WD90A MC311 MC397

I 0000

0 9000

OIOOO

0 7000

~ 0amp000

ta osoOO

~ 0 4000

0 3000

0 2000

0 1000

bull 0 0000

bull

I

middot

middot bull I

Ibull

-sibull ~ -middot-middot-middot-middot -middot-

----7----+-----~ I

Ibull bull bull I

rbull

--- Jk ISIK[blsuccess)

- bull - bull - lk 130Q[blsuccess) bull

lk 134W[blsuccess)

Jk-WD90A(blsuccc~I

- - - lk MC397[blsucccss)

--ltgt-- lk middotMC31 l[blsucccss)

--- Jk [bl 6 out of6succcsHascs)

05 I 5 2 250

b

Likelihood Functionsmiddot Success Cases - Phbullsc L

shy

bullbull bull bull bull bull bull -------------------~

bull bull bullbull bullbull bullbull bullbull bullbull bullbull bull

bull bullbull

bullbullbullbull

bullbull

bullbullbullbullbullbullbull

bullbullbullbullbullbull

bullbullbullbullbullbullbull

z 0

t z ) LL

c 0 0c i ~ i

J

~ sect ~--------------~ e l5

I

I

SOFTWARE

bull PF PROGRAM

bull CALREL

bull PF +CALREL

bull RELACS

bull

gt i

l

SENSITIVITY STUDIES

POSTERIOR DISTRIBUTION OF B

bull EFFECT OF CHANGE IN DISTRIBUTION OF PRIOR OF B

bull EFFECT OF REMOVAL OF ONE PLATFORM FROM EACH CATEGORY

bull EFFECT OF REMOVAL OF THREE DAMAGE CASES

bull EFFECT OF ADDITIONAL 13 SIMILAR PLATFORMS

-I

Prior and Posterior Distributions of B Variation In COV of PriorofB fromOl to04 -- -

4500

-4000

-

- -

3500 --middot - bull bull - bull bull Prior-r ICOV =02

3000

- - - - Prior-r ICOV =03 -

bull -~ e 2soo -

bull bull bull bull bull bull bull Pnor-r ICOV =o 4t frJ ~ 2000 Posterior rlCOV =0 2

bull bull Postenor rlCOV =0 3 bull bull 1500 bull middot bull - shy -middot- Posterior rlCOV =04

middot~ I 000 - bull

-middotmiddot ~ ~middotmiddot M -~

- middot 0 500

wbullbullbull _ I middot~ middotmiddotmiddotmiddotmiddot middotmiddotmiddotmiddotmiddot - -~ ~ __ middot_ _ bull -shy0 000

0 000 0200 0 400 0 600 0 800 IOOO 1200 1400 1600 1800 2000 -

- - b

-

=shy

Posterior Distribution of Bias Factor (B) bull Effect of Variation In COV of Prior of B

bull

bullPrior and Posterior Distributions of B Evaluating Effect of Removal of One Platform from Each Category from the Base Case of 13 Platforms

400

~ ~ 350

j IA i

3 00 bull I bull I ---Prior-r I bull l

250 1 I 1

I bull bull ---Posterior r lall 13 casesshy- ~E 2 00 bull bull bull bull bull bull bull Posterior r without STISIK

- ~ 1 bull - bull I - bull bull - bull bull Posterior r Iwithout T2I I 50 I

J-tt~i - bull - bull - Posterior r Iwithout T2S

tOO [-

I I ~ ~ I 0 50

0 00

___ r J i

l~T

~--

0 00 0 20 0 40 0 60 080 I 00 120 140 160 110 2 00

b

zshy

Posterior Distribution of Bias Factor (8) bull Effect of Removal of middotPlatform from Any Category

bullPrior and Posterior Distributions or B Evaluating Effect or Removal or 3 Damage Cases from the Base Case or 13 Platrorms

4

35

3

25

e L 2 ~ bulle shy

I 15

bull I

05

0

0 02 04 06 I 2 I 4 16 IS 2

I I

I

I

I Ibull

J -tmiddot L--

bullI

~

I

fl

II ~

I I v I

middot bull bull

bull L

~ ll- I-~lt

OS I

b

I

bull

1middot bullbullbullbull

~ __

bull

- - - - Prior-r

- - - - Postenor flall failure cases

Postenor flall success cases

--- Posterior flall 13 cases

---Posterior fI success failure cases only

IPosterior or B Mean =t11 COV =0121

-

_I

Posterir Distribution or Bias Factor (B) bull Effect or Removal or 3-Damage Cases

-

bullPrior and Posterior Distributions or B Evaluating Effect or Additional 13 Similar Observations as

Used in Andrew JIP The POsterior or B from Andrew-JIP Considered as New Prior or B

- 500

450

400

350

bull bull

bull bull bull bull J bull

bull bull f ~ I

I ~ Ibull bull

bullbull bull

l bull I -bullbull bull bull

I

I bull bull

bull I I~~bull bull bull

bull ~

I bull bull

~ ~_ ~~~

- - New Prior r bull

Posterior 114 new failure cases ~ 300- bull bull bull bull bull bull bull Posterior f 13 new damage casesl 150 ~

- bull bull - bull bull Posterior 116 new success cases i 2 00

Postenor 1113 new success failure -150 damage cases

I 00

0 50 -

0 00 0 00 0 20 0 40 0 60 0 80 I 00 120 I 40 160 110 2 00

Posterior ol B Mean= 120 COV= 0071 b

shy

shy

_rshy

Posterior Distribution of Bias Factor (B) bull Effect of Additional 13-Similar Observations

1

1 I

WEIGHTING APPROACH bull

bull ASSESSMENT OF PLATFORMS DATA

Select Platforms Most Mappable to the Analyzed Platforms

Unlikely SurvivalDamage Cases Identification

Based on Cut-Off Criteria Agemiddot Hs etc

bull INFORMATION REQUIRED FOR COMPARISON

bull CATEGORIZATION OF ADDITIONALmiddot PLATFORMS

bull UPDATING OF CALIBRATION RESULTS

--

OVERVIEW OF PLATFORMS SURVIVEDDAMAGED -~

Operator -

- Platforms

Survived CS)

Platforms

Damaged CD)

Platforms Used In

Phase IJIP -

JI~ ~Bill~IRBDll -

Amoco -

Chevron -

Exxon

Mob II

Phllllps

Shell

Unocal -

30

143

72 22

7

39 25shy -

1

12

bull

1 bull

1

1(S)+1 (D)

3 (S) +3(D)

2 (S)

bull shy

bull

bull

bull

Total Partlclnants Platforms 338 - 15 10C6S+4D) -

Other Operators

310

13

3 (D)

Total Platforms 648 28 13C6S+7D) -

Summary of Andrew JIP bull Phase I Platforms

Plairshy w v- N-ber Lea-le 1 shy car Hind_ o-w N-bull Deplh fl fl IJpe la bull n Yeadon Jadtet lbelledtf liefalit -- _ -OD

IL H - -shy _ICllshy IL IL - - rQClto ~ _

K STUIK 137 1963 8 365 608S4 Cellardeck

-middot STl-shy 170 1964 4 -~ K 39 62269 Cellardeck - --ST134W 137 1911 4 43 608S4 Smvlnlshy- _ - -

WD90A 184 1964 8 K 267Sbull34 83 SOS31 SUboftllar deck SmvMI

MC311 343 1978 8 x S7 62 - MC397 468 1991 4 x shy S33bull6SO 65911 -shySmeSmvlvol -- gt -

r-- T23ST52 63 1969 4 ton- shyKshy 52594 S03 -T25SSl39 62 1969 4 Kshy SOS94 S063 - middot ton-shy__ -

bullbull a ton--shyST161A 118 1964 8 K - 347S 51495 Cellarclecti

r~

ST1771 142 1965 8 K 39 60167 CeDordeck failme

STISIH 137 1964 8 K 395 608S4 CeDardeck Failme

K ST130A 140 1958 a SS 6095 I lblv Pai1uno - T21ST72 61 1969 4 Kshy S4S94 4974 I lhlv Pai1uno -

- -

JE

INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON

bull WATER DEPTH

bull bull YEAR OF INSTALLATIONDESIGN

bull PLATFORM ORIENTATION

bull PLATFORM LATITUDE AND LONGITUDE

bull KEY SKETCHES OF PLATFORM STRUCTURAL FRAMING

Number of Legs Plies Leg Batter Number of Conductors bull Deck Elevations (Including sub-cellar deck If any) Jacket Framing Elevations Brace type In Vertical Frames Sizes of Primary Members (leg pile vertical frames braces primary horizontals) Joint Details (can gap etc) LegPiie Annulus Grouting Status

bull middotPOST ANDREW INSPECTION REPORT Significant Wave Height Hs (ft)

bull INFORMATION ON DAMAGES PRIOR TO ANDREW

bull RESULTS FROM IN-HOUSE STUDIES IF AVAILABLE

bull ~RIGINAL DESIGN CRITERIA INFORMATION IF AVAILABLE

bull SOIL SHEAR STRENGTH PROFILE IF AVAILABLE

Platforms Identified for Evaluating Therr Likelihood of Falling Under Unexpected Survival Category

OPlRATOR

middotADA

BLOClt

llmucnJllE

NAME

YEAll

INSTAUJD

WA1ER

DBPTll

DISTANCE

JlllOM SHOllE

(MILU)

Bl

(M)

AMOCO 111113

STl11

WDI075

A

bull D

St

17

120

1n

44

20

10 I

CllllVllQfl STlllO c 61 110 21 11 STIUO D Q 1110 21 11

STl134 p Q 137 29 11

ST0151 G 61 137 11

ST0151 I 3 121 11

ST0151 J Q 140 11

STl151 It 137 11

ST0151 PJl001 Q UT 11

ST0151 PROD-2 Q UT 11

ST017 A 3 140 35 11

WDIUT c 65 214 34 10

llXXON STI055 II 51 6T 14 I

STl15 A t3 34

WDI073

WD0073

Amiddotshy Q

1111 1111

22

22

I

I

WDI073 c 1n 22 I

- WDOCl93-shy II

65 1110 2T A 3 194 31 MOBU 111121

SS IOl7

A B

501

50

50

31

T

I

T 6

MtlllPllY SS114 B T 50 T T

PlllllIPS SS0149 A 5 55 33 T

SAMllDAN STln A 107 T T

SBllLL WDl103 A 65 223 2T WDl103 B 65 221 2T WD0104 c 65 221 2T WD0104 D 6T 20 2T

UNOCAL SS020I

SS 020I SSl20I

SS 0209

SS0209

II p

PURll

A

B

3

n 61

103

97

97

95

100

34

34

34

34 34

bullbull bullbullbull

- SS0209 D u 95 34 I

- SS 1209 SSl253

SS0253

G A

c

6T

2

100

165

1T5

34

54

54

bullbull I

I I

I I

SOUTH MARCH

~LAND

N

SOUTH PaTOI ENE ISWltD AREA

MMS Defined Limits ror Offshore Structures Requiring Inspections Following Andrew

t---Jbulln3 T25bull I bullTtl

bull1$1Htl1Abull bull1SIK1779bull

~

middot-DaTA AREA~T

CRANO rsLE

AREA bull1300

bullll4W

=shy

bullPath of Hurricane Andrew with Platforms Used In Calibration

Likelihood Function vs b All Success Case STISI K STIJOQ ST134W WD90A MCJll MCJ97

I 0000

0 9000

0 8000

07000

-Bo6000

tiiii 0 5000 -

04000

0 3000

0 2000

01000

0 0000

0 05 25

ri1rt middot-middot-middot-middot-middot-

I 5 2

b

---11 middot151K(blsu~ccsI

- bull - bull - fk 130Q(bfU~lCSS)

lie bull134W(blsuccc]

lk-WD90A[blsultCS]

- - - lk middotMC397(bfUtlCS]

--o-- lk-MCJll(bbuctcsJ

--- ll (bl 6 out of 6 Mlttcs case]

_

( Likelihood Functions middotSuccess Cases - p - 1 c I

CATEGORIZATION OF ADDITIONAL PLATFORMS

I I I

bull CATEGORIES

Unexpected Survivals

Expected Survivals

bull APPROACH

Identify Corresponding Platform(s) Evaluated In the JIP

Determine Max Wave Height at Platform Location

Evaluate Whether Wave Hit the Deck

middot - Evaluate Likelihood of R and S to Differ from the Base Cases

Review Joint Information

Assess Special Features

bull IDENTIFY NUMBER OF PLATFORMS IN EACH CATEGORY

I

UPDATING OF CALIBRATION RESULTS

bull INCLUDE EFFECT OF ADDITIONAL PLATFORMS ON bullBbull

Consider Increased 1Uncertalntles In R and S

Develop bullNeW Ukellhood Functions for Different Categories

middotLet bullxbull Number of Ukely Unexpected Survival Cases and y Number of Ukely Expected Survival Cases

Determine Combined Likelihood Function lk (b Iadditional platforms) middot

bull REVISED LIKELIHOOD FUNCTION

lk(b I all cases) = lk(b I 9 platforms) middot x lk(b Ix platforms) x lk(b I y platforms)

bull ESTABLISH REVISED BIAS FACTOR B

f8 (b) u f8 (b) lk (b I all cases)

II

t t ~

shy

APPLICATIONS OF BIAS FACTORS

ii APPLICABLE TO PARTICULAR LOAD AND RESISTANCE RECIPE

bull USE BY THE API COMMITTEES AND MMS

Criteria Revisions

Recommended Design Guidelines Revisions bull

bullbull REASSESSMENT STUDIES AND DECISION MAKING

middot

bull RISK ASSESSMENT STUDIES BY THE OPERATORS

Individual Platforms

Fleet of Platforms

middot IMR Program Implementation

bullbull NOT TO BE USED IN DESIGN

I I fl J

EXAMPLE USE OF BIAS FACTORS

bull PHASE I EXAMPLE APPLICATION

bull EVALUATION OF ANNUAL PROBABILITY OF FAILURE

Establish Annual Maximum Seastate Data

Determine Coefficients to Define Base Shear

Determine Ultimate Capacity of Platform

Determine Probablllty of Failure Versus bullbbull

-bull DETERMINE UPDATED PROBABILITY OF FAILURE

Pf - P [S gt R]

Pf I P [~ gt Rb I b] f 8 (b) db

Pr - J P [S gt Rb I b] f bull 8 (b) db

bull TOTAL PROBABILITIES OF FAILURE

Total Pf - 1 bull Il [1 middotPf] - l Pf bull

Total Pf - 1 bull Il [1middot Pf] - l Pfmiddot

Total Pr = 1 - n [1middot PfbullJ = l Pfshy

I I I

EXAMPLE APPLICATION bull PHASE I

PLATFORM ST 151 K

bull GULF OF MEXICO

bull WATER DEPTH = 137 FT

bull ANNUAL MAXIMUM HS MEDIAN 175 ft COV 031

bull ASSUMED STORM DURATION 3 hoursmiddot

bull ASSUMED NUMBER OF WAVES 800

bull MEDIAN HS FOR STORM HOURS 1 2 AND 3

bull MEDIAN CURRENT VALUE bull

bull DETERMINE ULTIMATE CAPACITY FOR THREE DIRECTIONS

bull DETERMINE ANNUAL FAILURE PROBABILITIES

Broadsidemiddot Diagonal End-on

0037 0025 0007

0071 0044 0018

0022 0016 0004

DETERMINE TOTAL ANNUAL PROBABILITY OF FAILURE

Pf - 007

Pf - 013

Pr - 004

bull

EXAMPLE USE OF BIAS FACTORS - PHASE II middot

bull DETERMINE PREDICTIVE PROBABILITY OF FAILURE

Case - 1

Pf _ S S Pf (b b2) fa1a2 (b b2) db db2

g bull -1 (b1 R I S) - 1

92 - (b2 R2 I S) - 1

Case -2

Pf s ~ [bJ f11 (b) db

Pf [bJ p [g lt 0)

g - bull 1 (b1 R1 I S) - 1

1

bull1) l

I

BENEFITS OF ANDREW JIP bull PHASE II

lt

bull IMPROVED LOAD AND RESISTANCE RECIPE

bull IMPROVED CALIBRATION PROCEDURE

bull MULTIPLE BIAS FACTORS

bull PROCEDURE TO INCLUDE EFFECT OF OTHER PLATFORMS

bull PROCEDURE TO USE MULTIPLE BIAS FACTORS

  • Structure Bookmarks
    • ANDREW PHASE II JOINT1 INDUSTRY PROJECT
    • PROGRESS MEETING NUMBER 1
    • ANDREW PHASE II JOINT INDUSTRY PROJECT PROGRESS MEETING NUMBER 1 NOVEMBER 16 AGENDA
    • ANDREW PHASE II UST OF PARTICIPANTS
    • ANDREW PHASE I OBJECTIVES AND RESULTS
    • ANDREW PHASE II PROJECT OBJECTIVES
    • SUBSEQUENT PROJECTS
    • ANDREW PHASE II PROJECT TEAM
    • ANDREW PHASE II UPDATE MODELING middotAND ANALYSIS RECIPE
    • -ANDREW PHASE II CONDUCTOR RESISTANCE
    • ANDREW PHASE Il WA VE FORCE PROFILE
    • ANDREW PHASE II WA VE FORCE PROFILEmiddot
    • ANDREW PHASE II BRACE MODELING
    • ANDREW PHASE II FOUNDATIONS
    • ANDREW PHASE II middot middot JOINT MODELING
    • ANDREW PHASE Il JOINT MODELING OPTIONS DISCUSSION
    • ANDREW PHASE II JOINT MODEJJNG OPTIONS ILLUSTRATIONS
    • ANDREW PHASE II JOINT MODELING PHASE II METHOD
    • ANDREW PHASE II ANDREW STORM CONDmONS UPDATED
    • Platform Selection bullbullbull1111111111111
    • BasisCriteria for Selection bullbullbull1111111111111
    • CALIBRATION
    • WHERE TO INTRODUCE BIAS FACTOR (B
    • PHASE I -KEY CONCLUSIONS
    • bull CALIBRATION ISSUES PHASE II
    • PHASE II WORK SCOPE (CALIBRATION)
    • SOFTWARE
    • SENSITIVITY STUDIES POSTERIOR DISTRIBUTION OF B
    • WEIGHTING APPROACH bull
    • INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON
    • CATEGORIZATION OF ADDITIONAL PLATFORMS
    • UPDATING OF CALIBRATION RESULTS
    • APPLICATIONS OF BIAS FACTORS
    • EXAMPLE USE OF BIAS FACTORS
    • EXAMPLE APPLICATION bull PHASE I PLATFORM ST 151 K
    • EXAMPLE USE OF BIAS FACTORS -PHASE II middot
    • BENEFITS OF ANDREW JIP bull PHASE II
Page 59:  · . ' ANDREW PHASE II PROJECT OBJECTIVES . ' • ·UPDATE MODELING AND ANALYSIS RECIPE· JOINTS ' ' ·BRACES WAVE PROFILE FOUNDATIONS . ' ' CONDUCTORS· • ANALVZE PLATFORMS USING

bullbull

PHASE I PHASE II

CAPACITYbull ANALYSISRu bull

---------------------1----middot

Jc1 - iJbullraquoJ dbull

FttlS frt[f~l ratdhJJ

-

LIKELIHOOD FUNCTION

DEVELOPMENT

lk Jb Jfall1N) bull Jb)

111 Ill Isurvtvoq bull 1 - bl

-middot -------------------shy ~---------------------middot-middot

r111ir8 Cb1 111 Cbl BAYESIAN

UPDATING

~cove

R1R2 bull middot---------------------~----

Jblbull P(g cOJ

1 bull JbRI (MBSI I bull 1

~middot

llcJllJfalluro) bull P(gcOJ Ill 111 b21fallun) bull P ((g1 CO) U 1112 c OD

middot--------------------shy ----------------------shy bull

r111111r8 Cbl 111 Ill i-1nro1

deg ~1b2)laquottfrr (b1b21-lnfo) re Ill bull 1-11middot b2) bull lodegamp

middot

i1 covamp i2 cov~i

-

shy

--

Likelihood Function vs b All Success Case STISI K STl30Q STl34W WD90A MC311 MC397

I 0000

0 9000

OIOOO

0 7000

~ 0amp000

ta osoOO

~ 0 4000

0 3000

0 2000

0 1000

bull 0 0000

bull

I

middot

middot bull I

Ibull

-sibull ~ -middot-middot-middot-middot -middot-

----7----+-----~ I

Ibull bull bull I

rbull

--- Jk ISIK[blsuccess)

- bull - bull - lk 130Q[blsuccess) bull

lk 134W[blsuccess)

Jk-WD90A(blsuccc~I

- - - lk MC397[blsucccss)

--ltgt-- lk middotMC31 l[blsucccss)

--- Jk [bl 6 out of6succcsHascs)

05 I 5 2 250

b

Likelihood Functionsmiddot Success Cases - Phbullsc L

shy

bullbull bull bull bull bull bull -------------------~

bull bull bullbull bullbull bullbull bullbull bullbull bullbull bull

bull bullbull

bullbullbullbull

bullbull

bullbullbullbullbullbullbull

bullbullbullbullbullbull

bullbullbullbullbullbullbull

z 0

t z ) LL

c 0 0c i ~ i

J

~ sect ~--------------~ e l5

I

I

SOFTWARE

bull PF PROGRAM

bull CALREL

bull PF +CALREL

bull RELACS

bull

gt i

l

SENSITIVITY STUDIES

POSTERIOR DISTRIBUTION OF B

bull EFFECT OF CHANGE IN DISTRIBUTION OF PRIOR OF B

bull EFFECT OF REMOVAL OF ONE PLATFORM FROM EACH CATEGORY

bull EFFECT OF REMOVAL OF THREE DAMAGE CASES

bull EFFECT OF ADDITIONAL 13 SIMILAR PLATFORMS

-I

Prior and Posterior Distributions of B Variation In COV of PriorofB fromOl to04 -- -

4500

-4000

-

- -

3500 --middot - bull bull - bull bull Prior-r ICOV =02

3000

- - - - Prior-r ICOV =03 -

bull -~ e 2soo -

bull bull bull bull bull bull bull Pnor-r ICOV =o 4t frJ ~ 2000 Posterior rlCOV =0 2

bull bull Postenor rlCOV =0 3 bull bull 1500 bull middot bull - shy -middot- Posterior rlCOV =04

middot~ I 000 - bull

-middotmiddot ~ ~middotmiddot M -~

- middot 0 500

wbullbullbull _ I middot~ middotmiddotmiddotmiddotmiddot middotmiddotmiddotmiddotmiddot - -~ ~ __ middot_ _ bull -shy0 000

0 000 0200 0 400 0 600 0 800 IOOO 1200 1400 1600 1800 2000 -

- - b

-

=shy

Posterior Distribution of Bias Factor (B) bull Effect of Variation In COV of Prior of B

bull

bullPrior and Posterior Distributions of B Evaluating Effect of Removal of One Platform from Each Category from the Base Case of 13 Platforms

400

~ ~ 350

j IA i

3 00 bull I bull I ---Prior-r I bull l

250 1 I 1

I bull bull ---Posterior r lall 13 casesshy- ~E 2 00 bull bull bull bull bull bull bull Posterior r without STISIK

- ~ 1 bull - bull I - bull bull - bull bull Posterior r Iwithout T2I I 50 I

J-tt~i - bull - bull - Posterior r Iwithout T2S

tOO [-

I I ~ ~ I 0 50

0 00

___ r J i

l~T

~--

0 00 0 20 0 40 0 60 080 I 00 120 140 160 110 2 00

b

zshy

Posterior Distribution of Bias Factor (8) bull Effect of Removal of middotPlatform from Any Category

bullPrior and Posterior Distributions or B Evaluating Effect or Removal or 3 Damage Cases from the Base Case or 13 Platrorms

4

35

3

25

e L 2 ~ bulle shy

I 15

bull I

05

0

0 02 04 06 I 2 I 4 16 IS 2

I I

I

I

I Ibull

J -tmiddot L--

bullI

~

I

fl

II ~

I I v I

middot bull bull

bull L

~ ll- I-~lt

OS I

b

I

bull

1middot bullbullbullbull

~ __

bull

- - - - Prior-r

- - - - Postenor flall failure cases

Postenor flall success cases

--- Posterior flall 13 cases

---Posterior fI success failure cases only

IPosterior or B Mean =t11 COV =0121

-

_I

Posterir Distribution or Bias Factor (B) bull Effect or Removal or 3-Damage Cases

-

bullPrior and Posterior Distributions or B Evaluating Effect or Additional 13 Similar Observations as

Used in Andrew JIP The POsterior or B from Andrew-JIP Considered as New Prior or B

- 500

450

400

350

bull bull

bull bull bull bull J bull

bull bull f ~ I

I ~ Ibull bull

bullbull bull

l bull I -bullbull bull bull

I

I bull bull

bull I I~~bull bull bull

bull ~

I bull bull

~ ~_ ~~~

- - New Prior r bull

Posterior 114 new failure cases ~ 300- bull bull bull bull bull bull bull Posterior f 13 new damage casesl 150 ~

- bull bull - bull bull Posterior 116 new success cases i 2 00

Postenor 1113 new success failure -150 damage cases

I 00

0 50 -

0 00 0 00 0 20 0 40 0 60 0 80 I 00 120 I 40 160 110 2 00

Posterior ol B Mean= 120 COV= 0071 b

shy

shy

_rshy

Posterior Distribution of Bias Factor (B) bull Effect of Additional 13-Similar Observations

1

1 I

WEIGHTING APPROACH bull

bull ASSESSMENT OF PLATFORMS DATA

Select Platforms Most Mappable to the Analyzed Platforms

Unlikely SurvivalDamage Cases Identification

Based on Cut-Off Criteria Agemiddot Hs etc

bull INFORMATION REQUIRED FOR COMPARISON

bull CATEGORIZATION OF ADDITIONALmiddot PLATFORMS

bull UPDATING OF CALIBRATION RESULTS

--

OVERVIEW OF PLATFORMS SURVIVEDDAMAGED -~

Operator -

- Platforms

Survived CS)

Platforms

Damaged CD)

Platforms Used In

Phase IJIP -

JI~ ~Bill~IRBDll -

Amoco -

Chevron -

Exxon

Mob II

Phllllps

Shell

Unocal -

30

143

72 22

7

39 25shy -

1

12

bull

1 bull

1

1(S)+1 (D)

3 (S) +3(D)

2 (S)

bull shy

bull

bull

bull

Total Partlclnants Platforms 338 - 15 10C6S+4D) -

Other Operators

310

13

3 (D)

Total Platforms 648 28 13C6S+7D) -

Summary of Andrew JIP bull Phase I Platforms

Plairshy w v- N-ber Lea-le 1 shy car Hind_ o-w N-bull Deplh fl fl IJpe la bull n Yeadon Jadtet lbelledtf liefalit -- _ -OD

IL H - -shy _ICllshy IL IL - - rQClto ~ _

K STUIK 137 1963 8 365 608S4 Cellardeck

-middot STl-shy 170 1964 4 -~ K 39 62269 Cellardeck - --ST134W 137 1911 4 43 608S4 Smvlnlshy- _ - -

WD90A 184 1964 8 K 267Sbull34 83 SOS31 SUboftllar deck SmvMI

MC311 343 1978 8 x S7 62 - MC397 468 1991 4 x shy S33bull6SO 65911 -shySmeSmvlvol -- gt -

r-- T23ST52 63 1969 4 ton- shyKshy 52594 S03 -T25SSl39 62 1969 4 Kshy SOS94 S063 - middot ton-shy__ -

bullbull a ton--shyST161A 118 1964 8 K - 347S 51495 Cellarclecti

r~

ST1771 142 1965 8 K 39 60167 CeDordeck failme

STISIH 137 1964 8 K 395 608S4 CeDardeck Failme

K ST130A 140 1958 a SS 6095 I lblv Pai1uno - T21ST72 61 1969 4 Kshy S4S94 4974 I lhlv Pai1uno -

- -

JE

INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON

bull WATER DEPTH

bull bull YEAR OF INSTALLATIONDESIGN

bull PLATFORM ORIENTATION

bull PLATFORM LATITUDE AND LONGITUDE

bull KEY SKETCHES OF PLATFORM STRUCTURAL FRAMING

Number of Legs Plies Leg Batter Number of Conductors bull Deck Elevations (Including sub-cellar deck If any) Jacket Framing Elevations Brace type In Vertical Frames Sizes of Primary Members (leg pile vertical frames braces primary horizontals) Joint Details (can gap etc) LegPiie Annulus Grouting Status

bull middotPOST ANDREW INSPECTION REPORT Significant Wave Height Hs (ft)

bull INFORMATION ON DAMAGES PRIOR TO ANDREW

bull RESULTS FROM IN-HOUSE STUDIES IF AVAILABLE

bull ~RIGINAL DESIGN CRITERIA INFORMATION IF AVAILABLE

bull SOIL SHEAR STRENGTH PROFILE IF AVAILABLE

Platforms Identified for Evaluating Therr Likelihood of Falling Under Unexpected Survival Category

OPlRATOR

middotADA

BLOClt

llmucnJllE

NAME

YEAll

INSTAUJD

WA1ER

DBPTll

DISTANCE

JlllOM SHOllE

(MILU)

Bl

(M)

AMOCO 111113

STl11

WDI075

A

bull D

St

17

120

1n

44

20

10 I

CllllVllQfl STlllO c 61 110 21 11 STIUO D Q 1110 21 11

STl134 p Q 137 29 11

ST0151 G 61 137 11

ST0151 I 3 121 11

ST0151 J Q 140 11

STl151 It 137 11

ST0151 PJl001 Q UT 11

ST0151 PROD-2 Q UT 11

ST017 A 3 140 35 11

WDIUT c 65 214 34 10

llXXON STI055 II 51 6T 14 I

STl15 A t3 34

WDI073

WD0073

Amiddotshy Q

1111 1111

22

22

I

I

WDI073 c 1n 22 I

- WDOCl93-shy II

65 1110 2T A 3 194 31 MOBU 111121

SS IOl7

A B

501

50

50

31

T

I

T 6

MtlllPllY SS114 B T 50 T T

PlllllIPS SS0149 A 5 55 33 T

SAMllDAN STln A 107 T T

SBllLL WDl103 A 65 223 2T WDl103 B 65 221 2T WD0104 c 65 221 2T WD0104 D 6T 20 2T

UNOCAL SS020I

SS 020I SSl20I

SS 0209

SS0209

II p

PURll

A

B

3

n 61

103

97

97

95

100

34

34

34

34 34

bullbull bullbullbull

- SS0209 D u 95 34 I

- SS 1209 SSl253

SS0253

G A

c

6T

2

100

165

1T5

34

54

54

bullbull I

I I

I I

SOUTH MARCH

~LAND

N

SOUTH PaTOI ENE ISWltD AREA

MMS Defined Limits ror Offshore Structures Requiring Inspections Following Andrew

t---Jbulln3 T25bull I bullTtl

bull1$1Htl1Abull bull1SIK1779bull

~

middot-DaTA AREA~T

CRANO rsLE

AREA bull1300

bullll4W

=shy

bullPath of Hurricane Andrew with Platforms Used In Calibration

Likelihood Function vs b All Success Case STISI K STIJOQ ST134W WD90A MCJll MCJ97

I 0000

0 9000

0 8000

07000

-Bo6000

tiiii 0 5000 -

04000

0 3000

0 2000

01000

0 0000

0 05 25

ri1rt middot-middot-middot-middot-middot-

I 5 2

b

---11 middot151K(blsu~ccsI

- bull - bull - fk 130Q(bfU~lCSS)

lie bull134W(blsuccc]

lk-WD90A[blsultCS]

- - - lk middotMC397(bfUtlCS]

--o-- lk-MCJll(bbuctcsJ

--- ll (bl 6 out of 6 Mlttcs case]

_

( Likelihood Functions middotSuccess Cases - p - 1 c I

CATEGORIZATION OF ADDITIONAL PLATFORMS

I I I

bull CATEGORIES

Unexpected Survivals

Expected Survivals

bull APPROACH

Identify Corresponding Platform(s) Evaluated In the JIP

Determine Max Wave Height at Platform Location

Evaluate Whether Wave Hit the Deck

middot - Evaluate Likelihood of R and S to Differ from the Base Cases

Review Joint Information

Assess Special Features

bull IDENTIFY NUMBER OF PLATFORMS IN EACH CATEGORY

I

UPDATING OF CALIBRATION RESULTS

bull INCLUDE EFFECT OF ADDITIONAL PLATFORMS ON bullBbull

Consider Increased 1Uncertalntles In R and S

Develop bullNeW Ukellhood Functions for Different Categories

middotLet bullxbull Number of Ukely Unexpected Survival Cases and y Number of Ukely Expected Survival Cases

Determine Combined Likelihood Function lk (b Iadditional platforms) middot

bull REVISED LIKELIHOOD FUNCTION

lk(b I all cases) = lk(b I 9 platforms) middot x lk(b Ix platforms) x lk(b I y platforms)

bull ESTABLISH REVISED BIAS FACTOR B

f8 (b) u f8 (b) lk (b I all cases)

II

t t ~

shy

APPLICATIONS OF BIAS FACTORS

ii APPLICABLE TO PARTICULAR LOAD AND RESISTANCE RECIPE

bull USE BY THE API COMMITTEES AND MMS

Criteria Revisions

Recommended Design Guidelines Revisions bull

bullbull REASSESSMENT STUDIES AND DECISION MAKING

middot

bull RISK ASSESSMENT STUDIES BY THE OPERATORS

Individual Platforms

Fleet of Platforms

middot IMR Program Implementation

bullbull NOT TO BE USED IN DESIGN

I I fl J

EXAMPLE USE OF BIAS FACTORS

bull PHASE I EXAMPLE APPLICATION

bull EVALUATION OF ANNUAL PROBABILITY OF FAILURE

Establish Annual Maximum Seastate Data

Determine Coefficients to Define Base Shear

Determine Ultimate Capacity of Platform

Determine Probablllty of Failure Versus bullbbull

-bull DETERMINE UPDATED PROBABILITY OF FAILURE

Pf - P [S gt R]

Pf I P [~ gt Rb I b] f 8 (b) db

Pr - J P [S gt Rb I b] f bull 8 (b) db

bull TOTAL PROBABILITIES OF FAILURE

Total Pf - 1 bull Il [1 middotPf] - l Pf bull

Total Pf - 1 bull Il [1middot Pf] - l Pfmiddot

Total Pr = 1 - n [1middot PfbullJ = l Pfshy

I I I

EXAMPLE APPLICATION bull PHASE I

PLATFORM ST 151 K

bull GULF OF MEXICO

bull WATER DEPTH = 137 FT

bull ANNUAL MAXIMUM HS MEDIAN 175 ft COV 031

bull ASSUMED STORM DURATION 3 hoursmiddot

bull ASSUMED NUMBER OF WAVES 800

bull MEDIAN HS FOR STORM HOURS 1 2 AND 3

bull MEDIAN CURRENT VALUE bull

bull DETERMINE ULTIMATE CAPACITY FOR THREE DIRECTIONS

bull DETERMINE ANNUAL FAILURE PROBABILITIES

Broadsidemiddot Diagonal End-on

0037 0025 0007

0071 0044 0018

0022 0016 0004

DETERMINE TOTAL ANNUAL PROBABILITY OF FAILURE

Pf - 007

Pf - 013

Pr - 004

bull

EXAMPLE USE OF BIAS FACTORS - PHASE II middot

bull DETERMINE PREDICTIVE PROBABILITY OF FAILURE

Case - 1

Pf _ S S Pf (b b2) fa1a2 (b b2) db db2

g bull -1 (b1 R I S) - 1

92 - (b2 R2 I S) - 1

Case -2

Pf s ~ [bJ f11 (b) db

Pf [bJ p [g lt 0)

g - bull 1 (b1 R1 I S) - 1

1

bull1) l

I

BENEFITS OF ANDREW JIP bull PHASE II

lt

bull IMPROVED LOAD AND RESISTANCE RECIPE

bull IMPROVED CALIBRATION PROCEDURE

bull MULTIPLE BIAS FACTORS

bull PROCEDURE TO INCLUDE EFFECT OF OTHER PLATFORMS

bull PROCEDURE TO USE MULTIPLE BIAS FACTORS

  • Structure Bookmarks
    • ANDREW PHASE II JOINT1 INDUSTRY PROJECT
    • PROGRESS MEETING NUMBER 1
    • ANDREW PHASE II JOINT INDUSTRY PROJECT PROGRESS MEETING NUMBER 1 NOVEMBER 16 AGENDA
    • ANDREW PHASE II UST OF PARTICIPANTS
    • ANDREW PHASE I OBJECTIVES AND RESULTS
    • ANDREW PHASE II PROJECT OBJECTIVES
    • SUBSEQUENT PROJECTS
    • ANDREW PHASE II PROJECT TEAM
    • ANDREW PHASE II UPDATE MODELING middotAND ANALYSIS RECIPE
    • -ANDREW PHASE II CONDUCTOR RESISTANCE
    • ANDREW PHASE Il WA VE FORCE PROFILE
    • ANDREW PHASE II WA VE FORCE PROFILEmiddot
    • ANDREW PHASE II BRACE MODELING
    • ANDREW PHASE II FOUNDATIONS
    • ANDREW PHASE II middot middot JOINT MODELING
    • ANDREW PHASE Il JOINT MODELING OPTIONS DISCUSSION
    • ANDREW PHASE II JOINT MODEJJNG OPTIONS ILLUSTRATIONS
    • ANDREW PHASE II JOINT MODELING PHASE II METHOD
    • ANDREW PHASE II ANDREW STORM CONDmONS UPDATED
    • Platform Selection bullbullbull1111111111111
    • BasisCriteria for Selection bullbullbull1111111111111
    • CALIBRATION
    • WHERE TO INTRODUCE BIAS FACTOR (B
    • PHASE I -KEY CONCLUSIONS
    • bull CALIBRATION ISSUES PHASE II
    • PHASE II WORK SCOPE (CALIBRATION)
    • SOFTWARE
    • SENSITIVITY STUDIES POSTERIOR DISTRIBUTION OF B
    • WEIGHTING APPROACH bull
    • INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON
    • CATEGORIZATION OF ADDITIONAL PLATFORMS
    • UPDATING OF CALIBRATION RESULTS
    • APPLICATIONS OF BIAS FACTORS
    • EXAMPLE USE OF BIAS FACTORS
    • EXAMPLE APPLICATION bull PHASE I PLATFORM ST 151 K
    • EXAMPLE USE OF BIAS FACTORS -PHASE II middot
    • BENEFITS OF ANDREW JIP bull PHASE II
Page 60:  · . ' ANDREW PHASE II PROJECT OBJECTIVES . ' • ·UPDATE MODELING AND ANALYSIS RECIPE· JOINTS ' ' ·BRACES WAVE PROFILE FOUNDATIONS . ' ' CONDUCTORS· • ANALVZE PLATFORMS USING

--

Likelihood Function vs b All Success Case STISI K STl30Q STl34W WD90A MC311 MC397

I 0000

0 9000

OIOOO

0 7000

~ 0amp000

ta osoOO

~ 0 4000

0 3000

0 2000

0 1000

bull 0 0000

bull

I

middot

middot bull I

Ibull

-sibull ~ -middot-middot-middot-middot -middot-

----7----+-----~ I

Ibull bull bull I

rbull

--- Jk ISIK[blsuccess)

- bull - bull - lk 130Q[blsuccess) bull

lk 134W[blsuccess)

Jk-WD90A(blsuccc~I

- - - lk MC397[blsucccss)

--ltgt-- lk middotMC31 l[blsucccss)

--- Jk [bl 6 out of6succcsHascs)

05 I 5 2 250

b

Likelihood Functionsmiddot Success Cases - Phbullsc L

shy

bullbull bull bull bull bull bull -------------------~

bull bull bullbull bullbull bullbull bullbull bullbull bullbull bull

bull bullbull

bullbullbullbull

bullbull

bullbullbullbullbullbullbull

bullbullbullbullbullbull

bullbullbullbullbullbullbull

z 0

t z ) LL

c 0 0c i ~ i

J

~ sect ~--------------~ e l5

I

I

SOFTWARE

bull PF PROGRAM

bull CALREL

bull PF +CALREL

bull RELACS

bull

gt i

l

SENSITIVITY STUDIES

POSTERIOR DISTRIBUTION OF B

bull EFFECT OF CHANGE IN DISTRIBUTION OF PRIOR OF B

bull EFFECT OF REMOVAL OF ONE PLATFORM FROM EACH CATEGORY

bull EFFECT OF REMOVAL OF THREE DAMAGE CASES

bull EFFECT OF ADDITIONAL 13 SIMILAR PLATFORMS

-I

Prior and Posterior Distributions of B Variation In COV of PriorofB fromOl to04 -- -

4500

-4000

-

- -

3500 --middot - bull bull - bull bull Prior-r ICOV =02

3000

- - - - Prior-r ICOV =03 -

bull -~ e 2soo -

bull bull bull bull bull bull bull Pnor-r ICOV =o 4t frJ ~ 2000 Posterior rlCOV =0 2

bull bull Postenor rlCOV =0 3 bull bull 1500 bull middot bull - shy -middot- Posterior rlCOV =04

middot~ I 000 - bull

-middotmiddot ~ ~middotmiddot M -~

- middot 0 500

wbullbullbull _ I middot~ middotmiddotmiddotmiddotmiddot middotmiddotmiddotmiddotmiddot - -~ ~ __ middot_ _ bull -shy0 000

0 000 0200 0 400 0 600 0 800 IOOO 1200 1400 1600 1800 2000 -

- - b

-

=shy

Posterior Distribution of Bias Factor (B) bull Effect of Variation In COV of Prior of B

bull

bullPrior and Posterior Distributions of B Evaluating Effect of Removal of One Platform from Each Category from the Base Case of 13 Platforms

400

~ ~ 350

j IA i

3 00 bull I bull I ---Prior-r I bull l

250 1 I 1

I bull bull ---Posterior r lall 13 casesshy- ~E 2 00 bull bull bull bull bull bull bull Posterior r without STISIK

- ~ 1 bull - bull I - bull bull - bull bull Posterior r Iwithout T2I I 50 I

J-tt~i - bull - bull - Posterior r Iwithout T2S

tOO [-

I I ~ ~ I 0 50

0 00

___ r J i

l~T

~--

0 00 0 20 0 40 0 60 080 I 00 120 140 160 110 2 00

b

zshy

Posterior Distribution of Bias Factor (8) bull Effect of Removal of middotPlatform from Any Category

bullPrior and Posterior Distributions or B Evaluating Effect or Removal or 3 Damage Cases from the Base Case or 13 Platrorms

4

35

3

25

e L 2 ~ bulle shy

I 15

bull I

05

0

0 02 04 06 I 2 I 4 16 IS 2

I I

I

I

I Ibull

J -tmiddot L--

bullI

~

I

fl

II ~

I I v I

middot bull bull

bull L

~ ll- I-~lt

OS I

b

I

bull

1middot bullbullbullbull

~ __

bull

- - - - Prior-r

- - - - Postenor flall failure cases

Postenor flall success cases

--- Posterior flall 13 cases

---Posterior fI success failure cases only

IPosterior or B Mean =t11 COV =0121

-

_I

Posterir Distribution or Bias Factor (B) bull Effect or Removal or 3-Damage Cases

-

bullPrior and Posterior Distributions or B Evaluating Effect or Additional 13 Similar Observations as

Used in Andrew JIP The POsterior or B from Andrew-JIP Considered as New Prior or B

- 500

450

400

350

bull bull

bull bull bull bull J bull

bull bull f ~ I

I ~ Ibull bull

bullbull bull

l bull I -bullbull bull bull

I

I bull bull

bull I I~~bull bull bull

bull ~

I bull bull

~ ~_ ~~~

- - New Prior r bull

Posterior 114 new failure cases ~ 300- bull bull bull bull bull bull bull Posterior f 13 new damage casesl 150 ~

- bull bull - bull bull Posterior 116 new success cases i 2 00

Postenor 1113 new success failure -150 damage cases

I 00

0 50 -

0 00 0 00 0 20 0 40 0 60 0 80 I 00 120 I 40 160 110 2 00

Posterior ol B Mean= 120 COV= 0071 b

shy

shy

_rshy

Posterior Distribution of Bias Factor (B) bull Effect of Additional 13-Similar Observations

1

1 I

WEIGHTING APPROACH bull

bull ASSESSMENT OF PLATFORMS DATA

Select Platforms Most Mappable to the Analyzed Platforms

Unlikely SurvivalDamage Cases Identification

Based on Cut-Off Criteria Agemiddot Hs etc

bull INFORMATION REQUIRED FOR COMPARISON

bull CATEGORIZATION OF ADDITIONALmiddot PLATFORMS

bull UPDATING OF CALIBRATION RESULTS

--

OVERVIEW OF PLATFORMS SURVIVEDDAMAGED -~

Operator -

- Platforms

Survived CS)

Platforms

Damaged CD)

Platforms Used In

Phase IJIP -

JI~ ~Bill~IRBDll -

Amoco -

Chevron -

Exxon

Mob II

Phllllps

Shell

Unocal -

30

143

72 22

7

39 25shy -

1

12

bull

1 bull

1

1(S)+1 (D)

3 (S) +3(D)

2 (S)

bull shy

bull

bull

bull

Total Partlclnants Platforms 338 - 15 10C6S+4D) -

Other Operators

310

13

3 (D)

Total Platforms 648 28 13C6S+7D) -

Summary of Andrew JIP bull Phase I Platforms

Plairshy w v- N-ber Lea-le 1 shy car Hind_ o-w N-bull Deplh fl fl IJpe la bull n Yeadon Jadtet lbelledtf liefalit -- _ -OD

IL H - -shy _ICllshy IL IL - - rQClto ~ _

K STUIK 137 1963 8 365 608S4 Cellardeck

-middot STl-shy 170 1964 4 -~ K 39 62269 Cellardeck - --ST134W 137 1911 4 43 608S4 Smvlnlshy- _ - -

WD90A 184 1964 8 K 267Sbull34 83 SOS31 SUboftllar deck SmvMI

MC311 343 1978 8 x S7 62 - MC397 468 1991 4 x shy S33bull6SO 65911 -shySmeSmvlvol -- gt -

r-- T23ST52 63 1969 4 ton- shyKshy 52594 S03 -T25SSl39 62 1969 4 Kshy SOS94 S063 - middot ton-shy__ -

bullbull a ton--shyST161A 118 1964 8 K - 347S 51495 Cellarclecti

r~

ST1771 142 1965 8 K 39 60167 CeDordeck failme

STISIH 137 1964 8 K 395 608S4 CeDardeck Failme

K ST130A 140 1958 a SS 6095 I lblv Pai1uno - T21ST72 61 1969 4 Kshy S4S94 4974 I lhlv Pai1uno -

- -

JE

INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON

bull WATER DEPTH

bull bull YEAR OF INSTALLATIONDESIGN

bull PLATFORM ORIENTATION

bull PLATFORM LATITUDE AND LONGITUDE

bull KEY SKETCHES OF PLATFORM STRUCTURAL FRAMING

Number of Legs Plies Leg Batter Number of Conductors bull Deck Elevations (Including sub-cellar deck If any) Jacket Framing Elevations Brace type In Vertical Frames Sizes of Primary Members (leg pile vertical frames braces primary horizontals) Joint Details (can gap etc) LegPiie Annulus Grouting Status

bull middotPOST ANDREW INSPECTION REPORT Significant Wave Height Hs (ft)

bull INFORMATION ON DAMAGES PRIOR TO ANDREW

bull RESULTS FROM IN-HOUSE STUDIES IF AVAILABLE

bull ~RIGINAL DESIGN CRITERIA INFORMATION IF AVAILABLE

bull SOIL SHEAR STRENGTH PROFILE IF AVAILABLE

Platforms Identified for Evaluating Therr Likelihood of Falling Under Unexpected Survival Category

OPlRATOR

middotADA

BLOClt

llmucnJllE

NAME

YEAll

INSTAUJD

WA1ER

DBPTll

DISTANCE

JlllOM SHOllE

(MILU)

Bl

(M)

AMOCO 111113

STl11

WDI075

A

bull D

St

17

120

1n

44

20

10 I

CllllVllQfl STlllO c 61 110 21 11 STIUO D Q 1110 21 11

STl134 p Q 137 29 11

ST0151 G 61 137 11

ST0151 I 3 121 11

ST0151 J Q 140 11

STl151 It 137 11

ST0151 PJl001 Q UT 11

ST0151 PROD-2 Q UT 11

ST017 A 3 140 35 11

WDIUT c 65 214 34 10

llXXON STI055 II 51 6T 14 I

STl15 A t3 34

WDI073

WD0073

Amiddotshy Q

1111 1111

22

22

I

I

WDI073 c 1n 22 I

- WDOCl93-shy II

65 1110 2T A 3 194 31 MOBU 111121

SS IOl7

A B

501

50

50

31

T

I

T 6

MtlllPllY SS114 B T 50 T T

PlllllIPS SS0149 A 5 55 33 T

SAMllDAN STln A 107 T T

SBllLL WDl103 A 65 223 2T WDl103 B 65 221 2T WD0104 c 65 221 2T WD0104 D 6T 20 2T

UNOCAL SS020I

SS 020I SSl20I

SS 0209

SS0209

II p

PURll

A

B

3

n 61

103

97

97

95

100

34

34

34

34 34

bullbull bullbullbull

- SS0209 D u 95 34 I

- SS 1209 SSl253

SS0253

G A

c

6T

2

100

165

1T5

34

54

54

bullbull I

I I

I I

SOUTH MARCH

~LAND

N

SOUTH PaTOI ENE ISWltD AREA

MMS Defined Limits ror Offshore Structures Requiring Inspections Following Andrew

t---Jbulln3 T25bull I bullTtl

bull1$1Htl1Abull bull1SIK1779bull

~

middot-DaTA AREA~T

CRANO rsLE

AREA bull1300

bullll4W

=shy

bullPath of Hurricane Andrew with Platforms Used In Calibration

Likelihood Function vs b All Success Case STISI K STIJOQ ST134W WD90A MCJll MCJ97

I 0000

0 9000

0 8000

07000

-Bo6000

tiiii 0 5000 -

04000

0 3000

0 2000

01000

0 0000

0 05 25

ri1rt middot-middot-middot-middot-middot-

I 5 2

b

---11 middot151K(blsu~ccsI

- bull - bull - fk 130Q(bfU~lCSS)

lie bull134W(blsuccc]

lk-WD90A[blsultCS]

- - - lk middotMC397(bfUtlCS]

--o-- lk-MCJll(bbuctcsJ

--- ll (bl 6 out of 6 Mlttcs case]

_

( Likelihood Functions middotSuccess Cases - p - 1 c I

CATEGORIZATION OF ADDITIONAL PLATFORMS

I I I

bull CATEGORIES

Unexpected Survivals

Expected Survivals

bull APPROACH

Identify Corresponding Platform(s) Evaluated In the JIP

Determine Max Wave Height at Platform Location

Evaluate Whether Wave Hit the Deck

middot - Evaluate Likelihood of R and S to Differ from the Base Cases

Review Joint Information

Assess Special Features

bull IDENTIFY NUMBER OF PLATFORMS IN EACH CATEGORY

I

UPDATING OF CALIBRATION RESULTS

bull INCLUDE EFFECT OF ADDITIONAL PLATFORMS ON bullBbull

Consider Increased 1Uncertalntles In R and S

Develop bullNeW Ukellhood Functions for Different Categories

middotLet bullxbull Number of Ukely Unexpected Survival Cases and y Number of Ukely Expected Survival Cases

Determine Combined Likelihood Function lk (b Iadditional platforms) middot

bull REVISED LIKELIHOOD FUNCTION

lk(b I all cases) = lk(b I 9 platforms) middot x lk(b Ix platforms) x lk(b I y platforms)

bull ESTABLISH REVISED BIAS FACTOR B

f8 (b) u f8 (b) lk (b I all cases)

II

t t ~

shy

APPLICATIONS OF BIAS FACTORS

ii APPLICABLE TO PARTICULAR LOAD AND RESISTANCE RECIPE

bull USE BY THE API COMMITTEES AND MMS

Criteria Revisions

Recommended Design Guidelines Revisions bull

bullbull REASSESSMENT STUDIES AND DECISION MAKING

middot

bull RISK ASSESSMENT STUDIES BY THE OPERATORS

Individual Platforms

Fleet of Platforms

middot IMR Program Implementation

bullbull NOT TO BE USED IN DESIGN

I I fl J

EXAMPLE USE OF BIAS FACTORS

bull PHASE I EXAMPLE APPLICATION

bull EVALUATION OF ANNUAL PROBABILITY OF FAILURE

Establish Annual Maximum Seastate Data

Determine Coefficients to Define Base Shear

Determine Ultimate Capacity of Platform

Determine Probablllty of Failure Versus bullbbull

-bull DETERMINE UPDATED PROBABILITY OF FAILURE

Pf - P [S gt R]

Pf I P [~ gt Rb I b] f 8 (b) db

Pr - J P [S gt Rb I b] f bull 8 (b) db

bull TOTAL PROBABILITIES OF FAILURE

Total Pf - 1 bull Il [1 middotPf] - l Pf bull

Total Pf - 1 bull Il [1middot Pf] - l Pfmiddot

Total Pr = 1 - n [1middot PfbullJ = l Pfshy

I I I

EXAMPLE APPLICATION bull PHASE I

PLATFORM ST 151 K

bull GULF OF MEXICO

bull WATER DEPTH = 137 FT

bull ANNUAL MAXIMUM HS MEDIAN 175 ft COV 031

bull ASSUMED STORM DURATION 3 hoursmiddot

bull ASSUMED NUMBER OF WAVES 800

bull MEDIAN HS FOR STORM HOURS 1 2 AND 3

bull MEDIAN CURRENT VALUE bull

bull DETERMINE ULTIMATE CAPACITY FOR THREE DIRECTIONS

bull DETERMINE ANNUAL FAILURE PROBABILITIES

Broadsidemiddot Diagonal End-on

0037 0025 0007

0071 0044 0018

0022 0016 0004

DETERMINE TOTAL ANNUAL PROBABILITY OF FAILURE

Pf - 007

Pf - 013

Pr - 004

bull

EXAMPLE USE OF BIAS FACTORS - PHASE II middot

bull DETERMINE PREDICTIVE PROBABILITY OF FAILURE

Case - 1

Pf _ S S Pf (b b2) fa1a2 (b b2) db db2

g bull -1 (b1 R I S) - 1

92 - (b2 R2 I S) - 1

Case -2

Pf s ~ [bJ f11 (b) db

Pf [bJ p [g lt 0)

g - bull 1 (b1 R1 I S) - 1

1

bull1) l

I

BENEFITS OF ANDREW JIP bull PHASE II

lt

bull IMPROVED LOAD AND RESISTANCE RECIPE

bull IMPROVED CALIBRATION PROCEDURE

bull MULTIPLE BIAS FACTORS

bull PROCEDURE TO INCLUDE EFFECT OF OTHER PLATFORMS

bull PROCEDURE TO USE MULTIPLE BIAS FACTORS

  • Structure Bookmarks
    • ANDREW PHASE II JOINT1 INDUSTRY PROJECT
    • PROGRESS MEETING NUMBER 1
    • ANDREW PHASE II JOINT INDUSTRY PROJECT PROGRESS MEETING NUMBER 1 NOVEMBER 16 AGENDA
    • ANDREW PHASE II UST OF PARTICIPANTS
    • ANDREW PHASE I OBJECTIVES AND RESULTS
    • ANDREW PHASE II PROJECT OBJECTIVES
    • SUBSEQUENT PROJECTS
    • ANDREW PHASE II PROJECT TEAM
    • ANDREW PHASE II UPDATE MODELING middotAND ANALYSIS RECIPE
    • -ANDREW PHASE II CONDUCTOR RESISTANCE
    • ANDREW PHASE Il WA VE FORCE PROFILE
    • ANDREW PHASE II WA VE FORCE PROFILEmiddot
    • ANDREW PHASE II BRACE MODELING
    • ANDREW PHASE II FOUNDATIONS
    • ANDREW PHASE II middot middot JOINT MODELING
    • ANDREW PHASE Il JOINT MODELING OPTIONS DISCUSSION
    • ANDREW PHASE II JOINT MODEJJNG OPTIONS ILLUSTRATIONS
    • ANDREW PHASE II JOINT MODELING PHASE II METHOD
    • ANDREW PHASE II ANDREW STORM CONDmONS UPDATED
    • Platform Selection bullbullbull1111111111111
    • BasisCriteria for Selection bullbullbull1111111111111
    • CALIBRATION
    • WHERE TO INTRODUCE BIAS FACTOR (B
    • PHASE I -KEY CONCLUSIONS
    • bull CALIBRATION ISSUES PHASE II
    • PHASE II WORK SCOPE (CALIBRATION)
    • SOFTWARE
    • SENSITIVITY STUDIES POSTERIOR DISTRIBUTION OF B
    • WEIGHTING APPROACH bull
    • INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON
    • CATEGORIZATION OF ADDITIONAL PLATFORMS
    • UPDATING OF CALIBRATION RESULTS
    • APPLICATIONS OF BIAS FACTORS
    • EXAMPLE USE OF BIAS FACTORS
    • EXAMPLE APPLICATION bull PHASE I PLATFORM ST 151 K
    • EXAMPLE USE OF BIAS FACTORS -PHASE II middot
    • BENEFITS OF ANDREW JIP bull PHASE II
Page 61:  · . ' ANDREW PHASE II PROJECT OBJECTIVES . ' • ·UPDATE MODELING AND ANALYSIS RECIPE· JOINTS ' ' ·BRACES WAVE PROFILE FOUNDATIONS . ' ' CONDUCTORS· • ANALVZE PLATFORMS USING

bullbull bull bull bull bull bull -------------------~

bull bull bullbull bullbull bullbull bullbull bullbull bullbull bull

bull bullbull

bullbullbullbull

bullbull

bullbullbullbullbullbullbull

bullbullbullbullbullbull

bullbullbullbullbullbullbull

z 0

t z ) LL

c 0 0c i ~ i

J

~ sect ~--------------~ e l5

I

I

SOFTWARE

bull PF PROGRAM

bull CALREL

bull PF +CALREL

bull RELACS

bull

gt i

l

SENSITIVITY STUDIES

POSTERIOR DISTRIBUTION OF B

bull EFFECT OF CHANGE IN DISTRIBUTION OF PRIOR OF B

bull EFFECT OF REMOVAL OF ONE PLATFORM FROM EACH CATEGORY

bull EFFECT OF REMOVAL OF THREE DAMAGE CASES

bull EFFECT OF ADDITIONAL 13 SIMILAR PLATFORMS

-I

Prior and Posterior Distributions of B Variation In COV of PriorofB fromOl to04 -- -

4500

-4000

-

- -

3500 --middot - bull bull - bull bull Prior-r ICOV =02

3000

- - - - Prior-r ICOV =03 -

bull -~ e 2soo -

bull bull bull bull bull bull bull Pnor-r ICOV =o 4t frJ ~ 2000 Posterior rlCOV =0 2

bull bull Postenor rlCOV =0 3 bull bull 1500 bull middot bull - shy -middot- Posterior rlCOV =04

middot~ I 000 - bull

-middotmiddot ~ ~middotmiddot M -~

- middot 0 500

wbullbullbull _ I middot~ middotmiddotmiddotmiddotmiddot middotmiddotmiddotmiddotmiddot - -~ ~ __ middot_ _ bull -shy0 000

0 000 0200 0 400 0 600 0 800 IOOO 1200 1400 1600 1800 2000 -

- - b

-

=shy

Posterior Distribution of Bias Factor (B) bull Effect of Variation In COV of Prior of B

bull

bullPrior and Posterior Distributions of B Evaluating Effect of Removal of One Platform from Each Category from the Base Case of 13 Platforms

400

~ ~ 350

j IA i

3 00 bull I bull I ---Prior-r I bull l

250 1 I 1

I bull bull ---Posterior r lall 13 casesshy- ~E 2 00 bull bull bull bull bull bull bull Posterior r without STISIK

- ~ 1 bull - bull I - bull bull - bull bull Posterior r Iwithout T2I I 50 I

J-tt~i - bull - bull - Posterior r Iwithout T2S

tOO [-

I I ~ ~ I 0 50

0 00

___ r J i

l~T

~--

0 00 0 20 0 40 0 60 080 I 00 120 140 160 110 2 00

b

zshy

Posterior Distribution of Bias Factor (8) bull Effect of Removal of middotPlatform from Any Category

bullPrior and Posterior Distributions or B Evaluating Effect or Removal or 3 Damage Cases from the Base Case or 13 Platrorms

4

35

3

25

e L 2 ~ bulle shy

I 15

bull I

05

0

0 02 04 06 I 2 I 4 16 IS 2

I I

I

I

I Ibull

J -tmiddot L--

bullI

~

I

fl

II ~

I I v I

middot bull bull

bull L

~ ll- I-~lt

OS I

b

I

bull

1middot bullbullbullbull

~ __

bull

- - - - Prior-r

- - - - Postenor flall failure cases

Postenor flall success cases

--- Posterior flall 13 cases

---Posterior fI success failure cases only

IPosterior or B Mean =t11 COV =0121

-

_I

Posterir Distribution or Bias Factor (B) bull Effect or Removal or 3-Damage Cases

-

bullPrior and Posterior Distributions or B Evaluating Effect or Additional 13 Similar Observations as

Used in Andrew JIP The POsterior or B from Andrew-JIP Considered as New Prior or B

- 500

450

400

350

bull bull

bull bull bull bull J bull

bull bull f ~ I

I ~ Ibull bull

bullbull bull

l bull I -bullbull bull bull

I

I bull bull

bull I I~~bull bull bull

bull ~

I bull bull

~ ~_ ~~~

- - New Prior r bull

Posterior 114 new failure cases ~ 300- bull bull bull bull bull bull bull Posterior f 13 new damage casesl 150 ~

- bull bull - bull bull Posterior 116 new success cases i 2 00

Postenor 1113 new success failure -150 damage cases

I 00

0 50 -

0 00 0 00 0 20 0 40 0 60 0 80 I 00 120 I 40 160 110 2 00

Posterior ol B Mean= 120 COV= 0071 b

shy

shy

_rshy

Posterior Distribution of Bias Factor (B) bull Effect of Additional 13-Similar Observations

1

1 I

WEIGHTING APPROACH bull

bull ASSESSMENT OF PLATFORMS DATA

Select Platforms Most Mappable to the Analyzed Platforms

Unlikely SurvivalDamage Cases Identification

Based on Cut-Off Criteria Agemiddot Hs etc

bull INFORMATION REQUIRED FOR COMPARISON

bull CATEGORIZATION OF ADDITIONALmiddot PLATFORMS

bull UPDATING OF CALIBRATION RESULTS

--

OVERVIEW OF PLATFORMS SURVIVEDDAMAGED -~

Operator -

- Platforms

Survived CS)

Platforms

Damaged CD)

Platforms Used In

Phase IJIP -

JI~ ~Bill~IRBDll -

Amoco -

Chevron -

Exxon

Mob II

Phllllps

Shell

Unocal -

30

143

72 22

7

39 25shy -

1

12

bull

1 bull

1

1(S)+1 (D)

3 (S) +3(D)

2 (S)

bull shy

bull

bull

bull

Total Partlclnants Platforms 338 - 15 10C6S+4D) -

Other Operators

310

13

3 (D)

Total Platforms 648 28 13C6S+7D) -

Summary of Andrew JIP bull Phase I Platforms

Plairshy w v- N-ber Lea-le 1 shy car Hind_ o-w N-bull Deplh fl fl IJpe la bull n Yeadon Jadtet lbelledtf liefalit -- _ -OD

IL H - -shy _ICllshy IL IL - - rQClto ~ _

K STUIK 137 1963 8 365 608S4 Cellardeck

-middot STl-shy 170 1964 4 -~ K 39 62269 Cellardeck - --ST134W 137 1911 4 43 608S4 Smvlnlshy- _ - -

WD90A 184 1964 8 K 267Sbull34 83 SOS31 SUboftllar deck SmvMI

MC311 343 1978 8 x S7 62 - MC397 468 1991 4 x shy S33bull6SO 65911 -shySmeSmvlvol -- gt -

r-- T23ST52 63 1969 4 ton- shyKshy 52594 S03 -T25SSl39 62 1969 4 Kshy SOS94 S063 - middot ton-shy__ -

bullbull a ton--shyST161A 118 1964 8 K - 347S 51495 Cellarclecti

r~

ST1771 142 1965 8 K 39 60167 CeDordeck failme

STISIH 137 1964 8 K 395 608S4 CeDardeck Failme

K ST130A 140 1958 a SS 6095 I lblv Pai1uno - T21ST72 61 1969 4 Kshy S4S94 4974 I lhlv Pai1uno -

- -

JE

INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON

bull WATER DEPTH

bull bull YEAR OF INSTALLATIONDESIGN

bull PLATFORM ORIENTATION

bull PLATFORM LATITUDE AND LONGITUDE

bull KEY SKETCHES OF PLATFORM STRUCTURAL FRAMING

Number of Legs Plies Leg Batter Number of Conductors bull Deck Elevations (Including sub-cellar deck If any) Jacket Framing Elevations Brace type In Vertical Frames Sizes of Primary Members (leg pile vertical frames braces primary horizontals) Joint Details (can gap etc) LegPiie Annulus Grouting Status

bull middotPOST ANDREW INSPECTION REPORT Significant Wave Height Hs (ft)

bull INFORMATION ON DAMAGES PRIOR TO ANDREW

bull RESULTS FROM IN-HOUSE STUDIES IF AVAILABLE

bull ~RIGINAL DESIGN CRITERIA INFORMATION IF AVAILABLE

bull SOIL SHEAR STRENGTH PROFILE IF AVAILABLE

Platforms Identified for Evaluating Therr Likelihood of Falling Under Unexpected Survival Category

OPlRATOR

middotADA

BLOClt

llmucnJllE

NAME

YEAll

INSTAUJD

WA1ER

DBPTll

DISTANCE

JlllOM SHOllE

(MILU)

Bl

(M)

AMOCO 111113

STl11

WDI075

A

bull D

St

17

120

1n

44

20

10 I

CllllVllQfl STlllO c 61 110 21 11 STIUO D Q 1110 21 11

STl134 p Q 137 29 11

ST0151 G 61 137 11

ST0151 I 3 121 11

ST0151 J Q 140 11

STl151 It 137 11

ST0151 PJl001 Q UT 11

ST0151 PROD-2 Q UT 11

ST017 A 3 140 35 11

WDIUT c 65 214 34 10

llXXON STI055 II 51 6T 14 I

STl15 A t3 34

WDI073

WD0073

Amiddotshy Q

1111 1111

22

22

I

I

WDI073 c 1n 22 I

- WDOCl93-shy II

65 1110 2T A 3 194 31 MOBU 111121

SS IOl7

A B

501

50

50

31

T

I

T 6

MtlllPllY SS114 B T 50 T T

PlllllIPS SS0149 A 5 55 33 T

SAMllDAN STln A 107 T T

SBllLL WDl103 A 65 223 2T WDl103 B 65 221 2T WD0104 c 65 221 2T WD0104 D 6T 20 2T

UNOCAL SS020I

SS 020I SSl20I

SS 0209

SS0209

II p

PURll

A

B

3

n 61

103

97

97

95

100

34

34

34

34 34

bullbull bullbullbull

- SS0209 D u 95 34 I

- SS 1209 SSl253

SS0253

G A

c

6T

2

100

165

1T5

34

54

54

bullbull I

I I

I I

SOUTH MARCH

~LAND

N

SOUTH PaTOI ENE ISWltD AREA

MMS Defined Limits ror Offshore Structures Requiring Inspections Following Andrew

t---Jbulln3 T25bull I bullTtl

bull1$1Htl1Abull bull1SIK1779bull

~

middot-DaTA AREA~T

CRANO rsLE

AREA bull1300

bullll4W

=shy

bullPath of Hurricane Andrew with Platforms Used In Calibration

Likelihood Function vs b All Success Case STISI K STIJOQ ST134W WD90A MCJll MCJ97

I 0000

0 9000

0 8000

07000

-Bo6000

tiiii 0 5000 -

04000

0 3000

0 2000

01000

0 0000

0 05 25

ri1rt middot-middot-middot-middot-middot-

I 5 2

b

---11 middot151K(blsu~ccsI

- bull - bull - fk 130Q(bfU~lCSS)

lie bull134W(blsuccc]

lk-WD90A[blsultCS]

- - - lk middotMC397(bfUtlCS]

--o-- lk-MCJll(bbuctcsJ

--- ll (bl 6 out of 6 Mlttcs case]

_

( Likelihood Functions middotSuccess Cases - p - 1 c I

CATEGORIZATION OF ADDITIONAL PLATFORMS

I I I

bull CATEGORIES

Unexpected Survivals

Expected Survivals

bull APPROACH

Identify Corresponding Platform(s) Evaluated In the JIP

Determine Max Wave Height at Platform Location

Evaluate Whether Wave Hit the Deck

middot - Evaluate Likelihood of R and S to Differ from the Base Cases

Review Joint Information

Assess Special Features

bull IDENTIFY NUMBER OF PLATFORMS IN EACH CATEGORY

I

UPDATING OF CALIBRATION RESULTS

bull INCLUDE EFFECT OF ADDITIONAL PLATFORMS ON bullBbull

Consider Increased 1Uncertalntles In R and S

Develop bullNeW Ukellhood Functions for Different Categories

middotLet bullxbull Number of Ukely Unexpected Survival Cases and y Number of Ukely Expected Survival Cases

Determine Combined Likelihood Function lk (b Iadditional platforms) middot

bull REVISED LIKELIHOOD FUNCTION

lk(b I all cases) = lk(b I 9 platforms) middot x lk(b Ix platforms) x lk(b I y platforms)

bull ESTABLISH REVISED BIAS FACTOR B

f8 (b) u f8 (b) lk (b I all cases)

II

t t ~

shy

APPLICATIONS OF BIAS FACTORS

ii APPLICABLE TO PARTICULAR LOAD AND RESISTANCE RECIPE

bull USE BY THE API COMMITTEES AND MMS

Criteria Revisions

Recommended Design Guidelines Revisions bull

bullbull REASSESSMENT STUDIES AND DECISION MAKING

middot

bull RISK ASSESSMENT STUDIES BY THE OPERATORS

Individual Platforms

Fleet of Platforms

middot IMR Program Implementation

bullbull NOT TO BE USED IN DESIGN

I I fl J

EXAMPLE USE OF BIAS FACTORS

bull PHASE I EXAMPLE APPLICATION

bull EVALUATION OF ANNUAL PROBABILITY OF FAILURE

Establish Annual Maximum Seastate Data

Determine Coefficients to Define Base Shear

Determine Ultimate Capacity of Platform

Determine Probablllty of Failure Versus bullbbull

-bull DETERMINE UPDATED PROBABILITY OF FAILURE

Pf - P [S gt R]

Pf I P [~ gt Rb I b] f 8 (b) db

Pr - J P [S gt Rb I b] f bull 8 (b) db

bull TOTAL PROBABILITIES OF FAILURE

Total Pf - 1 bull Il [1 middotPf] - l Pf bull

Total Pf - 1 bull Il [1middot Pf] - l Pfmiddot

Total Pr = 1 - n [1middot PfbullJ = l Pfshy

I I I

EXAMPLE APPLICATION bull PHASE I

PLATFORM ST 151 K

bull GULF OF MEXICO

bull WATER DEPTH = 137 FT

bull ANNUAL MAXIMUM HS MEDIAN 175 ft COV 031

bull ASSUMED STORM DURATION 3 hoursmiddot

bull ASSUMED NUMBER OF WAVES 800

bull MEDIAN HS FOR STORM HOURS 1 2 AND 3

bull MEDIAN CURRENT VALUE bull

bull DETERMINE ULTIMATE CAPACITY FOR THREE DIRECTIONS

bull DETERMINE ANNUAL FAILURE PROBABILITIES

Broadsidemiddot Diagonal End-on

0037 0025 0007

0071 0044 0018

0022 0016 0004

DETERMINE TOTAL ANNUAL PROBABILITY OF FAILURE

Pf - 007

Pf - 013

Pr - 004

bull

EXAMPLE USE OF BIAS FACTORS - PHASE II middot

bull DETERMINE PREDICTIVE PROBABILITY OF FAILURE

Case - 1

Pf _ S S Pf (b b2) fa1a2 (b b2) db db2

g bull -1 (b1 R I S) - 1

92 - (b2 R2 I S) - 1

Case -2

Pf s ~ [bJ f11 (b) db

Pf [bJ p [g lt 0)

g - bull 1 (b1 R1 I S) - 1

1

bull1) l

I

BENEFITS OF ANDREW JIP bull PHASE II

lt

bull IMPROVED LOAD AND RESISTANCE RECIPE

bull IMPROVED CALIBRATION PROCEDURE

bull MULTIPLE BIAS FACTORS

bull PROCEDURE TO INCLUDE EFFECT OF OTHER PLATFORMS

bull PROCEDURE TO USE MULTIPLE BIAS FACTORS

  • Structure Bookmarks
    • ANDREW PHASE II JOINT1 INDUSTRY PROJECT
    • PROGRESS MEETING NUMBER 1
    • ANDREW PHASE II JOINT INDUSTRY PROJECT PROGRESS MEETING NUMBER 1 NOVEMBER 16 AGENDA
    • ANDREW PHASE II UST OF PARTICIPANTS
    • ANDREW PHASE I OBJECTIVES AND RESULTS
    • ANDREW PHASE II PROJECT OBJECTIVES
    • SUBSEQUENT PROJECTS
    • ANDREW PHASE II PROJECT TEAM
    • ANDREW PHASE II UPDATE MODELING middotAND ANALYSIS RECIPE
    • -ANDREW PHASE II CONDUCTOR RESISTANCE
    • ANDREW PHASE Il WA VE FORCE PROFILE
    • ANDREW PHASE II WA VE FORCE PROFILEmiddot
    • ANDREW PHASE II BRACE MODELING
    • ANDREW PHASE II FOUNDATIONS
    • ANDREW PHASE II middot middot JOINT MODELING
    • ANDREW PHASE Il JOINT MODELING OPTIONS DISCUSSION
    • ANDREW PHASE II JOINT MODEJJNG OPTIONS ILLUSTRATIONS
    • ANDREW PHASE II JOINT MODELING PHASE II METHOD
    • ANDREW PHASE II ANDREW STORM CONDmONS UPDATED
    • Platform Selection bullbullbull1111111111111
    • BasisCriteria for Selection bullbullbull1111111111111
    • CALIBRATION
    • WHERE TO INTRODUCE BIAS FACTOR (B
    • PHASE I -KEY CONCLUSIONS
    • bull CALIBRATION ISSUES PHASE II
    • PHASE II WORK SCOPE (CALIBRATION)
    • SOFTWARE
    • SENSITIVITY STUDIES POSTERIOR DISTRIBUTION OF B
    • WEIGHTING APPROACH bull
    • INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON
    • CATEGORIZATION OF ADDITIONAL PLATFORMS
    • UPDATING OF CALIBRATION RESULTS
    • APPLICATIONS OF BIAS FACTORS
    • EXAMPLE USE OF BIAS FACTORS
    • EXAMPLE APPLICATION bull PHASE I PLATFORM ST 151 K
    • EXAMPLE USE OF BIAS FACTORS -PHASE II middot
    • BENEFITS OF ANDREW JIP bull PHASE II
Page 62:  · . ' ANDREW PHASE II PROJECT OBJECTIVES . ' • ·UPDATE MODELING AND ANALYSIS RECIPE· JOINTS ' ' ·BRACES WAVE PROFILE FOUNDATIONS . ' ' CONDUCTORS· • ANALVZE PLATFORMS USING

I

SOFTWARE

bull PF PROGRAM

bull CALREL

bull PF +CALREL

bull RELACS

bull

gt i

l

SENSITIVITY STUDIES

POSTERIOR DISTRIBUTION OF B

bull EFFECT OF CHANGE IN DISTRIBUTION OF PRIOR OF B

bull EFFECT OF REMOVAL OF ONE PLATFORM FROM EACH CATEGORY

bull EFFECT OF REMOVAL OF THREE DAMAGE CASES

bull EFFECT OF ADDITIONAL 13 SIMILAR PLATFORMS

-I

Prior and Posterior Distributions of B Variation In COV of PriorofB fromOl to04 -- -

4500

-4000

-

- -

3500 --middot - bull bull - bull bull Prior-r ICOV =02

3000

- - - - Prior-r ICOV =03 -

bull -~ e 2soo -

bull bull bull bull bull bull bull Pnor-r ICOV =o 4t frJ ~ 2000 Posterior rlCOV =0 2

bull bull Postenor rlCOV =0 3 bull bull 1500 bull middot bull - shy -middot- Posterior rlCOV =04

middot~ I 000 - bull

-middotmiddot ~ ~middotmiddot M -~

- middot 0 500

wbullbullbull _ I middot~ middotmiddotmiddotmiddotmiddot middotmiddotmiddotmiddotmiddot - -~ ~ __ middot_ _ bull -shy0 000

0 000 0200 0 400 0 600 0 800 IOOO 1200 1400 1600 1800 2000 -

- - b

-

=shy

Posterior Distribution of Bias Factor (B) bull Effect of Variation In COV of Prior of B

bull

bullPrior and Posterior Distributions of B Evaluating Effect of Removal of One Platform from Each Category from the Base Case of 13 Platforms

400

~ ~ 350

j IA i

3 00 bull I bull I ---Prior-r I bull l

250 1 I 1

I bull bull ---Posterior r lall 13 casesshy- ~E 2 00 bull bull bull bull bull bull bull Posterior r without STISIK

- ~ 1 bull - bull I - bull bull - bull bull Posterior r Iwithout T2I I 50 I

J-tt~i - bull - bull - Posterior r Iwithout T2S

tOO [-

I I ~ ~ I 0 50

0 00

___ r J i

l~T

~--

0 00 0 20 0 40 0 60 080 I 00 120 140 160 110 2 00

b

zshy

Posterior Distribution of Bias Factor (8) bull Effect of Removal of middotPlatform from Any Category

bullPrior and Posterior Distributions or B Evaluating Effect or Removal or 3 Damage Cases from the Base Case or 13 Platrorms

4

35

3

25

e L 2 ~ bulle shy

I 15

bull I

05

0

0 02 04 06 I 2 I 4 16 IS 2

I I

I

I

I Ibull

J -tmiddot L--

bullI

~

I

fl

II ~

I I v I

middot bull bull

bull L

~ ll- I-~lt

OS I

b

I

bull

1middot bullbullbullbull

~ __

bull

- - - - Prior-r

- - - - Postenor flall failure cases

Postenor flall success cases

--- Posterior flall 13 cases

---Posterior fI success failure cases only

IPosterior or B Mean =t11 COV =0121

-

_I

Posterir Distribution or Bias Factor (B) bull Effect or Removal or 3-Damage Cases

-

bullPrior and Posterior Distributions or B Evaluating Effect or Additional 13 Similar Observations as

Used in Andrew JIP The POsterior or B from Andrew-JIP Considered as New Prior or B

- 500

450

400

350

bull bull

bull bull bull bull J bull

bull bull f ~ I

I ~ Ibull bull

bullbull bull

l bull I -bullbull bull bull

I

I bull bull

bull I I~~bull bull bull

bull ~

I bull bull

~ ~_ ~~~

- - New Prior r bull

Posterior 114 new failure cases ~ 300- bull bull bull bull bull bull bull Posterior f 13 new damage casesl 150 ~

- bull bull - bull bull Posterior 116 new success cases i 2 00

Postenor 1113 new success failure -150 damage cases

I 00

0 50 -

0 00 0 00 0 20 0 40 0 60 0 80 I 00 120 I 40 160 110 2 00

Posterior ol B Mean= 120 COV= 0071 b

shy

shy

_rshy

Posterior Distribution of Bias Factor (B) bull Effect of Additional 13-Similar Observations

1

1 I

WEIGHTING APPROACH bull

bull ASSESSMENT OF PLATFORMS DATA

Select Platforms Most Mappable to the Analyzed Platforms

Unlikely SurvivalDamage Cases Identification

Based on Cut-Off Criteria Agemiddot Hs etc

bull INFORMATION REQUIRED FOR COMPARISON

bull CATEGORIZATION OF ADDITIONALmiddot PLATFORMS

bull UPDATING OF CALIBRATION RESULTS

--

OVERVIEW OF PLATFORMS SURVIVEDDAMAGED -~

Operator -

- Platforms

Survived CS)

Platforms

Damaged CD)

Platforms Used In

Phase IJIP -

JI~ ~Bill~IRBDll -

Amoco -

Chevron -

Exxon

Mob II

Phllllps

Shell

Unocal -

30

143

72 22

7

39 25shy -

1

12

bull

1 bull

1

1(S)+1 (D)

3 (S) +3(D)

2 (S)

bull shy

bull

bull

bull

Total Partlclnants Platforms 338 - 15 10C6S+4D) -

Other Operators

310

13

3 (D)

Total Platforms 648 28 13C6S+7D) -

Summary of Andrew JIP bull Phase I Platforms

Plairshy w v- N-ber Lea-le 1 shy car Hind_ o-w N-bull Deplh fl fl IJpe la bull n Yeadon Jadtet lbelledtf liefalit -- _ -OD

IL H - -shy _ICllshy IL IL - - rQClto ~ _

K STUIK 137 1963 8 365 608S4 Cellardeck

-middot STl-shy 170 1964 4 -~ K 39 62269 Cellardeck - --ST134W 137 1911 4 43 608S4 Smvlnlshy- _ - -

WD90A 184 1964 8 K 267Sbull34 83 SOS31 SUboftllar deck SmvMI

MC311 343 1978 8 x S7 62 - MC397 468 1991 4 x shy S33bull6SO 65911 -shySmeSmvlvol -- gt -

r-- T23ST52 63 1969 4 ton- shyKshy 52594 S03 -T25SSl39 62 1969 4 Kshy SOS94 S063 - middot ton-shy__ -

bullbull a ton--shyST161A 118 1964 8 K - 347S 51495 Cellarclecti

r~

ST1771 142 1965 8 K 39 60167 CeDordeck failme

STISIH 137 1964 8 K 395 608S4 CeDardeck Failme

K ST130A 140 1958 a SS 6095 I lblv Pai1uno - T21ST72 61 1969 4 Kshy S4S94 4974 I lhlv Pai1uno -

- -

JE

INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON

bull WATER DEPTH

bull bull YEAR OF INSTALLATIONDESIGN

bull PLATFORM ORIENTATION

bull PLATFORM LATITUDE AND LONGITUDE

bull KEY SKETCHES OF PLATFORM STRUCTURAL FRAMING

Number of Legs Plies Leg Batter Number of Conductors bull Deck Elevations (Including sub-cellar deck If any) Jacket Framing Elevations Brace type In Vertical Frames Sizes of Primary Members (leg pile vertical frames braces primary horizontals) Joint Details (can gap etc) LegPiie Annulus Grouting Status

bull middotPOST ANDREW INSPECTION REPORT Significant Wave Height Hs (ft)

bull INFORMATION ON DAMAGES PRIOR TO ANDREW

bull RESULTS FROM IN-HOUSE STUDIES IF AVAILABLE

bull ~RIGINAL DESIGN CRITERIA INFORMATION IF AVAILABLE

bull SOIL SHEAR STRENGTH PROFILE IF AVAILABLE

Platforms Identified for Evaluating Therr Likelihood of Falling Under Unexpected Survival Category

OPlRATOR

middotADA

BLOClt

llmucnJllE

NAME

YEAll

INSTAUJD

WA1ER

DBPTll

DISTANCE

JlllOM SHOllE

(MILU)

Bl

(M)

AMOCO 111113

STl11

WDI075

A

bull D

St

17

120

1n

44

20

10 I

CllllVllQfl STlllO c 61 110 21 11 STIUO D Q 1110 21 11

STl134 p Q 137 29 11

ST0151 G 61 137 11

ST0151 I 3 121 11

ST0151 J Q 140 11

STl151 It 137 11

ST0151 PJl001 Q UT 11

ST0151 PROD-2 Q UT 11

ST017 A 3 140 35 11

WDIUT c 65 214 34 10

llXXON STI055 II 51 6T 14 I

STl15 A t3 34

WDI073

WD0073

Amiddotshy Q

1111 1111

22

22

I

I

WDI073 c 1n 22 I

- WDOCl93-shy II

65 1110 2T A 3 194 31 MOBU 111121

SS IOl7

A B

501

50

50

31

T

I

T 6

MtlllPllY SS114 B T 50 T T

PlllllIPS SS0149 A 5 55 33 T

SAMllDAN STln A 107 T T

SBllLL WDl103 A 65 223 2T WDl103 B 65 221 2T WD0104 c 65 221 2T WD0104 D 6T 20 2T

UNOCAL SS020I

SS 020I SSl20I

SS 0209

SS0209

II p

PURll

A

B

3

n 61

103

97

97

95

100

34

34

34

34 34

bullbull bullbullbull

- SS0209 D u 95 34 I

- SS 1209 SSl253

SS0253

G A

c

6T

2

100

165

1T5

34

54

54

bullbull I

I I

I I

SOUTH MARCH

~LAND

N

SOUTH PaTOI ENE ISWltD AREA

MMS Defined Limits ror Offshore Structures Requiring Inspections Following Andrew

t---Jbulln3 T25bull I bullTtl

bull1$1Htl1Abull bull1SIK1779bull

~

middot-DaTA AREA~T

CRANO rsLE

AREA bull1300

bullll4W

=shy

bullPath of Hurricane Andrew with Platforms Used In Calibration

Likelihood Function vs b All Success Case STISI K STIJOQ ST134W WD90A MCJll MCJ97

I 0000

0 9000

0 8000

07000

-Bo6000

tiiii 0 5000 -

04000

0 3000

0 2000

01000

0 0000

0 05 25

ri1rt middot-middot-middot-middot-middot-

I 5 2

b

---11 middot151K(blsu~ccsI

- bull - bull - fk 130Q(bfU~lCSS)

lie bull134W(blsuccc]

lk-WD90A[blsultCS]

- - - lk middotMC397(bfUtlCS]

--o-- lk-MCJll(bbuctcsJ

--- ll (bl 6 out of 6 Mlttcs case]

_

( Likelihood Functions middotSuccess Cases - p - 1 c I

CATEGORIZATION OF ADDITIONAL PLATFORMS

I I I

bull CATEGORIES

Unexpected Survivals

Expected Survivals

bull APPROACH

Identify Corresponding Platform(s) Evaluated In the JIP

Determine Max Wave Height at Platform Location

Evaluate Whether Wave Hit the Deck

middot - Evaluate Likelihood of R and S to Differ from the Base Cases

Review Joint Information

Assess Special Features

bull IDENTIFY NUMBER OF PLATFORMS IN EACH CATEGORY

I

UPDATING OF CALIBRATION RESULTS

bull INCLUDE EFFECT OF ADDITIONAL PLATFORMS ON bullBbull

Consider Increased 1Uncertalntles In R and S

Develop bullNeW Ukellhood Functions for Different Categories

middotLet bullxbull Number of Ukely Unexpected Survival Cases and y Number of Ukely Expected Survival Cases

Determine Combined Likelihood Function lk (b Iadditional platforms) middot

bull REVISED LIKELIHOOD FUNCTION

lk(b I all cases) = lk(b I 9 platforms) middot x lk(b Ix platforms) x lk(b I y platforms)

bull ESTABLISH REVISED BIAS FACTOR B

f8 (b) u f8 (b) lk (b I all cases)

II

t t ~

shy

APPLICATIONS OF BIAS FACTORS

ii APPLICABLE TO PARTICULAR LOAD AND RESISTANCE RECIPE

bull USE BY THE API COMMITTEES AND MMS

Criteria Revisions

Recommended Design Guidelines Revisions bull

bullbull REASSESSMENT STUDIES AND DECISION MAKING

middot

bull RISK ASSESSMENT STUDIES BY THE OPERATORS

Individual Platforms

Fleet of Platforms

middot IMR Program Implementation

bullbull NOT TO BE USED IN DESIGN

I I fl J

EXAMPLE USE OF BIAS FACTORS

bull PHASE I EXAMPLE APPLICATION

bull EVALUATION OF ANNUAL PROBABILITY OF FAILURE

Establish Annual Maximum Seastate Data

Determine Coefficients to Define Base Shear

Determine Ultimate Capacity of Platform

Determine Probablllty of Failure Versus bullbbull

-bull DETERMINE UPDATED PROBABILITY OF FAILURE

Pf - P [S gt R]

Pf I P [~ gt Rb I b] f 8 (b) db

Pr - J P [S gt Rb I b] f bull 8 (b) db

bull TOTAL PROBABILITIES OF FAILURE

Total Pf - 1 bull Il [1 middotPf] - l Pf bull

Total Pf - 1 bull Il [1middot Pf] - l Pfmiddot

Total Pr = 1 - n [1middot PfbullJ = l Pfshy

I I I

EXAMPLE APPLICATION bull PHASE I

PLATFORM ST 151 K

bull GULF OF MEXICO

bull WATER DEPTH = 137 FT

bull ANNUAL MAXIMUM HS MEDIAN 175 ft COV 031

bull ASSUMED STORM DURATION 3 hoursmiddot

bull ASSUMED NUMBER OF WAVES 800

bull MEDIAN HS FOR STORM HOURS 1 2 AND 3

bull MEDIAN CURRENT VALUE bull

bull DETERMINE ULTIMATE CAPACITY FOR THREE DIRECTIONS

bull DETERMINE ANNUAL FAILURE PROBABILITIES

Broadsidemiddot Diagonal End-on

0037 0025 0007

0071 0044 0018

0022 0016 0004

DETERMINE TOTAL ANNUAL PROBABILITY OF FAILURE

Pf - 007

Pf - 013

Pr - 004

bull

EXAMPLE USE OF BIAS FACTORS - PHASE II middot

bull DETERMINE PREDICTIVE PROBABILITY OF FAILURE

Case - 1

Pf _ S S Pf (b b2) fa1a2 (b b2) db db2

g bull -1 (b1 R I S) - 1

92 - (b2 R2 I S) - 1

Case -2

Pf s ~ [bJ f11 (b) db

Pf [bJ p [g lt 0)

g - bull 1 (b1 R1 I S) - 1

1

bull1) l

I

BENEFITS OF ANDREW JIP bull PHASE II

lt

bull IMPROVED LOAD AND RESISTANCE RECIPE

bull IMPROVED CALIBRATION PROCEDURE

bull MULTIPLE BIAS FACTORS

bull PROCEDURE TO INCLUDE EFFECT OF OTHER PLATFORMS

bull PROCEDURE TO USE MULTIPLE BIAS FACTORS

  • Structure Bookmarks
    • ANDREW PHASE II JOINT1 INDUSTRY PROJECT
    • PROGRESS MEETING NUMBER 1
    • ANDREW PHASE II JOINT INDUSTRY PROJECT PROGRESS MEETING NUMBER 1 NOVEMBER 16 AGENDA
    • ANDREW PHASE II UST OF PARTICIPANTS
    • ANDREW PHASE I OBJECTIVES AND RESULTS
    • ANDREW PHASE II PROJECT OBJECTIVES
    • SUBSEQUENT PROJECTS
    • ANDREW PHASE II PROJECT TEAM
    • ANDREW PHASE II UPDATE MODELING middotAND ANALYSIS RECIPE
    • -ANDREW PHASE II CONDUCTOR RESISTANCE
    • ANDREW PHASE Il WA VE FORCE PROFILE
    • ANDREW PHASE II WA VE FORCE PROFILEmiddot
    • ANDREW PHASE II BRACE MODELING
    • ANDREW PHASE II FOUNDATIONS
    • ANDREW PHASE II middot middot JOINT MODELING
    • ANDREW PHASE Il JOINT MODELING OPTIONS DISCUSSION
    • ANDREW PHASE II JOINT MODEJJNG OPTIONS ILLUSTRATIONS
    • ANDREW PHASE II JOINT MODELING PHASE II METHOD
    • ANDREW PHASE II ANDREW STORM CONDmONS UPDATED
    • Platform Selection bullbullbull1111111111111
    • BasisCriteria for Selection bullbullbull1111111111111
    • CALIBRATION
    • WHERE TO INTRODUCE BIAS FACTOR (B
    • PHASE I -KEY CONCLUSIONS
    • bull CALIBRATION ISSUES PHASE II
    • PHASE II WORK SCOPE (CALIBRATION)
    • SOFTWARE
    • SENSITIVITY STUDIES POSTERIOR DISTRIBUTION OF B
    • WEIGHTING APPROACH bull
    • INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON
    • CATEGORIZATION OF ADDITIONAL PLATFORMS
    • UPDATING OF CALIBRATION RESULTS
    • APPLICATIONS OF BIAS FACTORS
    • EXAMPLE USE OF BIAS FACTORS
    • EXAMPLE APPLICATION bull PHASE I PLATFORM ST 151 K
    • EXAMPLE USE OF BIAS FACTORS -PHASE II middot
    • BENEFITS OF ANDREW JIP bull PHASE II
Page 63:  · . ' ANDREW PHASE II PROJECT OBJECTIVES . ' • ·UPDATE MODELING AND ANALYSIS RECIPE· JOINTS ' ' ·BRACES WAVE PROFILE FOUNDATIONS . ' ' CONDUCTORS· • ANALVZE PLATFORMS USING

gt i

l

SENSITIVITY STUDIES

POSTERIOR DISTRIBUTION OF B

bull EFFECT OF CHANGE IN DISTRIBUTION OF PRIOR OF B

bull EFFECT OF REMOVAL OF ONE PLATFORM FROM EACH CATEGORY

bull EFFECT OF REMOVAL OF THREE DAMAGE CASES

bull EFFECT OF ADDITIONAL 13 SIMILAR PLATFORMS

-I

Prior and Posterior Distributions of B Variation In COV of PriorofB fromOl to04 -- -

4500

-4000

-

- -

3500 --middot - bull bull - bull bull Prior-r ICOV =02

3000

- - - - Prior-r ICOV =03 -

bull -~ e 2soo -

bull bull bull bull bull bull bull Pnor-r ICOV =o 4t frJ ~ 2000 Posterior rlCOV =0 2

bull bull Postenor rlCOV =0 3 bull bull 1500 bull middot bull - shy -middot- Posterior rlCOV =04

middot~ I 000 - bull

-middotmiddot ~ ~middotmiddot M -~

- middot 0 500

wbullbullbull _ I middot~ middotmiddotmiddotmiddotmiddot middotmiddotmiddotmiddotmiddot - -~ ~ __ middot_ _ bull -shy0 000

0 000 0200 0 400 0 600 0 800 IOOO 1200 1400 1600 1800 2000 -

- - b

-

=shy

Posterior Distribution of Bias Factor (B) bull Effect of Variation In COV of Prior of B

bull

bullPrior and Posterior Distributions of B Evaluating Effect of Removal of One Platform from Each Category from the Base Case of 13 Platforms

400

~ ~ 350

j IA i

3 00 bull I bull I ---Prior-r I bull l

250 1 I 1

I bull bull ---Posterior r lall 13 casesshy- ~E 2 00 bull bull bull bull bull bull bull Posterior r without STISIK

- ~ 1 bull - bull I - bull bull - bull bull Posterior r Iwithout T2I I 50 I

J-tt~i - bull - bull - Posterior r Iwithout T2S

tOO [-

I I ~ ~ I 0 50

0 00

___ r J i

l~T

~--

0 00 0 20 0 40 0 60 080 I 00 120 140 160 110 2 00

b

zshy

Posterior Distribution of Bias Factor (8) bull Effect of Removal of middotPlatform from Any Category

bullPrior and Posterior Distributions or B Evaluating Effect or Removal or 3 Damage Cases from the Base Case or 13 Platrorms

4

35

3

25

e L 2 ~ bulle shy

I 15

bull I

05

0

0 02 04 06 I 2 I 4 16 IS 2

I I

I

I

I Ibull

J -tmiddot L--

bullI

~

I

fl

II ~

I I v I

middot bull bull

bull L

~ ll- I-~lt

OS I

b

I

bull

1middot bullbullbullbull

~ __

bull

- - - - Prior-r

- - - - Postenor flall failure cases

Postenor flall success cases

--- Posterior flall 13 cases

---Posterior fI success failure cases only

IPosterior or B Mean =t11 COV =0121

-

_I

Posterir Distribution or Bias Factor (B) bull Effect or Removal or 3-Damage Cases

-

bullPrior and Posterior Distributions or B Evaluating Effect or Additional 13 Similar Observations as

Used in Andrew JIP The POsterior or B from Andrew-JIP Considered as New Prior or B

- 500

450

400

350

bull bull

bull bull bull bull J bull

bull bull f ~ I

I ~ Ibull bull

bullbull bull

l bull I -bullbull bull bull

I

I bull bull

bull I I~~bull bull bull

bull ~

I bull bull

~ ~_ ~~~

- - New Prior r bull

Posterior 114 new failure cases ~ 300- bull bull bull bull bull bull bull Posterior f 13 new damage casesl 150 ~

- bull bull - bull bull Posterior 116 new success cases i 2 00

Postenor 1113 new success failure -150 damage cases

I 00

0 50 -

0 00 0 00 0 20 0 40 0 60 0 80 I 00 120 I 40 160 110 2 00

Posterior ol B Mean= 120 COV= 0071 b

shy

shy

_rshy

Posterior Distribution of Bias Factor (B) bull Effect of Additional 13-Similar Observations

1

1 I

WEIGHTING APPROACH bull

bull ASSESSMENT OF PLATFORMS DATA

Select Platforms Most Mappable to the Analyzed Platforms

Unlikely SurvivalDamage Cases Identification

Based on Cut-Off Criteria Agemiddot Hs etc

bull INFORMATION REQUIRED FOR COMPARISON

bull CATEGORIZATION OF ADDITIONALmiddot PLATFORMS

bull UPDATING OF CALIBRATION RESULTS

--

OVERVIEW OF PLATFORMS SURVIVEDDAMAGED -~

Operator -

- Platforms

Survived CS)

Platforms

Damaged CD)

Platforms Used In

Phase IJIP -

JI~ ~Bill~IRBDll -

Amoco -

Chevron -

Exxon

Mob II

Phllllps

Shell

Unocal -

30

143

72 22

7

39 25shy -

1

12

bull

1 bull

1

1(S)+1 (D)

3 (S) +3(D)

2 (S)

bull shy

bull

bull

bull

Total Partlclnants Platforms 338 - 15 10C6S+4D) -

Other Operators

310

13

3 (D)

Total Platforms 648 28 13C6S+7D) -

Summary of Andrew JIP bull Phase I Platforms

Plairshy w v- N-ber Lea-le 1 shy car Hind_ o-w N-bull Deplh fl fl IJpe la bull n Yeadon Jadtet lbelledtf liefalit -- _ -OD

IL H - -shy _ICllshy IL IL - - rQClto ~ _

K STUIK 137 1963 8 365 608S4 Cellardeck

-middot STl-shy 170 1964 4 -~ K 39 62269 Cellardeck - --ST134W 137 1911 4 43 608S4 Smvlnlshy- _ - -

WD90A 184 1964 8 K 267Sbull34 83 SOS31 SUboftllar deck SmvMI

MC311 343 1978 8 x S7 62 - MC397 468 1991 4 x shy S33bull6SO 65911 -shySmeSmvlvol -- gt -

r-- T23ST52 63 1969 4 ton- shyKshy 52594 S03 -T25SSl39 62 1969 4 Kshy SOS94 S063 - middot ton-shy__ -

bullbull a ton--shyST161A 118 1964 8 K - 347S 51495 Cellarclecti

r~

ST1771 142 1965 8 K 39 60167 CeDordeck failme

STISIH 137 1964 8 K 395 608S4 CeDardeck Failme

K ST130A 140 1958 a SS 6095 I lblv Pai1uno - T21ST72 61 1969 4 Kshy S4S94 4974 I lhlv Pai1uno -

- -

JE

INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON

bull WATER DEPTH

bull bull YEAR OF INSTALLATIONDESIGN

bull PLATFORM ORIENTATION

bull PLATFORM LATITUDE AND LONGITUDE

bull KEY SKETCHES OF PLATFORM STRUCTURAL FRAMING

Number of Legs Plies Leg Batter Number of Conductors bull Deck Elevations (Including sub-cellar deck If any) Jacket Framing Elevations Brace type In Vertical Frames Sizes of Primary Members (leg pile vertical frames braces primary horizontals) Joint Details (can gap etc) LegPiie Annulus Grouting Status

bull middotPOST ANDREW INSPECTION REPORT Significant Wave Height Hs (ft)

bull INFORMATION ON DAMAGES PRIOR TO ANDREW

bull RESULTS FROM IN-HOUSE STUDIES IF AVAILABLE

bull ~RIGINAL DESIGN CRITERIA INFORMATION IF AVAILABLE

bull SOIL SHEAR STRENGTH PROFILE IF AVAILABLE

Platforms Identified for Evaluating Therr Likelihood of Falling Under Unexpected Survival Category

OPlRATOR

middotADA

BLOClt

llmucnJllE

NAME

YEAll

INSTAUJD

WA1ER

DBPTll

DISTANCE

JlllOM SHOllE

(MILU)

Bl

(M)

AMOCO 111113

STl11

WDI075

A

bull D

St

17

120

1n

44

20

10 I

CllllVllQfl STlllO c 61 110 21 11 STIUO D Q 1110 21 11

STl134 p Q 137 29 11

ST0151 G 61 137 11

ST0151 I 3 121 11

ST0151 J Q 140 11

STl151 It 137 11

ST0151 PJl001 Q UT 11

ST0151 PROD-2 Q UT 11

ST017 A 3 140 35 11

WDIUT c 65 214 34 10

llXXON STI055 II 51 6T 14 I

STl15 A t3 34

WDI073

WD0073

Amiddotshy Q

1111 1111

22

22

I

I

WDI073 c 1n 22 I

- WDOCl93-shy II

65 1110 2T A 3 194 31 MOBU 111121

SS IOl7

A B

501

50

50

31

T

I

T 6

MtlllPllY SS114 B T 50 T T

PlllllIPS SS0149 A 5 55 33 T

SAMllDAN STln A 107 T T

SBllLL WDl103 A 65 223 2T WDl103 B 65 221 2T WD0104 c 65 221 2T WD0104 D 6T 20 2T

UNOCAL SS020I

SS 020I SSl20I

SS 0209

SS0209

II p

PURll

A

B

3

n 61

103

97

97

95

100

34

34

34

34 34

bullbull bullbullbull

- SS0209 D u 95 34 I

- SS 1209 SSl253

SS0253

G A

c

6T

2

100

165

1T5

34

54

54

bullbull I

I I

I I

SOUTH MARCH

~LAND

N

SOUTH PaTOI ENE ISWltD AREA

MMS Defined Limits ror Offshore Structures Requiring Inspections Following Andrew

t---Jbulln3 T25bull I bullTtl

bull1$1Htl1Abull bull1SIK1779bull

~

middot-DaTA AREA~T

CRANO rsLE

AREA bull1300

bullll4W

=shy

bullPath of Hurricane Andrew with Platforms Used In Calibration

Likelihood Function vs b All Success Case STISI K STIJOQ ST134W WD90A MCJll MCJ97

I 0000

0 9000

0 8000

07000

-Bo6000

tiiii 0 5000 -

04000

0 3000

0 2000

01000

0 0000

0 05 25

ri1rt middot-middot-middot-middot-middot-

I 5 2

b

---11 middot151K(blsu~ccsI

- bull - bull - fk 130Q(bfU~lCSS)

lie bull134W(blsuccc]

lk-WD90A[blsultCS]

- - - lk middotMC397(bfUtlCS]

--o-- lk-MCJll(bbuctcsJ

--- ll (bl 6 out of 6 Mlttcs case]

_

( Likelihood Functions middotSuccess Cases - p - 1 c I

CATEGORIZATION OF ADDITIONAL PLATFORMS

I I I

bull CATEGORIES

Unexpected Survivals

Expected Survivals

bull APPROACH

Identify Corresponding Platform(s) Evaluated In the JIP

Determine Max Wave Height at Platform Location

Evaluate Whether Wave Hit the Deck

middot - Evaluate Likelihood of R and S to Differ from the Base Cases

Review Joint Information

Assess Special Features

bull IDENTIFY NUMBER OF PLATFORMS IN EACH CATEGORY

I

UPDATING OF CALIBRATION RESULTS

bull INCLUDE EFFECT OF ADDITIONAL PLATFORMS ON bullBbull

Consider Increased 1Uncertalntles In R and S

Develop bullNeW Ukellhood Functions for Different Categories

middotLet bullxbull Number of Ukely Unexpected Survival Cases and y Number of Ukely Expected Survival Cases

Determine Combined Likelihood Function lk (b Iadditional platforms) middot

bull REVISED LIKELIHOOD FUNCTION

lk(b I all cases) = lk(b I 9 platforms) middot x lk(b Ix platforms) x lk(b I y platforms)

bull ESTABLISH REVISED BIAS FACTOR B

f8 (b) u f8 (b) lk (b I all cases)

II

t t ~

shy

APPLICATIONS OF BIAS FACTORS

ii APPLICABLE TO PARTICULAR LOAD AND RESISTANCE RECIPE

bull USE BY THE API COMMITTEES AND MMS

Criteria Revisions

Recommended Design Guidelines Revisions bull

bullbull REASSESSMENT STUDIES AND DECISION MAKING

middot

bull RISK ASSESSMENT STUDIES BY THE OPERATORS

Individual Platforms

Fleet of Platforms

middot IMR Program Implementation

bullbull NOT TO BE USED IN DESIGN

I I fl J

EXAMPLE USE OF BIAS FACTORS

bull PHASE I EXAMPLE APPLICATION

bull EVALUATION OF ANNUAL PROBABILITY OF FAILURE

Establish Annual Maximum Seastate Data

Determine Coefficients to Define Base Shear

Determine Ultimate Capacity of Platform

Determine Probablllty of Failure Versus bullbbull

-bull DETERMINE UPDATED PROBABILITY OF FAILURE

Pf - P [S gt R]

Pf I P [~ gt Rb I b] f 8 (b) db

Pr - J P [S gt Rb I b] f bull 8 (b) db

bull TOTAL PROBABILITIES OF FAILURE

Total Pf - 1 bull Il [1 middotPf] - l Pf bull

Total Pf - 1 bull Il [1middot Pf] - l Pfmiddot

Total Pr = 1 - n [1middot PfbullJ = l Pfshy

I I I

EXAMPLE APPLICATION bull PHASE I

PLATFORM ST 151 K

bull GULF OF MEXICO

bull WATER DEPTH = 137 FT

bull ANNUAL MAXIMUM HS MEDIAN 175 ft COV 031

bull ASSUMED STORM DURATION 3 hoursmiddot

bull ASSUMED NUMBER OF WAVES 800

bull MEDIAN HS FOR STORM HOURS 1 2 AND 3

bull MEDIAN CURRENT VALUE bull

bull DETERMINE ULTIMATE CAPACITY FOR THREE DIRECTIONS

bull DETERMINE ANNUAL FAILURE PROBABILITIES

Broadsidemiddot Diagonal End-on

0037 0025 0007

0071 0044 0018

0022 0016 0004

DETERMINE TOTAL ANNUAL PROBABILITY OF FAILURE

Pf - 007

Pf - 013

Pr - 004

bull

EXAMPLE USE OF BIAS FACTORS - PHASE II middot

bull DETERMINE PREDICTIVE PROBABILITY OF FAILURE

Case - 1

Pf _ S S Pf (b b2) fa1a2 (b b2) db db2

g bull -1 (b1 R I S) - 1

92 - (b2 R2 I S) - 1

Case -2

Pf s ~ [bJ f11 (b) db

Pf [bJ p [g lt 0)

g - bull 1 (b1 R1 I S) - 1

1

bull1) l

I

BENEFITS OF ANDREW JIP bull PHASE II

lt

bull IMPROVED LOAD AND RESISTANCE RECIPE

bull IMPROVED CALIBRATION PROCEDURE

bull MULTIPLE BIAS FACTORS

bull PROCEDURE TO INCLUDE EFFECT OF OTHER PLATFORMS

bull PROCEDURE TO USE MULTIPLE BIAS FACTORS

  • Structure Bookmarks
    • ANDREW PHASE II JOINT1 INDUSTRY PROJECT
    • PROGRESS MEETING NUMBER 1
    • ANDREW PHASE II JOINT INDUSTRY PROJECT PROGRESS MEETING NUMBER 1 NOVEMBER 16 AGENDA
    • ANDREW PHASE II UST OF PARTICIPANTS
    • ANDREW PHASE I OBJECTIVES AND RESULTS
    • ANDREW PHASE II PROJECT OBJECTIVES
    • SUBSEQUENT PROJECTS
    • ANDREW PHASE II PROJECT TEAM
    • ANDREW PHASE II UPDATE MODELING middotAND ANALYSIS RECIPE
    • -ANDREW PHASE II CONDUCTOR RESISTANCE
    • ANDREW PHASE Il WA VE FORCE PROFILE
    • ANDREW PHASE II WA VE FORCE PROFILEmiddot
    • ANDREW PHASE II BRACE MODELING
    • ANDREW PHASE II FOUNDATIONS
    • ANDREW PHASE II middot middot JOINT MODELING
    • ANDREW PHASE Il JOINT MODELING OPTIONS DISCUSSION
    • ANDREW PHASE II JOINT MODEJJNG OPTIONS ILLUSTRATIONS
    • ANDREW PHASE II JOINT MODELING PHASE II METHOD
    • ANDREW PHASE II ANDREW STORM CONDmONS UPDATED
    • Platform Selection bullbullbull1111111111111
    • BasisCriteria for Selection bullbullbull1111111111111
    • CALIBRATION
    • WHERE TO INTRODUCE BIAS FACTOR (B
    • PHASE I -KEY CONCLUSIONS
    • bull CALIBRATION ISSUES PHASE II
    • PHASE II WORK SCOPE (CALIBRATION)
    • SOFTWARE
    • SENSITIVITY STUDIES POSTERIOR DISTRIBUTION OF B
    • WEIGHTING APPROACH bull
    • INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON
    • CATEGORIZATION OF ADDITIONAL PLATFORMS
    • UPDATING OF CALIBRATION RESULTS
    • APPLICATIONS OF BIAS FACTORS
    • EXAMPLE USE OF BIAS FACTORS
    • EXAMPLE APPLICATION bull PHASE I PLATFORM ST 151 K
    • EXAMPLE USE OF BIAS FACTORS -PHASE II middot
    • BENEFITS OF ANDREW JIP bull PHASE II
Page 64:  · . ' ANDREW PHASE II PROJECT OBJECTIVES . ' • ·UPDATE MODELING AND ANALYSIS RECIPE· JOINTS ' ' ·BRACES WAVE PROFILE FOUNDATIONS . ' ' CONDUCTORS· • ANALVZE PLATFORMS USING

-I

Prior and Posterior Distributions of B Variation In COV of PriorofB fromOl to04 -- -

4500

-4000

-

- -

3500 --middot - bull bull - bull bull Prior-r ICOV =02

3000

- - - - Prior-r ICOV =03 -

bull -~ e 2soo -

bull bull bull bull bull bull bull Pnor-r ICOV =o 4t frJ ~ 2000 Posterior rlCOV =0 2

bull bull Postenor rlCOV =0 3 bull bull 1500 bull middot bull - shy -middot- Posterior rlCOV =04

middot~ I 000 - bull

-middotmiddot ~ ~middotmiddot M -~

- middot 0 500

wbullbullbull _ I middot~ middotmiddotmiddotmiddotmiddot middotmiddotmiddotmiddotmiddot - -~ ~ __ middot_ _ bull -shy0 000

0 000 0200 0 400 0 600 0 800 IOOO 1200 1400 1600 1800 2000 -

- - b

-

=shy

Posterior Distribution of Bias Factor (B) bull Effect of Variation In COV of Prior of B

bull

bullPrior and Posterior Distributions of B Evaluating Effect of Removal of One Platform from Each Category from the Base Case of 13 Platforms

400

~ ~ 350

j IA i

3 00 bull I bull I ---Prior-r I bull l

250 1 I 1

I bull bull ---Posterior r lall 13 casesshy- ~E 2 00 bull bull bull bull bull bull bull Posterior r without STISIK

- ~ 1 bull - bull I - bull bull - bull bull Posterior r Iwithout T2I I 50 I

J-tt~i - bull - bull - Posterior r Iwithout T2S

tOO [-

I I ~ ~ I 0 50

0 00

___ r J i

l~T

~--

0 00 0 20 0 40 0 60 080 I 00 120 140 160 110 2 00

b

zshy

Posterior Distribution of Bias Factor (8) bull Effect of Removal of middotPlatform from Any Category

bullPrior and Posterior Distributions or B Evaluating Effect or Removal or 3 Damage Cases from the Base Case or 13 Platrorms

4

35

3

25

e L 2 ~ bulle shy

I 15

bull I

05

0

0 02 04 06 I 2 I 4 16 IS 2

I I

I

I

I Ibull

J -tmiddot L--

bullI

~

I

fl

II ~

I I v I

middot bull bull

bull L

~ ll- I-~lt

OS I

b

I

bull

1middot bullbullbullbull

~ __

bull

- - - - Prior-r

- - - - Postenor flall failure cases

Postenor flall success cases

--- Posterior flall 13 cases

---Posterior fI success failure cases only

IPosterior or B Mean =t11 COV =0121

-

_I

Posterir Distribution or Bias Factor (B) bull Effect or Removal or 3-Damage Cases

-

bullPrior and Posterior Distributions or B Evaluating Effect or Additional 13 Similar Observations as

Used in Andrew JIP The POsterior or B from Andrew-JIP Considered as New Prior or B

- 500

450

400

350

bull bull

bull bull bull bull J bull

bull bull f ~ I

I ~ Ibull bull

bullbull bull

l bull I -bullbull bull bull

I

I bull bull

bull I I~~bull bull bull

bull ~

I bull bull

~ ~_ ~~~

- - New Prior r bull

Posterior 114 new failure cases ~ 300- bull bull bull bull bull bull bull Posterior f 13 new damage casesl 150 ~

- bull bull - bull bull Posterior 116 new success cases i 2 00

Postenor 1113 new success failure -150 damage cases

I 00

0 50 -

0 00 0 00 0 20 0 40 0 60 0 80 I 00 120 I 40 160 110 2 00

Posterior ol B Mean= 120 COV= 0071 b

shy

shy

_rshy

Posterior Distribution of Bias Factor (B) bull Effect of Additional 13-Similar Observations

1

1 I

WEIGHTING APPROACH bull

bull ASSESSMENT OF PLATFORMS DATA

Select Platforms Most Mappable to the Analyzed Platforms

Unlikely SurvivalDamage Cases Identification

Based on Cut-Off Criteria Agemiddot Hs etc

bull INFORMATION REQUIRED FOR COMPARISON

bull CATEGORIZATION OF ADDITIONALmiddot PLATFORMS

bull UPDATING OF CALIBRATION RESULTS

--

OVERVIEW OF PLATFORMS SURVIVEDDAMAGED -~

Operator -

- Platforms

Survived CS)

Platforms

Damaged CD)

Platforms Used In

Phase IJIP -

JI~ ~Bill~IRBDll -

Amoco -

Chevron -

Exxon

Mob II

Phllllps

Shell

Unocal -

30

143

72 22

7

39 25shy -

1

12

bull

1 bull

1

1(S)+1 (D)

3 (S) +3(D)

2 (S)

bull shy

bull

bull

bull

Total Partlclnants Platforms 338 - 15 10C6S+4D) -

Other Operators

310

13

3 (D)

Total Platforms 648 28 13C6S+7D) -

Summary of Andrew JIP bull Phase I Platforms

Plairshy w v- N-ber Lea-le 1 shy car Hind_ o-w N-bull Deplh fl fl IJpe la bull n Yeadon Jadtet lbelledtf liefalit -- _ -OD

IL H - -shy _ICllshy IL IL - - rQClto ~ _

K STUIK 137 1963 8 365 608S4 Cellardeck

-middot STl-shy 170 1964 4 -~ K 39 62269 Cellardeck - --ST134W 137 1911 4 43 608S4 Smvlnlshy- _ - -

WD90A 184 1964 8 K 267Sbull34 83 SOS31 SUboftllar deck SmvMI

MC311 343 1978 8 x S7 62 - MC397 468 1991 4 x shy S33bull6SO 65911 -shySmeSmvlvol -- gt -

r-- T23ST52 63 1969 4 ton- shyKshy 52594 S03 -T25SSl39 62 1969 4 Kshy SOS94 S063 - middot ton-shy__ -

bullbull a ton--shyST161A 118 1964 8 K - 347S 51495 Cellarclecti

r~

ST1771 142 1965 8 K 39 60167 CeDordeck failme

STISIH 137 1964 8 K 395 608S4 CeDardeck Failme

K ST130A 140 1958 a SS 6095 I lblv Pai1uno - T21ST72 61 1969 4 Kshy S4S94 4974 I lhlv Pai1uno -

- -

JE

INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON

bull WATER DEPTH

bull bull YEAR OF INSTALLATIONDESIGN

bull PLATFORM ORIENTATION

bull PLATFORM LATITUDE AND LONGITUDE

bull KEY SKETCHES OF PLATFORM STRUCTURAL FRAMING

Number of Legs Plies Leg Batter Number of Conductors bull Deck Elevations (Including sub-cellar deck If any) Jacket Framing Elevations Brace type In Vertical Frames Sizes of Primary Members (leg pile vertical frames braces primary horizontals) Joint Details (can gap etc) LegPiie Annulus Grouting Status

bull middotPOST ANDREW INSPECTION REPORT Significant Wave Height Hs (ft)

bull INFORMATION ON DAMAGES PRIOR TO ANDREW

bull RESULTS FROM IN-HOUSE STUDIES IF AVAILABLE

bull ~RIGINAL DESIGN CRITERIA INFORMATION IF AVAILABLE

bull SOIL SHEAR STRENGTH PROFILE IF AVAILABLE

Platforms Identified for Evaluating Therr Likelihood of Falling Under Unexpected Survival Category

OPlRATOR

middotADA

BLOClt

llmucnJllE

NAME

YEAll

INSTAUJD

WA1ER

DBPTll

DISTANCE

JlllOM SHOllE

(MILU)

Bl

(M)

AMOCO 111113

STl11

WDI075

A

bull D

St

17

120

1n

44

20

10 I

CllllVllQfl STlllO c 61 110 21 11 STIUO D Q 1110 21 11

STl134 p Q 137 29 11

ST0151 G 61 137 11

ST0151 I 3 121 11

ST0151 J Q 140 11

STl151 It 137 11

ST0151 PJl001 Q UT 11

ST0151 PROD-2 Q UT 11

ST017 A 3 140 35 11

WDIUT c 65 214 34 10

llXXON STI055 II 51 6T 14 I

STl15 A t3 34

WDI073

WD0073

Amiddotshy Q

1111 1111

22

22

I

I

WDI073 c 1n 22 I

- WDOCl93-shy II

65 1110 2T A 3 194 31 MOBU 111121

SS IOl7

A B

501

50

50

31

T

I

T 6

MtlllPllY SS114 B T 50 T T

PlllllIPS SS0149 A 5 55 33 T

SAMllDAN STln A 107 T T

SBllLL WDl103 A 65 223 2T WDl103 B 65 221 2T WD0104 c 65 221 2T WD0104 D 6T 20 2T

UNOCAL SS020I

SS 020I SSl20I

SS 0209

SS0209

II p

PURll

A

B

3

n 61

103

97

97

95

100

34

34

34

34 34

bullbull bullbullbull

- SS0209 D u 95 34 I

- SS 1209 SSl253

SS0253

G A

c

6T

2

100

165

1T5

34

54

54

bullbull I

I I

I I

SOUTH MARCH

~LAND

N

SOUTH PaTOI ENE ISWltD AREA

MMS Defined Limits ror Offshore Structures Requiring Inspections Following Andrew

t---Jbulln3 T25bull I bullTtl

bull1$1Htl1Abull bull1SIK1779bull

~

middot-DaTA AREA~T

CRANO rsLE

AREA bull1300

bullll4W

=shy

bullPath of Hurricane Andrew with Platforms Used In Calibration

Likelihood Function vs b All Success Case STISI K STIJOQ ST134W WD90A MCJll MCJ97

I 0000

0 9000

0 8000

07000

-Bo6000

tiiii 0 5000 -

04000

0 3000

0 2000

01000

0 0000

0 05 25

ri1rt middot-middot-middot-middot-middot-

I 5 2

b

---11 middot151K(blsu~ccsI

- bull - bull - fk 130Q(bfU~lCSS)

lie bull134W(blsuccc]

lk-WD90A[blsultCS]

- - - lk middotMC397(bfUtlCS]

--o-- lk-MCJll(bbuctcsJ

--- ll (bl 6 out of 6 Mlttcs case]

_

( Likelihood Functions middotSuccess Cases - p - 1 c I

CATEGORIZATION OF ADDITIONAL PLATFORMS

I I I

bull CATEGORIES

Unexpected Survivals

Expected Survivals

bull APPROACH

Identify Corresponding Platform(s) Evaluated In the JIP

Determine Max Wave Height at Platform Location

Evaluate Whether Wave Hit the Deck

middot - Evaluate Likelihood of R and S to Differ from the Base Cases

Review Joint Information

Assess Special Features

bull IDENTIFY NUMBER OF PLATFORMS IN EACH CATEGORY

I

UPDATING OF CALIBRATION RESULTS

bull INCLUDE EFFECT OF ADDITIONAL PLATFORMS ON bullBbull

Consider Increased 1Uncertalntles In R and S

Develop bullNeW Ukellhood Functions for Different Categories

middotLet bullxbull Number of Ukely Unexpected Survival Cases and y Number of Ukely Expected Survival Cases

Determine Combined Likelihood Function lk (b Iadditional platforms) middot

bull REVISED LIKELIHOOD FUNCTION

lk(b I all cases) = lk(b I 9 platforms) middot x lk(b Ix platforms) x lk(b I y platforms)

bull ESTABLISH REVISED BIAS FACTOR B

f8 (b) u f8 (b) lk (b I all cases)

II

t t ~

shy

APPLICATIONS OF BIAS FACTORS

ii APPLICABLE TO PARTICULAR LOAD AND RESISTANCE RECIPE

bull USE BY THE API COMMITTEES AND MMS

Criteria Revisions

Recommended Design Guidelines Revisions bull

bullbull REASSESSMENT STUDIES AND DECISION MAKING

middot

bull RISK ASSESSMENT STUDIES BY THE OPERATORS

Individual Platforms

Fleet of Platforms

middot IMR Program Implementation

bullbull NOT TO BE USED IN DESIGN

I I fl J

EXAMPLE USE OF BIAS FACTORS

bull PHASE I EXAMPLE APPLICATION

bull EVALUATION OF ANNUAL PROBABILITY OF FAILURE

Establish Annual Maximum Seastate Data

Determine Coefficients to Define Base Shear

Determine Ultimate Capacity of Platform

Determine Probablllty of Failure Versus bullbbull

-bull DETERMINE UPDATED PROBABILITY OF FAILURE

Pf - P [S gt R]

Pf I P [~ gt Rb I b] f 8 (b) db

Pr - J P [S gt Rb I b] f bull 8 (b) db

bull TOTAL PROBABILITIES OF FAILURE

Total Pf - 1 bull Il [1 middotPf] - l Pf bull

Total Pf - 1 bull Il [1middot Pf] - l Pfmiddot

Total Pr = 1 - n [1middot PfbullJ = l Pfshy

I I I

EXAMPLE APPLICATION bull PHASE I

PLATFORM ST 151 K

bull GULF OF MEXICO

bull WATER DEPTH = 137 FT

bull ANNUAL MAXIMUM HS MEDIAN 175 ft COV 031

bull ASSUMED STORM DURATION 3 hoursmiddot

bull ASSUMED NUMBER OF WAVES 800

bull MEDIAN HS FOR STORM HOURS 1 2 AND 3

bull MEDIAN CURRENT VALUE bull

bull DETERMINE ULTIMATE CAPACITY FOR THREE DIRECTIONS

bull DETERMINE ANNUAL FAILURE PROBABILITIES

Broadsidemiddot Diagonal End-on

0037 0025 0007

0071 0044 0018

0022 0016 0004

DETERMINE TOTAL ANNUAL PROBABILITY OF FAILURE

Pf - 007

Pf - 013

Pr - 004

bull

EXAMPLE USE OF BIAS FACTORS - PHASE II middot

bull DETERMINE PREDICTIVE PROBABILITY OF FAILURE

Case - 1

Pf _ S S Pf (b b2) fa1a2 (b b2) db db2

g bull -1 (b1 R I S) - 1

92 - (b2 R2 I S) - 1

Case -2

Pf s ~ [bJ f11 (b) db

Pf [bJ p [g lt 0)

g - bull 1 (b1 R1 I S) - 1

1

bull1) l

I

BENEFITS OF ANDREW JIP bull PHASE II

lt

bull IMPROVED LOAD AND RESISTANCE RECIPE

bull IMPROVED CALIBRATION PROCEDURE

bull MULTIPLE BIAS FACTORS

bull PROCEDURE TO INCLUDE EFFECT OF OTHER PLATFORMS

bull PROCEDURE TO USE MULTIPLE BIAS FACTORS

  • Structure Bookmarks
    • ANDREW PHASE II JOINT1 INDUSTRY PROJECT
    • PROGRESS MEETING NUMBER 1
    • ANDREW PHASE II JOINT INDUSTRY PROJECT PROGRESS MEETING NUMBER 1 NOVEMBER 16 AGENDA
    • ANDREW PHASE II UST OF PARTICIPANTS
    • ANDREW PHASE I OBJECTIVES AND RESULTS
    • ANDREW PHASE II PROJECT OBJECTIVES
    • SUBSEQUENT PROJECTS
    • ANDREW PHASE II PROJECT TEAM
    • ANDREW PHASE II UPDATE MODELING middotAND ANALYSIS RECIPE
    • -ANDREW PHASE II CONDUCTOR RESISTANCE
    • ANDREW PHASE Il WA VE FORCE PROFILE
    • ANDREW PHASE II WA VE FORCE PROFILEmiddot
    • ANDREW PHASE II BRACE MODELING
    • ANDREW PHASE II FOUNDATIONS
    • ANDREW PHASE II middot middot JOINT MODELING
    • ANDREW PHASE Il JOINT MODELING OPTIONS DISCUSSION
    • ANDREW PHASE II JOINT MODEJJNG OPTIONS ILLUSTRATIONS
    • ANDREW PHASE II JOINT MODELING PHASE II METHOD
    • ANDREW PHASE II ANDREW STORM CONDmONS UPDATED
    • Platform Selection bullbullbull1111111111111
    • BasisCriteria for Selection bullbullbull1111111111111
    • CALIBRATION
    • WHERE TO INTRODUCE BIAS FACTOR (B
    • PHASE I -KEY CONCLUSIONS
    • bull CALIBRATION ISSUES PHASE II
    • PHASE II WORK SCOPE (CALIBRATION)
    • SOFTWARE
    • SENSITIVITY STUDIES POSTERIOR DISTRIBUTION OF B
    • WEIGHTING APPROACH bull
    • INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON
    • CATEGORIZATION OF ADDITIONAL PLATFORMS
    • UPDATING OF CALIBRATION RESULTS
    • APPLICATIONS OF BIAS FACTORS
    • EXAMPLE USE OF BIAS FACTORS
    • EXAMPLE APPLICATION bull PHASE I PLATFORM ST 151 K
    • EXAMPLE USE OF BIAS FACTORS -PHASE II middot
    • BENEFITS OF ANDREW JIP bull PHASE II
Page 65:  · . ' ANDREW PHASE II PROJECT OBJECTIVES . ' • ·UPDATE MODELING AND ANALYSIS RECIPE· JOINTS ' ' ·BRACES WAVE PROFILE FOUNDATIONS . ' ' CONDUCTORS· • ANALVZE PLATFORMS USING

bull

bullPrior and Posterior Distributions of B Evaluating Effect of Removal of One Platform from Each Category from the Base Case of 13 Platforms

400

~ ~ 350

j IA i

3 00 bull I bull I ---Prior-r I bull l

250 1 I 1

I bull bull ---Posterior r lall 13 casesshy- ~E 2 00 bull bull bull bull bull bull bull Posterior r without STISIK

- ~ 1 bull - bull I - bull bull - bull bull Posterior r Iwithout T2I I 50 I

J-tt~i - bull - bull - Posterior r Iwithout T2S

tOO [-

I I ~ ~ I 0 50

0 00

___ r J i

l~T

~--

0 00 0 20 0 40 0 60 080 I 00 120 140 160 110 2 00

b

zshy

Posterior Distribution of Bias Factor (8) bull Effect of Removal of middotPlatform from Any Category

bullPrior and Posterior Distributions or B Evaluating Effect or Removal or 3 Damage Cases from the Base Case or 13 Platrorms

4

35

3

25

e L 2 ~ bulle shy

I 15

bull I

05

0

0 02 04 06 I 2 I 4 16 IS 2

I I

I

I

I Ibull

J -tmiddot L--

bullI

~

I

fl

II ~

I I v I

middot bull bull

bull L

~ ll- I-~lt

OS I

b

I

bull

1middot bullbullbullbull

~ __

bull

- - - - Prior-r

- - - - Postenor flall failure cases

Postenor flall success cases

--- Posterior flall 13 cases

---Posterior fI success failure cases only

IPosterior or B Mean =t11 COV =0121

-

_I

Posterir Distribution or Bias Factor (B) bull Effect or Removal or 3-Damage Cases

-

bullPrior and Posterior Distributions or B Evaluating Effect or Additional 13 Similar Observations as

Used in Andrew JIP The POsterior or B from Andrew-JIP Considered as New Prior or B

- 500

450

400

350

bull bull

bull bull bull bull J bull

bull bull f ~ I

I ~ Ibull bull

bullbull bull

l bull I -bullbull bull bull

I

I bull bull

bull I I~~bull bull bull

bull ~

I bull bull

~ ~_ ~~~

- - New Prior r bull

Posterior 114 new failure cases ~ 300- bull bull bull bull bull bull bull Posterior f 13 new damage casesl 150 ~

- bull bull - bull bull Posterior 116 new success cases i 2 00

Postenor 1113 new success failure -150 damage cases

I 00

0 50 -

0 00 0 00 0 20 0 40 0 60 0 80 I 00 120 I 40 160 110 2 00

Posterior ol B Mean= 120 COV= 0071 b

shy

shy

_rshy

Posterior Distribution of Bias Factor (B) bull Effect of Additional 13-Similar Observations

1

1 I

WEIGHTING APPROACH bull

bull ASSESSMENT OF PLATFORMS DATA

Select Platforms Most Mappable to the Analyzed Platforms

Unlikely SurvivalDamage Cases Identification

Based on Cut-Off Criteria Agemiddot Hs etc

bull INFORMATION REQUIRED FOR COMPARISON

bull CATEGORIZATION OF ADDITIONALmiddot PLATFORMS

bull UPDATING OF CALIBRATION RESULTS

--

OVERVIEW OF PLATFORMS SURVIVEDDAMAGED -~

Operator -

- Platforms

Survived CS)

Platforms

Damaged CD)

Platforms Used In

Phase IJIP -

JI~ ~Bill~IRBDll -

Amoco -

Chevron -

Exxon

Mob II

Phllllps

Shell

Unocal -

30

143

72 22

7

39 25shy -

1

12

bull

1 bull

1

1(S)+1 (D)

3 (S) +3(D)

2 (S)

bull shy

bull

bull

bull

Total Partlclnants Platforms 338 - 15 10C6S+4D) -

Other Operators

310

13

3 (D)

Total Platforms 648 28 13C6S+7D) -

Summary of Andrew JIP bull Phase I Platforms

Plairshy w v- N-ber Lea-le 1 shy car Hind_ o-w N-bull Deplh fl fl IJpe la bull n Yeadon Jadtet lbelledtf liefalit -- _ -OD

IL H - -shy _ICllshy IL IL - - rQClto ~ _

K STUIK 137 1963 8 365 608S4 Cellardeck

-middot STl-shy 170 1964 4 -~ K 39 62269 Cellardeck - --ST134W 137 1911 4 43 608S4 Smvlnlshy- _ - -

WD90A 184 1964 8 K 267Sbull34 83 SOS31 SUboftllar deck SmvMI

MC311 343 1978 8 x S7 62 - MC397 468 1991 4 x shy S33bull6SO 65911 -shySmeSmvlvol -- gt -

r-- T23ST52 63 1969 4 ton- shyKshy 52594 S03 -T25SSl39 62 1969 4 Kshy SOS94 S063 - middot ton-shy__ -

bullbull a ton--shyST161A 118 1964 8 K - 347S 51495 Cellarclecti

r~

ST1771 142 1965 8 K 39 60167 CeDordeck failme

STISIH 137 1964 8 K 395 608S4 CeDardeck Failme

K ST130A 140 1958 a SS 6095 I lblv Pai1uno - T21ST72 61 1969 4 Kshy S4S94 4974 I lhlv Pai1uno -

- -

JE

INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON

bull WATER DEPTH

bull bull YEAR OF INSTALLATIONDESIGN

bull PLATFORM ORIENTATION

bull PLATFORM LATITUDE AND LONGITUDE

bull KEY SKETCHES OF PLATFORM STRUCTURAL FRAMING

Number of Legs Plies Leg Batter Number of Conductors bull Deck Elevations (Including sub-cellar deck If any) Jacket Framing Elevations Brace type In Vertical Frames Sizes of Primary Members (leg pile vertical frames braces primary horizontals) Joint Details (can gap etc) LegPiie Annulus Grouting Status

bull middotPOST ANDREW INSPECTION REPORT Significant Wave Height Hs (ft)

bull INFORMATION ON DAMAGES PRIOR TO ANDREW

bull RESULTS FROM IN-HOUSE STUDIES IF AVAILABLE

bull ~RIGINAL DESIGN CRITERIA INFORMATION IF AVAILABLE

bull SOIL SHEAR STRENGTH PROFILE IF AVAILABLE

Platforms Identified for Evaluating Therr Likelihood of Falling Under Unexpected Survival Category

OPlRATOR

middotADA

BLOClt

llmucnJllE

NAME

YEAll

INSTAUJD

WA1ER

DBPTll

DISTANCE

JlllOM SHOllE

(MILU)

Bl

(M)

AMOCO 111113

STl11

WDI075

A

bull D

St

17

120

1n

44

20

10 I

CllllVllQfl STlllO c 61 110 21 11 STIUO D Q 1110 21 11

STl134 p Q 137 29 11

ST0151 G 61 137 11

ST0151 I 3 121 11

ST0151 J Q 140 11

STl151 It 137 11

ST0151 PJl001 Q UT 11

ST0151 PROD-2 Q UT 11

ST017 A 3 140 35 11

WDIUT c 65 214 34 10

llXXON STI055 II 51 6T 14 I

STl15 A t3 34

WDI073

WD0073

Amiddotshy Q

1111 1111

22

22

I

I

WDI073 c 1n 22 I

- WDOCl93-shy II

65 1110 2T A 3 194 31 MOBU 111121

SS IOl7

A B

501

50

50

31

T

I

T 6

MtlllPllY SS114 B T 50 T T

PlllllIPS SS0149 A 5 55 33 T

SAMllDAN STln A 107 T T

SBllLL WDl103 A 65 223 2T WDl103 B 65 221 2T WD0104 c 65 221 2T WD0104 D 6T 20 2T

UNOCAL SS020I

SS 020I SSl20I

SS 0209

SS0209

II p

PURll

A

B

3

n 61

103

97

97

95

100

34

34

34

34 34

bullbull bullbullbull

- SS0209 D u 95 34 I

- SS 1209 SSl253

SS0253

G A

c

6T

2

100

165

1T5

34

54

54

bullbull I

I I

I I

SOUTH MARCH

~LAND

N

SOUTH PaTOI ENE ISWltD AREA

MMS Defined Limits ror Offshore Structures Requiring Inspections Following Andrew

t---Jbulln3 T25bull I bullTtl

bull1$1Htl1Abull bull1SIK1779bull

~

middot-DaTA AREA~T

CRANO rsLE

AREA bull1300

bullll4W

=shy

bullPath of Hurricane Andrew with Platforms Used In Calibration

Likelihood Function vs b All Success Case STISI K STIJOQ ST134W WD90A MCJll MCJ97

I 0000

0 9000

0 8000

07000

-Bo6000

tiiii 0 5000 -

04000

0 3000

0 2000

01000

0 0000

0 05 25

ri1rt middot-middot-middot-middot-middot-

I 5 2

b

---11 middot151K(blsu~ccsI

- bull - bull - fk 130Q(bfU~lCSS)

lie bull134W(blsuccc]

lk-WD90A[blsultCS]

- - - lk middotMC397(bfUtlCS]

--o-- lk-MCJll(bbuctcsJ

--- ll (bl 6 out of 6 Mlttcs case]

_

( Likelihood Functions middotSuccess Cases - p - 1 c I

CATEGORIZATION OF ADDITIONAL PLATFORMS

I I I

bull CATEGORIES

Unexpected Survivals

Expected Survivals

bull APPROACH

Identify Corresponding Platform(s) Evaluated In the JIP

Determine Max Wave Height at Platform Location

Evaluate Whether Wave Hit the Deck

middot - Evaluate Likelihood of R and S to Differ from the Base Cases

Review Joint Information

Assess Special Features

bull IDENTIFY NUMBER OF PLATFORMS IN EACH CATEGORY

I

UPDATING OF CALIBRATION RESULTS

bull INCLUDE EFFECT OF ADDITIONAL PLATFORMS ON bullBbull

Consider Increased 1Uncertalntles In R and S

Develop bullNeW Ukellhood Functions for Different Categories

middotLet bullxbull Number of Ukely Unexpected Survival Cases and y Number of Ukely Expected Survival Cases

Determine Combined Likelihood Function lk (b Iadditional platforms) middot

bull REVISED LIKELIHOOD FUNCTION

lk(b I all cases) = lk(b I 9 platforms) middot x lk(b Ix platforms) x lk(b I y platforms)

bull ESTABLISH REVISED BIAS FACTOR B

f8 (b) u f8 (b) lk (b I all cases)

II

t t ~

shy

APPLICATIONS OF BIAS FACTORS

ii APPLICABLE TO PARTICULAR LOAD AND RESISTANCE RECIPE

bull USE BY THE API COMMITTEES AND MMS

Criteria Revisions

Recommended Design Guidelines Revisions bull

bullbull REASSESSMENT STUDIES AND DECISION MAKING

middot

bull RISK ASSESSMENT STUDIES BY THE OPERATORS

Individual Platforms

Fleet of Platforms

middot IMR Program Implementation

bullbull NOT TO BE USED IN DESIGN

I I fl J

EXAMPLE USE OF BIAS FACTORS

bull PHASE I EXAMPLE APPLICATION

bull EVALUATION OF ANNUAL PROBABILITY OF FAILURE

Establish Annual Maximum Seastate Data

Determine Coefficients to Define Base Shear

Determine Ultimate Capacity of Platform

Determine Probablllty of Failure Versus bullbbull

-bull DETERMINE UPDATED PROBABILITY OF FAILURE

Pf - P [S gt R]

Pf I P [~ gt Rb I b] f 8 (b) db

Pr - J P [S gt Rb I b] f bull 8 (b) db

bull TOTAL PROBABILITIES OF FAILURE

Total Pf - 1 bull Il [1 middotPf] - l Pf bull

Total Pf - 1 bull Il [1middot Pf] - l Pfmiddot

Total Pr = 1 - n [1middot PfbullJ = l Pfshy

I I I

EXAMPLE APPLICATION bull PHASE I

PLATFORM ST 151 K

bull GULF OF MEXICO

bull WATER DEPTH = 137 FT

bull ANNUAL MAXIMUM HS MEDIAN 175 ft COV 031

bull ASSUMED STORM DURATION 3 hoursmiddot

bull ASSUMED NUMBER OF WAVES 800

bull MEDIAN HS FOR STORM HOURS 1 2 AND 3

bull MEDIAN CURRENT VALUE bull

bull DETERMINE ULTIMATE CAPACITY FOR THREE DIRECTIONS

bull DETERMINE ANNUAL FAILURE PROBABILITIES

Broadsidemiddot Diagonal End-on

0037 0025 0007

0071 0044 0018

0022 0016 0004

DETERMINE TOTAL ANNUAL PROBABILITY OF FAILURE

Pf - 007

Pf - 013

Pr - 004

bull

EXAMPLE USE OF BIAS FACTORS - PHASE II middot

bull DETERMINE PREDICTIVE PROBABILITY OF FAILURE

Case - 1

Pf _ S S Pf (b b2) fa1a2 (b b2) db db2

g bull -1 (b1 R I S) - 1

92 - (b2 R2 I S) - 1

Case -2

Pf s ~ [bJ f11 (b) db

Pf [bJ p [g lt 0)

g - bull 1 (b1 R1 I S) - 1

1

bull1) l

I

BENEFITS OF ANDREW JIP bull PHASE II

lt

bull IMPROVED LOAD AND RESISTANCE RECIPE

bull IMPROVED CALIBRATION PROCEDURE

bull MULTIPLE BIAS FACTORS

bull PROCEDURE TO INCLUDE EFFECT OF OTHER PLATFORMS

bull PROCEDURE TO USE MULTIPLE BIAS FACTORS

  • Structure Bookmarks
    • ANDREW PHASE II JOINT1 INDUSTRY PROJECT
    • PROGRESS MEETING NUMBER 1
    • ANDREW PHASE II JOINT INDUSTRY PROJECT PROGRESS MEETING NUMBER 1 NOVEMBER 16 AGENDA
    • ANDREW PHASE II UST OF PARTICIPANTS
    • ANDREW PHASE I OBJECTIVES AND RESULTS
    • ANDREW PHASE II PROJECT OBJECTIVES
    • SUBSEQUENT PROJECTS
    • ANDREW PHASE II PROJECT TEAM
    • ANDREW PHASE II UPDATE MODELING middotAND ANALYSIS RECIPE
    • -ANDREW PHASE II CONDUCTOR RESISTANCE
    • ANDREW PHASE Il WA VE FORCE PROFILE
    • ANDREW PHASE II WA VE FORCE PROFILEmiddot
    • ANDREW PHASE II BRACE MODELING
    • ANDREW PHASE II FOUNDATIONS
    • ANDREW PHASE II middot middot JOINT MODELING
    • ANDREW PHASE Il JOINT MODELING OPTIONS DISCUSSION
    • ANDREW PHASE II JOINT MODEJJNG OPTIONS ILLUSTRATIONS
    • ANDREW PHASE II JOINT MODELING PHASE II METHOD
    • ANDREW PHASE II ANDREW STORM CONDmONS UPDATED
    • Platform Selection bullbullbull1111111111111
    • BasisCriteria for Selection bullbullbull1111111111111
    • CALIBRATION
    • WHERE TO INTRODUCE BIAS FACTOR (B
    • PHASE I -KEY CONCLUSIONS
    • bull CALIBRATION ISSUES PHASE II
    • PHASE II WORK SCOPE (CALIBRATION)
    • SOFTWARE
    • SENSITIVITY STUDIES POSTERIOR DISTRIBUTION OF B
    • WEIGHTING APPROACH bull
    • INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON
    • CATEGORIZATION OF ADDITIONAL PLATFORMS
    • UPDATING OF CALIBRATION RESULTS
    • APPLICATIONS OF BIAS FACTORS
    • EXAMPLE USE OF BIAS FACTORS
    • EXAMPLE APPLICATION bull PHASE I PLATFORM ST 151 K
    • EXAMPLE USE OF BIAS FACTORS -PHASE II middot
    • BENEFITS OF ANDREW JIP bull PHASE II
Page 66:  · . ' ANDREW PHASE II PROJECT OBJECTIVES . ' • ·UPDATE MODELING AND ANALYSIS RECIPE· JOINTS ' ' ·BRACES WAVE PROFILE FOUNDATIONS . ' ' CONDUCTORS· • ANALVZE PLATFORMS USING

bullPrior and Posterior Distributions or B Evaluating Effect or Removal or 3 Damage Cases from the Base Case or 13 Platrorms

4

35

3

25

e L 2 ~ bulle shy

I 15

bull I

05

0

0 02 04 06 I 2 I 4 16 IS 2

I I

I

I

I Ibull

J -tmiddot L--

bullI

~

I

fl

II ~

I I v I

middot bull bull

bull L

~ ll- I-~lt

OS I

b

I

bull

1middot bullbullbullbull

~ __

bull

- - - - Prior-r

- - - - Postenor flall failure cases

Postenor flall success cases

--- Posterior flall 13 cases

---Posterior fI success failure cases only

IPosterior or B Mean =t11 COV =0121

-

_I

Posterir Distribution or Bias Factor (B) bull Effect or Removal or 3-Damage Cases

-

bullPrior and Posterior Distributions or B Evaluating Effect or Additional 13 Similar Observations as

Used in Andrew JIP The POsterior or B from Andrew-JIP Considered as New Prior or B

- 500

450

400

350

bull bull

bull bull bull bull J bull

bull bull f ~ I

I ~ Ibull bull

bullbull bull

l bull I -bullbull bull bull

I

I bull bull

bull I I~~bull bull bull

bull ~

I bull bull

~ ~_ ~~~

- - New Prior r bull

Posterior 114 new failure cases ~ 300- bull bull bull bull bull bull bull Posterior f 13 new damage casesl 150 ~

- bull bull - bull bull Posterior 116 new success cases i 2 00

Postenor 1113 new success failure -150 damage cases

I 00

0 50 -

0 00 0 00 0 20 0 40 0 60 0 80 I 00 120 I 40 160 110 2 00

Posterior ol B Mean= 120 COV= 0071 b

shy

shy

_rshy

Posterior Distribution of Bias Factor (B) bull Effect of Additional 13-Similar Observations

1

1 I

WEIGHTING APPROACH bull

bull ASSESSMENT OF PLATFORMS DATA

Select Platforms Most Mappable to the Analyzed Platforms

Unlikely SurvivalDamage Cases Identification

Based on Cut-Off Criteria Agemiddot Hs etc

bull INFORMATION REQUIRED FOR COMPARISON

bull CATEGORIZATION OF ADDITIONALmiddot PLATFORMS

bull UPDATING OF CALIBRATION RESULTS

--

OVERVIEW OF PLATFORMS SURVIVEDDAMAGED -~

Operator -

- Platforms

Survived CS)

Platforms

Damaged CD)

Platforms Used In

Phase IJIP -

JI~ ~Bill~IRBDll -

Amoco -

Chevron -

Exxon

Mob II

Phllllps

Shell

Unocal -

30

143

72 22

7

39 25shy -

1

12

bull

1 bull

1

1(S)+1 (D)

3 (S) +3(D)

2 (S)

bull shy

bull

bull

bull

Total Partlclnants Platforms 338 - 15 10C6S+4D) -

Other Operators

310

13

3 (D)

Total Platforms 648 28 13C6S+7D) -

Summary of Andrew JIP bull Phase I Platforms

Plairshy w v- N-ber Lea-le 1 shy car Hind_ o-w N-bull Deplh fl fl IJpe la bull n Yeadon Jadtet lbelledtf liefalit -- _ -OD

IL H - -shy _ICllshy IL IL - - rQClto ~ _

K STUIK 137 1963 8 365 608S4 Cellardeck

-middot STl-shy 170 1964 4 -~ K 39 62269 Cellardeck - --ST134W 137 1911 4 43 608S4 Smvlnlshy- _ - -

WD90A 184 1964 8 K 267Sbull34 83 SOS31 SUboftllar deck SmvMI

MC311 343 1978 8 x S7 62 - MC397 468 1991 4 x shy S33bull6SO 65911 -shySmeSmvlvol -- gt -

r-- T23ST52 63 1969 4 ton- shyKshy 52594 S03 -T25SSl39 62 1969 4 Kshy SOS94 S063 - middot ton-shy__ -

bullbull a ton--shyST161A 118 1964 8 K - 347S 51495 Cellarclecti

r~

ST1771 142 1965 8 K 39 60167 CeDordeck failme

STISIH 137 1964 8 K 395 608S4 CeDardeck Failme

K ST130A 140 1958 a SS 6095 I lblv Pai1uno - T21ST72 61 1969 4 Kshy S4S94 4974 I lhlv Pai1uno -

- -

JE

INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON

bull WATER DEPTH

bull bull YEAR OF INSTALLATIONDESIGN

bull PLATFORM ORIENTATION

bull PLATFORM LATITUDE AND LONGITUDE

bull KEY SKETCHES OF PLATFORM STRUCTURAL FRAMING

Number of Legs Plies Leg Batter Number of Conductors bull Deck Elevations (Including sub-cellar deck If any) Jacket Framing Elevations Brace type In Vertical Frames Sizes of Primary Members (leg pile vertical frames braces primary horizontals) Joint Details (can gap etc) LegPiie Annulus Grouting Status

bull middotPOST ANDREW INSPECTION REPORT Significant Wave Height Hs (ft)

bull INFORMATION ON DAMAGES PRIOR TO ANDREW

bull RESULTS FROM IN-HOUSE STUDIES IF AVAILABLE

bull ~RIGINAL DESIGN CRITERIA INFORMATION IF AVAILABLE

bull SOIL SHEAR STRENGTH PROFILE IF AVAILABLE

Platforms Identified for Evaluating Therr Likelihood of Falling Under Unexpected Survival Category

OPlRATOR

middotADA

BLOClt

llmucnJllE

NAME

YEAll

INSTAUJD

WA1ER

DBPTll

DISTANCE

JlllOM SHOllE

(MILU)

Bl

(M)

AMOCO 111113

STl11

WDI075

A

bull D

St

17

120

1n

44

20

10 I

CllllVllQfl STlllO c 61 110 21 11 STIUO D Q 1110 21 11

STl134 p Q 137 29 11

ST0151 G 61 137 11

ST0151 I 3 121 11

ST0151 J Q 140 11

STl151 It 137 11

ST0151 PJl001 Q UT 11

ST0151 PROD-2 Q UT 11

ST017 A 3 140 35 11

WDIUT c 65 214 34 10

llXXON STI055 II 51 6T 14 I

STl15 A t3 34

WDI073

WD0073

Amiddotshy Q

1111 1111

22

22

I

I

WDI073 c 1n 22 I

- WDOCl93-shy II

65 1110 2T A 3 194 31 MOBU 111121

SS IOl7

A B

501

50

50

31

T

I

T 6

MtlllPllY SS114 B T 50 T T

PlllllIPS SS0149 A 5 55 33 T

SAMllDAN STln A 107 T T

SBllLL WDl103 A 65 223 2T WDl103 B 65 221 2T WD0104 c 65 221 2T WD0104 D 6T 20 2T

UNOCAL SS020I

SS 020I SSl20I

SS 0209

SS0209

II p

PURll

A

B

3

n 61

103

97

97

95

100

34

34

34

34 34

bullbull bullbullbull

- SS0209 D u 95 34 I

- SS 1209 SSl253

SS0253

G A

c

6T

2

100

165

1T5

34

54

54

bullbull I

I I

I I

SOUTH MARCH

~LAND

N

SOUTH PaTOI ENE ISWltD AREA

MMS Defined Limits ror Offshore Structures Requiring Inspections Following Andrew

t---Jbulln3 T25bull I bullTtl

bull1$1Htl1Abull bull1SIK1779bull

~

middot-DaTA AREA~T

CRANO rsLE

AREA bull1300

bullll4W

=shy

bullPath of Hurricane Andrew with Platforms Used In Calibration

Likelihood Function vs b All Success Case STISI K STIJOQ ST134W WD90A MCJll MCJ97

I 0000

0 9000

0 8000

07000

-Bo6000

tiiii 0 5000 -

04000

0 3000

0 2000

01000

0 0000

0 05 25

ri1rt middot-middot-middot-middot-middot-

I 5 2

b

---11 middot151K(blsu~ccsI

- bull - bull - fk 130Q(bfU~lCSS)

lie bull134W(blsuccc]

lk-WD90A[blsultCS]

- - - lk middotMC397(bfUtlCS]

--o-- lk-MCJll(bbuctcsJ

--- ll (bl 6 out of 6 Mlttcs case]

_

( Likelihood Functions middotSuccess Cases - p - 1 c I

CATEGORIZATION OF ADDITIONAL PLATFORMS

I I I

bull CATEGORIES

Unexpected Survivals

Expected Survivals

bull APPROACH

Identify Corresponding Platform(s) Evaluated In the JIP

Determine Max Wave Height at Platform Location

Evaluate Whether Wave Hit the Deck

middot - Evaluate Likelihood of R and S to Differ from the Base Cases

Review Joint Information

Assess Special Features

bull IDENTIFY NUMBER OF PLATFORMS IN EACH CATEGORY

I

UPDATING OF CALIBRATION RESULTS

bull INCLUDE EFFECT OF ADDITIONAL PLATFORMS ON bullBbull

Consider Increased 1Uncertalntles In R and S

Develop bullNeW Ukellhood Functions for Different Categories

middotLet bullxbull Number of Ukely Unexpected Survival Cases and y Number of Ukely Expected Survival Cases

Determine Combined Likelihood Function lk (b Iadditional platforms) middot

bull REVISED LIKELIHOOD FUNCTION

lk(b I all cases) = lk(b I 9 platforms) middot x lk(b Ix platforms) x lk(b I y platforms)

bull ESTABLISH REVISED BIAS FACTOR B

f8 (b) u f8 (b) lk (b I all cases)

II

t t ~

shy

APPLICATIONS OF BIAS FACTORS

ii APPLICABLE TO PARTICULAR LOAD AND RESISTANCE RECIPE

bull USE BY THE API COMMITTEES AND MMS

Criteria Revisions

Recommended Design Guidelines Revisions bull

bullbull REASSESSMENT STUDIES AND DECISION MAKING

middot

bull RISK ASSESSMENT STUDIES BY THE OPERATORS

Individual Platforms

Fleet of Platforms

middot IMR Program Implementation

bullbull NOT TO BE USED IN DESIGN

I I fl J

EXAMPLE USE OF BIAS FACTORS

bull PHASE I EXAMPLE APPLICATION

bull EVALUATION OF ANNUAL PROBABILITY OF FAILURE

Establish Annual Maximum Seastate Data

Determine Coefficients to Define Base Shear

Determine Ultimate Capacity of Platform

Determine Probablllty of Failure Versus bullbbull

-bull DETERMINE UPDATED PROBABILITY OF FAILURE

Pf - P [S gt R]

Pf I P [~ gt Rb I b] f 8 (b) db

Pr - J P [S gt Rb I b] f bull 8 (b) db

bull TOTAL PROBABILITIES OF FAILURE

Total Pf - 1 bull Il [1 middotPf] - l Pf bull

Total Pf - 1 bull Il [1middot Pf] - l Pfmiddot

Total Pr = 1 - n [1middot PfbullJ = l Pfshy

I I I

EXAMPLE APPLICATION bull PHASE I

PLATFORM ST 151 K

bull GULF OF MEXICO

bull WATER DEPTH = 137 FT

bull ANNUAL MAXIMUM HS MEDIAN 175 ft COV 031

bull ASSUMED STORM DURATION 3 hoursmiddot

bull ASSUMED NUMBER OF WAVES 800

bull MEDIAN HS FOR STORM HOURS 1 2 AND 3

bull MEDIAN CURRENT VALUE bull

bull DETERMINE ULTIMATE CAPACITY FOR THREE DIRECTIONS

bull DETERMINE ANNUAL FAILURE PROBABILITIES

Broadsidemiddot Diagonal End-on

0037 0025 0007

0071 0044 0018

0022 0016 0004

DETERMINE TOTAL ANNUAL PROBABILITY OF FAILURE

Pf - 007

Pf - 013

Pr - 004

bull

EXAMPLE USE OF BIAS FACTORS - PHASE II middot

bull DETERMINE PREDICTIVE PROBABILITY OF FAILURE

Case - 1

Pf _ S S Pf (b b2) fa1a2 (b b2) db db2

g bull -1 (b1 R I S) - 1

92 - (b2 R2 I S) - 1

Case -2

Pf s ~ [bJ f11 (b) db

Pf [bJ p [g lt 0)

g - bull 1 (b1 R1 I S) - 1

1

bull1) l

I

BENEFITS OF ANDREW JIP bull PHASE II

lt

bull IMPROVED LOAD AND RESISTANCE RECIPE

bull IMPROVED CALIBRATION PROCEDURE

bull MULTIPLE BIAS FACTORS

bull PROCEDURE TO INCLUDE EFFECT OF OTHER PLATFORMS

bull PROCEDURE TO USE MULTIPLE BIAS FACTORS

  • Structure Bookmarks
    • ANDREW PHASE II JOINT1 INDUSTRY PROJECT
    • PROGRESS MEETING NUMBER 1
    • ANDREW PHASE II JOINT INDUSTRY PROJECT PROGRESS MEETING NUMBER 1 NOVEMBER 16 AGENDA
    • ANDREW PHASE II UST OF PARTICIPANTS
    • ANDREW PHASE I OBJECTIVES AND RESULTS
    • ANDREW PHASE II PROJECT OBJECTIVES
    • SUBSEQUENT PROJECTS
    • ANDREW PHASE II PROJECT TEAM
    • ANDREW PHASE II UPDATE MODELING middotAND ANALYSIS RECIPE
    • -ANDREW PHASE II CONDUCTOR RESISTANCE
    • ANDREW PHASE Il WA VE FORCE PROFILE
    • ANDREW PHASE II WA VE FORCE PROFILEmiddot
    • ANDREW PHASE II BRACE MODELING
    • ANDREW PHASE II FOUNDATIONS
    • ANDREW PHASE II middot middot JOINT MODELING
    • ANDREW PHASE Il JOINT MODELING OPTIONS DISCUSSION
    • ANDREW PHASE II JOINT MODEJJNG OPTIONS ILLUSTRATIONS
    • ANDREW PHASE II JOINT MODELING PHASE II METHOD
    • ANDREW PHASE II ANDREW STORM CONDmONS UPDATED
    • Platform Selection bullbullbull1111111111111
    • BasisCriteria for Selection bullbullbull1111111111111
    • CALIBRATION
    • WHERE TO INTRODUCE BIAS FACTOR (B
    • PHASE I -KEY CONCLUSIONS
    • bull CALIBRATION ISSUES PHASE II
    • PHASE II WORK SCOPE (CALIBRATION)
    • SOFTWARE
    • SENSITIVITY STUDIES POSTERIOR DISTRIBUTION OF B
    • WEIGHTING APPROACH bull
    • INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON
    • CATEGORIZATION OF ADDITIONAL PLATFORMS
    • UPDATING OF CALIBRATION RESULTS
    • APPLICATIONS OF BIAS FACTORS
    • EXAMPLE USE OF BIAS FACTORS
    • EXAMPLE APPLICATION bull PHASE I PLATFORM ST 151 K
    • EXAMPLE USE OF BIAS FACTORS -PHASE II middot
    • BENEFITS OF ANDREW JIP bull PHASE II
Page 67:  · . ' ANDREW PHASE II PROJECT OBJECTIVES . ' • ·UPDATE MODELING AND ANALYSIS RECIPE· JOINTS ' ' ·BRACES WAVE PROFILE FOUNDATIONS . ' ' CONDUCTORS· • ANALVZE PLATFORMS USING

-

bullPrior and Posterior Distributions or B Evaluating Effect or Additional 13 Similar Observations as

Used in Andrew JIP The POsterior or B from Andrew-JIP Considered as New Prior or B

- 500

450

400

350

bull bull

bull bull bull bull J bull

bull bull f ~ I

I ~ Ibull bull

bullbull bull

l bull I -bullbull bull bull

I

I bull bull

bull I I~~bull bull bull

bull ~

I bull bull

~ ~_ ~~~

- - New Prior r bull

Posterior 114 new failure cases ~ 300- bull bull bull bull bull bull bull Posterior f 13 new damage casesl 150 ~

- bull bull - bull bull Posterior 116 new success cases i 2 00

Postenor 1113 new success failure -150 damage cases

I 00

0 50 -

0 00 0 00 0 20 0 40 0 60 0 80 I 00 120 I 40 160 110 2 00

Posterior ol B Mean= 120 COV= 0071 b

shy

shy

_rshy

Posterior Distribution of Bias Factor (B) bull Effect of Additional 13-Similar Observations

1

1 I

WEIGHTING APPROACH bull

bull ASSESSMENT OF PLATFORMS DATA

Select Platforms Most Mappable to the Analyzed Platforms

Unlikely SurvivalDamage Cases Identification

Based on Cut-Off Criteria Agemiddot Hs etc

bull INFORMATION REQUIRED FOR COMPARISON

bull CATEGORIZATION OF ADDITIONALmiddot PLATFORMS

bull UPDATING OF CALIBRATION RESULTS

--

OVERVIEW OF PLATFORMS SURVIVEDDAMAGED -~

Operator -

- Platforms

Survived CS)

Platforms

Damaged CD)

Platforms Used In

Phase IJIP -

JI~ ~Bill~IRBDll -

Amoco -

Chevron -

Exxon

Mob II

Phllllps

Shell

Unocal -

30

143

72 22

7

39 25shy -

1

12

bull

1 bull

1

1(S)+1 (D)

3 (S) +3(D)

2 (S)

bull shy

bull

bull

bull

Total Partlclnants Platforms 338 - 15 10C6S+4D) -

Other Operators

310

13

3 (D)

Total Platforms 648 28 13C6S+7D) -

Summary of Andrew JIP bull Phase I Platforms

Plairshy w v- N-ber Lea-le 1 shy car Hind_ o-w N-bull Deplh fl fl IJpe la bull n Yeadon Jadtet lbelledtf liefalit -- _ -OD

IL H - -shy _ICllshy IL IL - - rQClto ~ _

K STUIK 137 1963 8 365 608S4 Cellardeck

-middot STl-shy 170 1964 4 -~ K 39 62269 Cellardeck - --ST134W 137 1911 4 43 608S4 Smvlnlshy- _ - -

WD90A 184 1964 8 K 267Sbull34 83 SOS31 SUboftllar deck SmvMI

MC311 343 1978 8 x S7 62 - MC397 468 1991 4 x shy S33bull6SO 65911 -shySmeSmvlvol -- gt -

r-- T23ST52 63 1969 4 ton- shyKshy 52594 S03 -T25SSl39 62 1969 4 Kshy SOS94 S063 - middot ton-shy__ -

bullbull a ton--shyST161A 118 1964 8 K - 347S 51495 Cellarclecti

r~

ST1771 142 1965 8 K 39 60167 CeDordeck failme

STISIH 137 1964 8 K 395 608S4 CeDardeck Failme

K ST130A 140 1958 a SS 6095 I lblv Pai1uno - T21ST72 61 1969 4 Kshy S4S94 4974 I lhlv Pai1uno -

- -

JE

INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON

bull WATER DEPTH

bull bull YEAR OF INSTALLATIONDESIGN

bull PLATFORM ORIENTATION

bull PLATFORM LATITUDE AND LONGITUDE

bull KEY SKETCHES OF PLATFORM STRUCTURAL FRAMING

Number of Legs Plies Leg Batter Number of Conductors bull Deck Elevations (Including sub-cellar deck If any) Jacket Framing Elevations Brace type In Vertical Frames Sizes of Primary Members (leg pile vertical frames braces primary horizontals) Joint Details (can gap etc) LegPiie Annulus Grouting Status

bull middotPOST ANDREW INSPECTION REPORT Significant Wave Height Hs (ft)

bull INFORMATION ON DAMAGES PRIOR TO ANDREW

bull RESULTS FROM IN-HOUSE STUDIES IF AVAILABLE

bull ~RIGINAL DESIGN CRITERIA INFORMATION IF AVAILABLE

bull SOIL SHEAR STRENGTH PROFILE IF AVAILABLE

Platforms Identified for Evaluating Therr Likelihood of Falling Under Unexpected Survival Category

OPlRATOR

middotADA

BLOClt

llmucnJllE

NAME

YEAll

INSTAUJD

WA1ER

DBPTll

DISTANCE

JlllOM SHOllE

(MILU)

Bl

(M)

AMOCO 111113

STl11

WDI075

A

bull D

St

17

120

1n

44

20

10 I

CllllVllQfl STlllO c 61 110 21 11 STIUO D Q 1110 21 11

STl134 p Q 137 29 11

ST0151 G 61 137 11

ST0151 I 3 121 11

ST0151 J Q 140 11

STl151 It 137 11

ST0151 PJl001 Q UT 11

ST0151 PROD-2 Q UT 11

ST017 A 3 140 35 11

WDIUT c 65 214 34 10

llXXON STI055 II 51 6T 14 I

STl15 A t3 34

WDI073

WD0073

Amiddotshy Q

1111 1111

22

22

I

I

WDI073 c 1n 22 I

- WDOCl93-shy II

65 1110 2T A 3 194 31 MOBU 111121

SS IOl7

A B

501

50

50

31

T

I

T 6

MtlllPllY SS114 B T 50 T T

PlllllIPS SS0149 A 5 55 33 T

SAMllDAN STln A 107 T T

SBllLL WDl103 A 65 223 2T WDl103 B 65 221 2T WD0104 c 65 221 2T WD0104 D 6T 20 2T

UNOCAL SS020I

SS 020I SSl20I

SS 0209

SS0209

II p

PURll

A

B

3

n 61

103

97

97

95

100

34

34

34

34 34

bullbull bullbullbull

- SS0209 D u 95 34 I

- SS 1209 SSl253

SS0253

G A

c

6T

2

100

165

1T5

34

54

54

bullbull I

I I

I I

SOUTH MARCH

~LAND

N

SOUTH PaTOI ENE ISWltD AREA

MMS Defined Limits ror Offshore Structures Requiring Inspections Following Andrew

t---Jbulln3 T25bull I bullTtl

bull1$1Htl1Abull bull1SIK1779bull

~

middot-DaTA AREA~T

CRANO rsLE

AREA bull1300

bullll4W

=shy

bullPath of Hurricane Andrew with Platforms Used In Calibration

Likelihood Function vs b All Success Case STISI K STIJOQ ST134W WD90A MCJll MCJ97

I 0000

0 9000

0 8000

07000

-Bo6000

tiiii 0 5000 -

04000

0 3000

0 2000

01000

0 0000

0 05 25

ri1rt middot-middot-middot-middot-middot-

I 5 2

b

---11 middot151K(blsu~ccsI

- bull - bull - fk 130Q(bfU~lCSS)

lie bull134W(blsuccc]

lk-WD90A[blsultCS]

- - - lk middotMC397(bfUtlCS]

--o-- lk-MCJll(bbuctcsJ

--- ll (bl 6 out of 6 Mlttcs case]

_

( Likelihood Functions middotSuccess Cases - p - 1 c I

CATEGORIZATION OF ADDITIONAL PLATFORMS

I I I

bull CATEGORIES

Unexpected Survivals

Expected Survivals

bull APPROACH

Identify Corresponding Platform(s) Evaluated In the JIP

Determine Max Wave Height at Platform Location

Evaluate Whether Wave Hit the Deck

middot - Evaluate Likelihood of R and S to Differ from the Base Cases

Review Joint Information

Assess Special Features

bull IDENTIFY NUMBER OF PLATFORMS IN EACH CATEGORY

I

UPDATING OF CALIBRATION RESULTS

bull INCLUDE EFFECT OF ADDITIONAL PLATFORMS ON bullBbull

Consider Increased 1Uncertalntles In R and S

Develop bullNeW Ukellhood Functions for Different Categories

middotLet bullxbull Number of Ukely Unexpected Survival Cases and y Number of Ukely Expected Survival Cases

Determine Combined Likelihood Function lk (b Iadditional platforms) middot

bull REVISED LIKELIHOOD FUNCTION

lk(b I all cases) = lk(b I 9 platforms) middot x lk(b Ix platforms) x lk(b I y platforms)

bull ESTABLISH REVISED BIAS FACTOR B

f8 (b) u f8 (b) lk (b I all cases)

II

t t ~

shy

APPLICATIONS OF BIAS FACTORS

ii APPLICABLE TO PARTICULAR LOAD AND RESISTANCE RECIPE

bull USE BY THE API COMMITTEES AND MMS

Criteria Revisions

Recommended Design Guidelines Revisions bull

bullbull REASSESSMENT STUDIES AND DECISION MAKING

middot

bull RISK ASSESSMENT STUDIES BY THE OPERATORS

Individual Platforms

Fleet of Platforms

middot IMR Program Implementation

bullbull NOT TO BE USED IN DESIGN

I I fl J

EXAMPLE USE OF BIAS FACTORS

bull PHASE I EXAMPLE APPLICATION

bull EVALUATION OF ANNUAL PROBABILITY OF FAILURE

Establish Annual Maximum Seastate Data

Determine Coefficients to Define Base Shear

Determine Ultimate Capacity of Platform

Determine Probablllty of Failure Versus bullbbull

-bull DETERMINE UPDATED PROBABILITY OF FAILURE

Pf - P [S gt R]

Pf I P [~ gt Rb I b] f 8 (b) db

Pr - J P [S gt Rb I b] f bull 8 (b) db

bull TOTAL PROBABILITIES OF FAILURE

Total Pf - 1 bull Il [1 middotPf] - l Pf bull

Total Pf - 1 bull Il [1middot Pf] - l Pfmiddot

Total Pr = 1 - n [1middot PfbullJ = l Pfshy

I I I

EXAMPLE APPLICATION bull PHASE I

PLATFORM ST 151 K

bull GULF OF MEXICO

bull WATER DEPTH = 137 FT

bull ANNUAL MAXIMUM HS MEDIAN 175 ft COV 031

bull ASSUMED STORM DURATION 3 hoursmiddot

bull ASSUMED NUMBER OF WAVES 800

bull MEDIAN HS FOR STORM HOURS 1 2 AND 3

bull MEDIAN CURRENT VALUE bull

bull DETERMINE ULTIMATE CAPACITY FOR THREE DIRECTIONS

bull DETERMINE ANNUAL FAILURE PROBABILITIES

Broadsidemiddot Diagonal End-on

0037 0025 0007

0071 0044 0018

0022 0016 0004

DETERMINE TOTAL ANNUAL PROBABILITY OF FAILURE

Pf - 007

Pf - 013

Pr - 004

bull

EXAMPLE USE OF BIAS FACTORS - PHASE II middot

bull DETERMINE PREDICTIVE PROBABILITY OF FAILURE

Case - 1

Pf _ S S Pf (b b2) fa1a2 (b b2) db db2

g bull -1 (b1 R I S) - 1

92 - (b2 R2 I S) - 1

Case -2

Pf s ~ [bJ f11 (b) db

Pf [bJ p [g lt 0)

g - bull 1 (b1 R1 I S) - 1

1

bull1) l

I

BENEFITS OF ANDREW JIP bull PHASE II

lt

bull IMPROVED LOAD AND RESISTANCE RECIPE

bull IMPROVED CALIBRATION PROCEDURE

bull MULTIPLE BIAS FACTORS

bull PROCEDURE TO INCLUDE EFFECT OF OTHER PLATFORMS

bull PROCEDURE TO USE MULTIPLE BIAS FACTORS

  • Structure Bookmarks
    • ANDREW PHASE II JOINT1 INDUSTRY PROJECT
    • PROGRESS MEETING NUMBER 1
    • ANDREW PHASE II JOINT INDUSTRY PROJECT PROGRESS MEETING NUMBER 1 NOVEMBER 16 AGENDA
    • ANDREW PHASE II UST OF PARTICIPANTS
    • ANDREW PHASE I OBJECTIVES AND RESULTS
    • ANDREW PHASE II PROJECT OBJECTIVES
    • SUBSEQUENT PROJECTS
    • ANDREW PHASE II PROJECT TEAM
    • ANDREW PHASE II UPDATE MODELING middotAND ANALYSIS RECIPE
    • -ANDREW PHASE II CONDUCTOR RESISTANCE
    • ANDREW PHASE Il WA VE FORCE PROFILE
    • ANDREW PHASE II WA VE FORCE PROFILEmiddot
    • ANDREW PHASE II BRACE MODELING
    • ANDREW PHASE II FOUNDATIONS
    • ANDREW PHASE II middot middot JOINT MODELING
    • ANDREW PHASE Il JOINT MODELING OPTIONS DISCUSSION
    • ANDREW PHASE II JOINT MODEJJNG OPTIONS ILLUSTRATIONS
    • ANDREW PHASE II JOINT MODELING PHASE II METHOD
    • ANDREW PHASE II ANDREW STORM CONDmONS UPDATED
    • Platform Selection bullbullbull1111111111111
    • BasisCriteria for Selection bullbullbull1111111111111
    • CALIBRATION
    • WHERE TO INTRODUCE BIAS FACTOR (B
    • PHASE I -KEY CONCLUSIONS
    • bull CALIBRATION ISSUES PHASE II
    • PHASE II WORK SCOPE (CALIBRATION)
    • SOFTWARE
    • SENSITIVITY STUDIES POSTERIOR DISTRIBUTION OF B
    • WEIGHTING APPROACH bull
    • INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON
    • CATEGORIZATION OF ADDITIONAL PLATFORMS
    • UPDATING OF CALIBRATION RESULTS
    • APPLICATIONS OF BIAS FACTORS
    • EXAMPLE USE OF BIAS FACTORS
    • EXAMPLE APPLICATION bull PHASE I PLATFORM ST 151 K
    • EXAMPLE USE OF BIAS FACTORS -PHASE II middot
    • BENEFITS OF ANDREW JIP bull PHASE II
Page 68:  · . ' ANDREW PHASE II PROJECT OBJECTIVES . ' • ·UPDATE MODELING AND ANALYSIS RECIPE· JOINTS ' ' ·BRACES WAVE PROFILE FOUNDATIONS . ' ' CONDUCTORS· • ANALVZE PLATFORMS USING

1

1 I

WEIGHTING APPROACH bull

bull ASSESSMENT OF PLATFORMS DATA

Select Platforms Most Mappable to the Analyzed Platforms

Unlikely SurvivalDamage Cases Identification

Based on Cut-Off Criteria Agemiddot Hs etc

bull INFORMATION REQUIRED FOR COMPARISON

bull CATEGORIZATION OF ADDITIONALmiddot PLATFORMS

bull UPDATING OF CALIBRATION RESULTS

--

OVERVIEW OF PLATFORMS SURVIVEDDAMAGED -~

Operator -

- Platforms

Survived CS)

Platforms

Damaged CD)

Platforms Used In

Phase IJIP -

JI~ ~Bill~IRBDll -

Amoco -

Chevron -

Exxon

Mob II

Phllllps

Shell

Unocal -

30

143

72 22

7

39 25shy -

1

12

bull

1 bull

1

1(S)+1 (D)

3 (S) +3(D)

2 (S)

bull shy

bull

bull

bull

Total Partlclnants Platforms 338 - 15 10C6S+4D) -

Other Operators

310

13

3 (D)

Total Platforms 648 28 13C6S+7D) -

Summary of Andrew JIP bull Phase I Platforms

Plairshy w v- N-ber Lea-le 1 shy car Hind_ o-w N-bull Deplh fl fl IJpe la bull n Yeadon Jadtet lbelledtf liefalit -- _ -OD

IL H - -shy _ICllshy IL IL - - rQClto ~ _

K STUIK 137 1963 8 365 608S4 Cellardeck

-middot STl-shy 170 1964 4 -~ K 39 62269 Cellardeck - --ST134W 137 1911 4 43 608S4 Smvlnlshy- _ - -

WD90A 184 1964 8 K 267Sbull34 83 SOS31 SUboftllar deck SmvMI

MC311 343 1978 8 x S7 62 - MC397 468 1991 4 x shy S33bull6SO 65911 -shySmeSmvlvol -- gt -

r-- T23ST52 63 1969 4 ton- shyKshy 52594 S03 -T25SSl39 62 1969 4 Kshy SOS94 S063 - middot ton-shy__ -

bullbull a ton--shyST161A 118 1964 8 K - 347S 51495 Cellarclecti

r~

ST1771 142 1965 8 K 39 60167 CeDordeck failme

STISIH 137 1964 8 K 395 608S4 CeDardeck Failme

K ST130A 140 1958 a SS 6095 I lblv Pai1uno - T21ST72 61 1969 4 Kshy S4S94 4974 I lhlv Pai1uno -

- -

JE

INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON

bull WATER DEPTH

bull bull YEAR OF INSTALLATIONDESIGN

bull PLATFORM ORIENTATION

bull PLATFORM LATITUDE AND LONGITUDE

bull KEY SKETCHES OF PLATFORM STRUCTURAL FRAMING

Number of Legs Plies Leg Batter Number of Conductors bull Deck Elevations (Including sub-cellar deck If any) Jacket Framing Elevations Brace type In Vertical Frames Sizes of Primary Members (leg pile vertical frames braces primary horizontals) Joint Details (can gap etc) LegPiie Annulus Grouting Status

bull middotPOST ANDREW INSPECTION REPORT Significant Wave Height Hs (ft)

bull INFORMATION ON DAMAGES PRIOR TO ANDREW

bull RESULTS FROM IN-HOUSE STUDIES IF AVAILABLE

bull ~RIGINAL DESIGN CRITERIA INFORMATION IF AVAILABLE

bull SOIL SHEAR STRENGTH PROFILE IF AVAILABLE

Platforms Identified for Evaluating Therr Likelihood of Falling Under Unexpected Survival Category

OPlRATOR

middotADA

BLOClt

llmucnJllE

NAME

YEAll

INSTAUJD

WA1ER

DBPTll

DISTANCE

JlllOM SHOllE

(MILU)

Bl

(M)

AMOCO 111113

STl11

WDI075

A

bull D

St

17

120

1n

44

20

10 I

CllllVllQfl STlllO c 61 110 21 11 STIUO D Q 1110 21 11

STl134 p Q 137 29 11

ST0151 G 61 137 11

ST0151 I 3 121 11

ST0151 J Q 140 11

STl151 It 137 11

ST0151 PJl001 Q UT 11

ST0151 PROD-2 Q UT 11

ST017 A 3 140 35 11

WDIUT c 65 214 34 10

llXXON STI055 II 51 6T 14 I

STl15 A t3 34

WDI073

WD0073

Amiddotshy Q

1111 1111

22

22

I

I

WDI073 c 1n 22 I

- WDOCl93-shy II

65 1110 2T A 3 194 31 MOBU 111121

SS IOl7

A B

501

50

50

31

T

I

T 6

MtlllPllY SS114 B T 50 T T

PlllllIPS SS0149 A 5 55 33 T

SAMllDAN STln A 107 T T

SBllLL WDl103 A 65 223 2T WDl103 B 65 221 2T WD0104 c 65 221 2T WD0104 D 6T 20 2T

UNOCAL SS020I

SS 020I SSl20I

SS 0209

SS0209

II p

PURll

A

B

3

n 61

103

97

97

95

100

34

34

34

34 34

bullbull bullbullbull

- SS0209 D u 95 34 I

- SS 1209 SSl253

SS0253

G A

c

6T

2

100

165

1T5

34

54

54

bullbull I

I I

I I

SOUTH MARCH

~LAND

N

SOUTH PaTOI ENE ISWltD AREA

MMS Defined Limits ror Offshore Structures Requiring Inspections Following Andrew

t---Jbulln3 T25bull I bullTtl

bull1$1Htl1Abull bull1SIK1779bull

~

middot-DaTA AREA~T

CRANO rsLE

AREA bull1300

bullll4W

=shy

bullPath of Hurricane Andrew with Platforms Used In Calibration

Likelihood Function vs b All Success Case STISI K STIJOQ ST134W WD90A MCJll MCJ97

I 0000

0 9000

0 8000

07000

-Bo6000

tiiii 0 5000 -

04000

0 3000

0 2000

01000

0 0000

0 05 25

ri1rt middot-middot-middot-middot-middot-

I 5 2

b

---11 middot151K(blsu~ccsI

- bull - bull - fk 130Q(bfU~lCSS)

lie bull134W(blsuccc]

lk-WD90A[blsultCS]

- - - lk middotMC397(bfUtlCS]

--o-- lk-MCJll(bbuctcsJ

--- ll (bl 6 out of 6 Mlttcs case]

_

( Likelihood Functions middotSuccess Cases - p - 1 c I

CATEGORIZATION OF ADDITIONAL PLATFORMS

I I I

bull CATEGORIES

Unexpected Survivals

Expected Survivals

bull APPROACH

Identify Corresponding Platform(s) Evaluated In the JIP

Determine Max Wave Height at Platform Location

Evaluate Whether Wave Hit the Deck

middot - Evaluate Likelihood of R and S to Differ from the Base Cases

Review Joint Information

Assess Special Features

bull IDENTIFY NUMBER OF PLATFORMS IN EACH CATEGORY

I

UPDATING OF CALIBRATION RESULTS

bull INCLUDE EFFECT OF ADDITIONAL PLATFORMS ON bullBbull

Consider Increased 1Uncertalntles In R and S

Develop bullNeW Ukellhood Functions for Different Categories

middotLet bullxbull Number of Ukely Unexpected Survival Cases and y Number of Ukely Expected Survival Cases

Determine Combined Likelihood Function lk (b Iadditional platforms) middot

bull REVISED LIKELIHOOD FUNCTION

lk(b I all cases) = lk(b I 9 platforms) middot x lk(b Ix platforms) x lk(b I y platforms)

bull ESTABLISH REVISED BIAS FACTOR B

f8 (b) u f8 (b) lk (b I all cases)

II

t t ~

shy

APPLICATIONS OF BIAS FACTORS

ii APPLICABLE TO PARTICULAR LOAD AND RESISTANCE RECIPE

bull USE BY THE API COMMITTEES AND MMS

Criteria Revisions

Recommended Design Guidelines Revisions bull

bullbull REASSESSMENT STUDIES AND DECISION MAKING

middot

bull RISK ASSESSMENT STUDIES BY THE OPERATORS

Individual Platforms

Fleet of Platforms

middot IMR Program Implementation

bullbull NOT TO BE USED IN DESIGN

I I fl J

EXAMPLE USE OF BIAS FACTORS

bull PHASE I EXAMPLE APPLICATION

bull EVALUATION OF ANNUAL PROBABILITY OF FAILURE

Establish Annual Maximum Seastate Data

Determine Coefficients to Define Base Shear

Determine Ultimate Capacity of Platform

Determine Probablllty of Failure Versus bullbbull

-bull DETERMINE UPDATED PROBABILITY OF FAILURE

Pf - P [S gt R]

Pf I P [~ gt Rb I b] f 8 (b) db

Pr - J P [S gt Rb I b] f bull 8 (b) db

bull TOTAL PROBABILITIES OF FAILURE

Total Pf - 1 bull Il [1 middotPf] - l Pf bull

Total Pf - 1 bull Il [1middot Pf] - l Pfmiddot

Total Pr = 1 - n [1middot PfbullJ = l Pfshy

I I I

EXAMPLE APPLICATION bull PHASE I

PLATFORM ST 151 K

bull GULF OF MEXICO

bull WATER DEPTH = 137 FT

bull ANNUAL MAXIMUM HS MEDIAN 175 ft COV 031

bull ASSUMED STORM DURATION 3 hoursmiddot

bull ASSUMED NUMBER OF WAVES 800

bull MEDIAN HS FOR STORM HOURS 1 2 AND 3

bull MEDIAN CURRENT VALUE bull

bull DETERMINE ULTIMATE CAPACITY FOR THREE DIRECTIONS

bull DETERMINE ANNUAL FAILURE PROBABILITIES

Broadsidemiddot Diagonal End-on

0037 0025 0007

0071 0044 0018

0022 0016 0004

DETERMINE TOTAL ANNUAL PROBABILITY OF FAILURE

Pf - 007

Pf - 013

Pr - 004

bull

EXAMPLE USE OF BIAS FACTORS - PHASE II middot

bull DETERMINE PREDICTIVE PROBABILITY OF FAILURE

Case - 1

Pf _ S S Pf (b b2) fa1a2 (b b2) db db2

g bull -1 (b1 R I S) - 1

92 - (b2 R2 I S) - 1

Case -2

Pf s ~ [bJ f11 (b) db

Pf [bJ p [g lt 0)

g - bull 1 (b1 R1 I S) - 1

1

bull1) l

I

BENEFITS OF ANDREW JIP bull PHASE II

lt

bull IMPROVED LOAD AND RESISTANCE RECIPE

bull IMPROVED CALIBRATION PROCEDURE

bull MULTIPLE BIAS FACTORS

bull PROCEDURE TO INCLUDE EFFECT OF OTHER PLATFORMS

bull PROCEDURE TO USE MULTIPLE BIAS FACTORS

  • Structure Bookmarks
    • ANDREW PHASE II JOINT1 INDUSTRY PROJECT
    • PROGRESS MEETING NUMBER 1
    • ANDREW PHASE II JOINT INDUSTRY PROJECT PROGRESS MEETING NUMBER 1 NOVEMBER 16 AGENDA
    • ANDREW PHASE II UST OF PARTICIPANTS
    • ANDREW PHASE I OBJECTIVES AND RESULTS
    • ANDREW PHASE II PROJECT OBJECTIVES
    • SUBSEQUENT PROJECTS
    • ANDREW PHASE II PROJECT TEAM
    • ANDREW PHASE II UPDATE MODELING middotAND ANALYSIS RECIPE
    • -ANDREW PHASE II CONDUCTOR RESISTANCE
    • ANDREW PHASE Il WA VE FORCE PROFILE
    • ANDREW PHASE II WA VE FORCE PROFILEmiddot
    • ANDREW PHASE II BRACE MODELING
    • ANDREW PHASE II FOUNDATIONS
    • ANDREW PHASE II middot middot JOINT MODELING
    • ANDREW PHASE Il JOINT MODELING OPTIONS DISCUSSION
    • ANDREW PHASE II JOINT MODEJJNG OPTIONS ILLUSTRATIONS
    • ANDREW PHASE II JOINT MODELING PHASE II METHOD
    • ANDREW PHASE II ANDREW STORM CONDmONS UPDATED
    • Platform Selection bullbullbull1111111111111
    • BasisCriteria for Selection bullbullbull1111111111111
    • CALIBRATION
    • WHERE TO INTRODUCE BIAS FACTOR (B
    • PHASE I -KEY CONCLUSIONS
    • bull CALIBRATION ISSUES PHASE II
    • PHASE II WORK SCOPE (CALIBRATION)
    • SOFTWARE
    • SENSITIVITY STUDIES POSTERIOR DISTRIBUTION OF B
    • WEIGHTING APPROACH bull
    • INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON
    • CATEGORIZATION OF ADDITIONAL PLATFORMS
    • UPDATING OF CALIBRATION RESULTS
    • APPLICATIONS OF BIAS FACTORS
    • EXAMPLE USE OF BIAS FACTORS
    • EXAMPLE APPLICATION bull PHASE I PLATFORM ST 151 K
    • EXAMPLE USE OF BIAS FACTORS -PHASE II middot
    • BENEFITS OF ANDREW JIP bull PHASE II
Page 69:  · . ' ANDREW PHASE II PROJECT OBJECTIVES . ' • ·UPDATE MODELING AND ANALYSIS RECIPE· JOINTS ' ' ·BRACES WAVE PROFILE FOUNDATIONS . ' ' CONDUCTORS· • ANALVZE PLATFORMS USING

--

OVERVIEW OF PLATFORMS SURVIVEDDAMAGED -~

Operator -

- Platforms

Survived CS)

Platforms

Damaged CD)

Platforms Used In

Phase IJIP -

JI~ ~Bill~IRBDll -

Amoco -

Chevron -

Exxon

Mob II

Phllllps

Shell

Unocal -

30

143

72 22

7

39 25shy -

1

12

bull

1 bull

1

1(S)+1 (D)

3 (S) +3(D)

2 (S)

bull shy

bull

bull

bull

Total Partlclnants Platforms 338 - 15 10C6S+4D) -

Other Operators

310

13

3 (D)

Total Platforms 648 28 13C6S+7D) -

Summary of Andrew JIP bull Phase I Platforms

Plairshy w v- N-ber Lea-le 1 shy car Hind_ o-w N-bull Deplh fl fl IJpe la bull n Yeadon Jadtet lbelledtf liefalit -- _ -OD

IL H - -shy _ICllshy IL IL - - rQClto ~ _

K STUIK 137 1963 8 365 608S4 Cellardeck

-middot STl-shy 170 1964 4 -~ K 39 62269 Cellardeck - --ST134W 137 1911 4 43 608S4 Smvlnlshy- _ - -

WD90A 184 1964 8 K 267Sbull34 83 SOS31 SUboftllar deck SmvMI

MC311 343 1978 8 x S7 62 - MC397 468 1991 4 x shy S33bull6SO 65911 -shySmeSmvlvol -- gt -

r-- T23ST52 63 1969 4 ton- shyKshy 52594 S03 -T25SSl39 62 1969 4 Kshy SOS94 S063 - middot ton-shy__ -

bullbull a ton--shyST161A 118 1964 8 K - 347S 51495 Cellarclecti

r~

ST1771 142 1965 8 K 39 60167 CeDordeck failme

STISIH 137 1964 8 K 395 608S4 CeDardeck Failme

K ST130A 140 1958 a SS 6095 I lblv Pai1uno - T21ST72 61 1969 4 Kshy S4S94 4974 I lhlv Pai1uno -

- -

JE

INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON

bull WATER DEPTH

bull bull YEAR OF INSTALLATIONDESIGN

bull PLATFORM ORIENTATION

bull PLATFORM LATITUDE AND LONGITUDE

bull KEY SKETCHES OF PLATFORM STRUCTURAL FRAMING

Number of Legs Plies Leg Batter Number of Conductors bull Deck Elevations (Including sub-cellar deck If any) Jacket Framing Elevations Brace type In Vertical Frames Sizes of Primary Members (leg pile vertical frames braces primary horizontals) Joint Details (can gap etc) LegPiie Annulus Grouting Status

bull middotPOST ANDREW INSPECTION REPORT Significant Wave Height Hs (ft)

bull INFORMATION ON DAMAGES PRIOR TO ANDREW

bull RESULTS FROM IN-HOUSE STUDIES IF AVAILABLE

bull ~RIGINAL DESIGN CRITERIA INFORMATION IF AVAILABLE

bull SOIL SHEAR STRENGTH PROFILE IF AVAILABLE

Platforms Identified for Evaluating Therr Likelihood of Falling Under Unexpected Survival Category

OPlRATOR

middotADA

BLOClt

llmucnJllE

NAME

YEAll

INSTAUJD

WA1ER

DBPTll

DISTANCE

JlllOM SHOllE

(MILU)

Bl

(M)

AMOCO 111113

STl11

WDI075

A

bull D

St

17

120

1n

44

20

10 I

CllllVllQfl STlllO c 61 110 21 11 STIUO D Q 1110 21 11

STl134 p Q 137 29 11

ST0151 G 61 137 11

ST0151 I 3 121 11

ST0151 J Q 140 11

STl151 It 137 11

ST0151 PJl001 Q UT 11

ST0151 PROD-2 Q UT 11

ST017 A 3 140 35 11

WDIUT c 65 214 34 10

llXXON STI055 II 51 6T 14 I

STl15 A t3 34

WDI073

WD0073

Amiddotshy Q

1111 1111

22

22

I

I

WDI073 c 1n 22 I

- WDOCl93-shy II

65 1110 2T A 3 194 31 MOBU 111121

SS IOl7

A B

501

50

50

31

T

I

T 6

MtlllPllY SS114 B T 50 T T

PlllllIPS SS0149 A 5 55 33 T

SAMllDAN STln A 107 T T

SBllLL WDl103 A 65 223 2T WDl103 B 65 221 2T WD0104 c 65 221 2T WD0104 D 6T 20 2T

UNOCAL SS020I

SS 020I SSl20I

SS 0209

SS0209

II p

PURll

A

B

3

n 61

103

97

97

95

100

34

34

34

34 34

bullbull bullbullbull

- SS0209 D u 95 34 I

- SS 1209 SSl253

SS0253

G A

c

6T

2

100

165

1T5

34

54

54

bullbull I

I I

I I

SOUTH MARCH

~LAND

N

SOUTH PaTOI ENE ISWltD AREA

MMS Defined Limits ror Offshore Structures Requiring Inspections Following Andrew

t---Jbulln3 T25bull I bullTtl

bull1$1Htl1Abull bull1SIK1779bull

~

middot-DaTA AREA~T

CRANO rsLE

AREA bull1300

bullll4W

=shy

bullPath of Hurricane Andrew with Platforms Used In Calibration

Likelihood Function vs b All Success Case STISI K STIJOQ ST134W WD90A MCJll MCJ97

I 0000

0 9000

0 8000

07000

-Bo6000

tiiii 0 5000 -

04000

0 3000

0 2000

01000

0 0000

0 05 25

ri1rt middot-middot-middot-middot-middot-

I 5 2

b

---11 middot151K(blsu~ccsI

- bull - bull - fk 130Q(bfU~lCSS)

lie bull134W(blsuccc]

lk-WD90A[blsultCS]

- - - lk middotMC397(bfUtlCS]

--o-- lk-MCJll(bbuctcsJ

--- ll (bl 6 out of 6 Mlttcs case]

_

( Likelihood Functions middotSuccess Cases - p - 1 c I

CATEGORIZATION OF ADDITIONAL PLATFORMS

I I I

bull CATEGORIES

Unexpected Survivals

Expected Survivals

bull APPROACH

Identify Corresponding Platform(s) Evaluated In the JIP

Determine Max Wave Height at Platform Location

Evaluate Whether Wave Hit the Deck

middot - Evaluate Likelihood of R and S to Differ from the Base Cases

Review Joint Information

Assess Special Features

bull IDENTIFY NUMBER OF PLATFORMS IN EACH CATEGORY

I

UPDATING OF CALIBRATION RESULTS

bull INCLUDE EFFECT OF ADDITIONAL PLATFORMS ON bullBbull

Consider Increased 1Uncertalntles In R and S

Develop bullNeW Ukellhood Functions for Different Categories

middotLet bullxbull Number of Ukely Unexpected Survival Cases and y Number of Ukely Expected Survival Cases

Determine Combined Likelihood Function lk (b Iadditional platforms) middot

bull REVISED LIKELIHOOD FUNCTION

lk(b I all cases) = lk(b I 9 platforms) middot x lk(b Ix platforms) x lk(b I y platforms)

bull ESTABLISH REVISED BIAS FACTOR B

f8 (b) u f8 (b) lk (b I all cases)

II

t t ~

shy

APPLICATIONS OF BIAS FACTORS

ii APPLICABLE TO PARTICULAR LOAD AND RESISTANCE RECIPE

bull USE BY THE API COMMITTEES AND MMS

Criteria Revisions

Recommended Design Guidelines Revisions bull

bullbull REASSESSMENT STUDIES AND DECISION MAKING

middot

bull RISK ASSESSMENT STUDIES BY THE OPERATORS

Individual Platforms

Fleet of Platforms

middot IMR Program Implementation

bullbull NOT TO BE USED IN DESIGN

I I fl J

EXAMPLE USE OF BIAS FACTORS

bull PHASE I EXAMPLE APPLICATION

bull EVALUATION OF ANNUAL PROBABILITY OF FAILURE

Establish Annual Maximum Seastate Data

Determine Coefficients to Define Base Shear

Determine Ultimate Capacity of Platform

Determine Probablllty of Failure Versus bullbbull

-bull DETERMINE UPDATED PROBABILITY OF FAILURE

Pf - P [S gt R]

Pf I P [~ gt Rb I b] f 8 (b) db

Pr - J P [S gt Rb I b] f bull 8 (b) db

bull TOTAL PROBABILITIES OF FAILURE

Total Pf - 1 bull Il [1 middotPf] - l Pf bull

Total Pf - 1 bull Il [1middot Pf] - l Pfmiddot

Total Pr = 1 - n [1middot PfbullJ = l Pfshy

I I I

EXAMPLE APPLICATION bull PHASE I

PLATFORM ST 151 K

bull GULF OF MEXICO

bull WATER DEPTH = 137 FT

bull ANNUAL MAXIMUM HS MEDIAN 175 ft COV 031

bull ASSUMED STORM DURATION 3 hoursmiddot

bull ASSUMED NUMBER OF WAVES 800

bull MEDIAN HS FOR STORM HOURS 1 2 AND 3

bull MEDIAN CURRENT VALUE bull

bull DETERMINE ULTIMATE CAPACITY FOR THREE DIRECTIONS

bull DETERMINE ANNUAL FAILURE PROBABILITIES

Broadsidemiddot Diagonal End-on

0037 0025 0007

0071 0044 0018

0022 0016 0004

DETERMINE TOTAL ANNUAL PROBABILITY OF FAILURE

Pf - 007

Pf - 013

Pr - 004

bull

EXAMPLE USE OF BIAS FACTORS - PHASE II middot

bull DETERMINE PREDICTIVE PROBABILITY OF FAILURE

Case - 1

Pf _ S S Pf (b b2) fa1a2 (b b2) db db2

g bull -1 (b1 R I S) - 1

92 - (b2 R2 I S) - 1

Case -2

Pf s ~ [bJ f11 (b) db

Pf [bJ p [g lt 0)

g - bull 1 (b1 R1 I S) - 1

1

bull1) l

I

BENEFITS OF ANDREW JIP bull PHASE II

lt

bull IMPROVED LOAD AND RESISTANCE RECIPE

bull IMPROVED CALIBRATION PROCEDURE

bull MULTIPLE BIAS FACTORS

bull PROCEDURE TO INCLUDE EFFECT OF OTHER PLATFORMS

bull PROCEDURE TO USE MULTIPLE BIAS FACTORS

  • Structure Bookmarks
    • ANDREW PHASE II JOINT1 INDUSTRY PROJECT
    • PROGRESS MEETING NUMBER 1
    • ANDREW PHASE II JOINT INDUSTRY PROJECT PROGRESS MEETING NUMBER 1 NOVEMBER 16 AGENDA
    • ANDREW PHASE II UST OF PARTICIPANTS
    • ANDREW PHASE I OBJECTIVES AND RESULTS
    • ANDREW PHASE II PROJECT OBJECTIVES
    • SUBSEQUENT PROJECTS
    • ANDREW PHASE II PROJECT TEAM
    • ANDREW PHASE II UPDATE MODELING middotAND ANALYSIS RECIPE
    • -ANDREW PHASE II CONDUCTOR RESISTANCE
    • ANDREW PHASE Il WA VE FORCE PROFILE
    • ANDREW PHASE II WA VE FORCE PROFILEmiddot
    • ANDREW PHASE II BRACE MODELING
    • ANDREW PHASE II FOUNDATIONS
    • ANDREW PHASE II middot middot JOINT MODELING
    • ANDREW PHASE Il JOINT MODELING OPTIONS DISCUSSION
    • ANDREW PHASE II JOINT MODEJJNG OPTIONS ILLUSTRATIONS
    • ANDREW PHASE II JOINT MODELING PHASE II METHOD
    • ANDREW PHASE II ANDREW STORM CONDmONS UPDATED
    • Platform Selection bullbullbull1111111111111
    • BasisCriteria for Selection bullbullbull1111111111111
    • CALIBRATION
    • WHERE TO INTRODUCE BIAS FACTOR (B
    • PHASE I -KEY CONCLUSIONS
    • bull CALIBRATION ISSUES PHASE II
    • PHASE II WORK SCOPE (CALIBRATION)
    • SOFTWARE
    • SENSITIVITY STUDIES POSTERIOR DISTRIBUTION OF B
    • WEIGHTING APPROACH bull
    • INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON
    • CATEGORIZATION OF ADDITIONAL PLATFORMS
    • UPDATING OF CALIBRATION RESULTS
    • APPLICATIONS OF BIAS FACTORS
    • EXAMPLE USE OF BIAS FACTORS
    • EXAMPLE APPLICATION bull PHASE I PLATFORM ST 151 K
    • EXAMPLE USE OF BIAS FACTORS -PHASE II middot
    • BENEFITS OF ANDREW JIP bull PHASE II
Page 70:  · . ' ANDREW PHASE II PROJECT OBJECTIVES . ' • ·UPDATE MODELING AND ANALYSIS RECIPE· JOINTS ' ' ·BRACES WAVE PROFILE FOUNDATIONS . ' ' CONDUCTORS· • ANALVZE PLATFORMS USING

Summary of Andrew JIP bull Phase I Platforms

Plairshy w v- N-ber Lea-le 1 shy car Hind_ o-w N-bull Deplh fl fl IJpe la bull n Yeadon Jadtet lbelledtf liefalit -- _ -OD

IL H - -shy _ICllshy IL IL - - rQClto ~ _

K STUIK 137 1963 8 365 608S4 Cellardeck

-middot STl-shy 170 1964 4 -~ K 39 62269 Cellardeck - --ST134W 137 1911 4 43 608S4 Smvlnlshy- _ - -

WD90A 184 1964 8 K 267Sbull34 83 SOS31 SUboftllar deck SmvMI

MC311 343 1978 8 x S7 62 - MC397 468 1991 4 x shy S33bull6SO 65911 -shySmeSmvlvol -- gt -

r-- T23ST52 63 1969 4 ton- shyKshy 52594 S03 -T25SSl39 62 1969 4 Kshy SOS94 S063 - middot ton-shy__ -

bullbull a ton--shyST161A 118 1964 8 K - 347S 51495 Cellarclecti

r~

ST1771 142 1965 8 K 39 60167 CeDordeck failme

STISIH 137 1964 8 K 395 608S4 CeDardeck Failme

K ST130A 140 1958 a SS 6095 I lblv Pai1uno - T21ST72 61 1969 4 Kshy S4S94 4974 I lhlv Pai1uno -

- -

JE

INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON

bull WATER DEPTH

bull bull YEAR OF INSTALLATIONDESIGN

bull PLATFORM ORIENTATION

bull PLATFORM LATITUDE AND LONGITUDE

bull KEY SKETCHES OF PLATFORM STRUCTURAL FRAMING

Number of Legs Plies Leg Batter Number of Conductors bull Deck Elevations (Including sub-cellar deck If any) Jacket Framing Elevations Brace type In Vertical Frames Sizes of Primary Members (leg pile vertical frames braces primary horizontals) Joint Details (can gap etc) LegPiie Annulus Grouting Status

bull middotPOST ANDREW INSPECTION REPORT Significant Wave Height Hs (ft)

bull INFORMATION ON DAMAGES PRIOR TO ANDREW

bull RESULTS FROM IN-HOUSE STUDIES IF AVAILABLE

bull ~RIGINAL DESIGN CRITERIA INFORMATION IF AVAILABLE

bull SOIL SHEAR STRENGTH PROFILE IF AVAILABLE

Platforms Identified for Evaluating Therr Likelihood of Falling Under Unexpected Survival Category

OPlRATOR

middotADA

BLOClt

llmucnJllE

NAME

YEAll

INSTAUJD

WA1ER

DBPTll

DISTANCE

JlllOM SHOllE

(MILU)

Bl

(M)

AMOCO 111113

STl11

WDI075

A

bull D

St

17

120

1n

44

20

10 I

CllllVllQfl STlllO c 61 110 21 11 STIUO D Q 1110 21 11

STl134 p Q 137 29 11

ST0151 G 61 137 11

ST0151 I 3 121 11

ST0151 J Q 140 11

STl151 It 137 11

ST0151 PJl001 Q UT 11

ST0151 PROD-2 Q UT 11

ST017 A 3 140 35 11

WDIUT c 65 214 34 10

llXXON STI055 II 51 6T 14 I

STl15 A t3 34

WDI073

WD0073

Amiddotshy Q

1111 1111

22

22

I

I

WDI073 c 1n 22 I

- WDOCl93-shy II

65 1110 2T A 3 194 31 MOBU 111121

SS IOl7

A B

501

50

50

31

T

I

T 6

MtlllPllY SS114 B T 50 T T

PlllllIPS SS0149 A 5 55 33 T

SAMllDAN STln A 107 T T

SBllLL WDl103 A 65 223 2T WDl103 B 65 221 2T WD0104 c 65 221 2T WD0104 D 6T 20 2T

UNOCAL SS020I

SS 020I SSl20I

SS 0209

SS0209

II p

PURll

A

B

3

n 61

103

97

97

95

100

34

34

34

34 34

bullbull bullbullbull

- SS0209 D u 95 34 I

- SS 1209 SSl253

SS0253

G A

c

6T

2

100

165

1T5

34

54

54

bullbull I

I I

I I

SOUTH MARCH

~LAND

N

SOUTH PaTOI ENE ISWltD AREA

MMS Defined Limits ror Offshore Structures Requiring Inspections Following Andrew

t---Jbulln3 T25bull I bullTtl

bull1$1Htl1Abull bull1SIK1779bull

~

middot-DaTA AREA~T

CRANO rsLE

AREA bull1300

bullll4W

=shy

bullPath of Hurricane Andrew with Platforms Used In Calibration

Likelihood Function vs b All Success Case STISI K STIJOQ ST134W WD90A MCJll MCJ97

I 0000

0 9000

0 8000

07000

-Bo6000

tiiii 0 5000 -

04000

0 3000

0 2000

01000

0 0000

0 05 25

ri1rt middot-middot-middot-middot-middot-

I 5 2

b

---11 middot151K(blsu~ccsI

- bull - bull - fk 130Q(bfU~lCSS)

lie bull134W(blsuccc]

lk-WD90A[blsultCS]

- - - lk middotMC397(bfUtlCS]

--o-- lk-MCJll(bbuctcsJ

--- ll (bl 6 out of 6 Mlttcs case]

_

( Likelihood Functions middotSuccess Cases - p - 1 c I

CATEGORIZATION OF ADDITIONAL PLATFORMS

I I I

bull CATEGORIES

Unexpected Survivals

Expected Survivals

bull APPROACH

Identify Corresponding Platform(s) Evaluated In the JIP

Determine Max Wave Height at Platform Location

Evaluate Whether Wave Hit the Deck

middot - Evaluate Likelihood of R and S to Differ from the Base Cases

Review Joint Information

Assess Special Features

bull IDENTIFY NUMBER OF PLATFORMS IN EACH CATEGORY

I

UPDATING OF CALIBRATION RESULTS

bull INCLUDE EFFECT OF ADDITIONAL PLATFORMS ON bullBbull

Consider Increased 1Uncertalntles In R and S

Develop bullNeW Ukellhood Functions for Different Categories

middotLet bullxbull Number of Ukely Unexpected Survival Cases and y Number of Ukely Expected Survival Cases

Determine Combined Likelihood Function lk (b Iadditional platforms) middot

bull REVISED LIKELIHOOD FUNCTION

lk(b I all cases) = lk(b I 9 platforms) middot x lk(b Ix platforms) x lk(b I y platforms)

bull ESTABLISH REVISED BIAS FACTOR B

f8 (b) u f8 (b) lk (b I all cases)

II

t t ~

shy

APPLICATIONS OF BIAS FACTORS

ii APPLICABLE TO PARTICULAR LOAD AND RESISTANCE RECIPE

bull USE BY THE API COMMITTEES AND MMS

Criteria Revisions

Recommended Design Guidelines Revisions bull

bullbull REASSESSMENT STUDIES AND DECISION MAKING

middot

bull RISK ASSESSMENT STUDIES BY THE OPERATORS

Individual Platforms

Fleet of Platforms

middot IMR Program Implementation

bullbull NOT TO BE USED IN DESIGN

I I fl J

EXAMPLE USE OF BIAS FACTORS

bull PHASE I EXAMPLE APPLICATION

bull EVALUATION OF ANNUAL PROBABILITY OF FAILURE

Establish Annual Maximum Seastate Data

Determine Coefficients to Define Base Shear

Determine Ultimate Capacity of Platform

Determine Probablllty of Failure Versus bullbbull

-bull DETERMINE UPDATED PROBABILITY OF FAILURE

Pf - P [S gt R]

Pf I P [~ gt Rb I b] f 8 (b) db

Pr - J P [S gt Rb I b] f bull 8 (b) db

bull TOTAL PROBABILITIES OF FAILURE

Total Pf - 1 bull Il [1 middotPf] - l Pf bull

Total Pf - 1 bull Il [1middot Pf] - l Pfmiddot

Total Pr = 1 - n [1middot PfbullJ = l Pfshy

I I I

EXAMPLE APPLICATION bull PHASE I

PLATFORM ST 151 K

bull GULF OF MEXICO

bull WATER DEPTH = 137 FT

bull ANNUAL MAXIMUM HS MEDIAN 175 ft COV 031

bull ASSUMED STORM DURATION 3 hoursmiddot

bull ASSUMED NUMBER OF WAVES 800

bull MEDIAN HS FOR STORM HOURS 1 2 AND 3

bull MEDIAN CURRENT VALUE bull

bull DETERMINE ULTIMATE CAPACITY FOR THREE DIRECTIONS

bull DETERMINE ANNUAL FAILURE PROBABILITIES

Broadsidemiddot Diagonal End-on

0037 0025 0007

0071 0044 0018

0022 0016 0004

DETERMINE TOTAL ANNUAL PROBABILITY OF FAILURE

Pf - 007

Pf - 013

Pr - 004

bull

EXAMPLE USE OF BIAS FACTORS - PHASE II middot

bull DETERMINE PREDICTIVE PROBABILITY OF FAILURE

Case - 1

Pf _ S S Pf (b b2) fa1a2 (b b2) db db2

g bull -1 (b1 R I S) - 1

92 - (b2 R2 I S) - 1

Case -2

Pf s ~ [bJ f11 (b) db

Pf [bJ p [g lt 0)

g - bull 1 (b1 R1 I S) - 1

1

bull1) l

I

BENEFITS OF ANDREW JIP bull PHASE II

lt

bull IMPROVED LOAD AND RESISTANCE RECIPE

bull IMPROVED CALIBRATION PROCEDURE

bull MULTIPLE BIAS FACTORS

bull PROCEDURE TO INCLUDE EFFECT OF OTHER PLATFORMS

bull PROCEDURE TO USE MULTIPLE BIAS FACTORS

  • Structure Bookmarks
    • ANDREW PHASE II JOINT1 INDUSTRY PROJECT
    • PROGRESS MEETING NUMBER 1
    • ANDREW PHASE II JOINT INDUSTRY PROJECT PROGRESS MEETING NUMBER 1 NOVEMBER 16 AGENDA
    • ANDREW PHASE II UST OF PARTICIPANTS
    • ANDREW PHASE I OBJECTIVES AND RESULTS
    • ANDREW PHASE II PROJECT OBJECTIVES
    • SUBSEQUENT PROJECTS
    • ANDREW PHASE II PROJECT TEAM
    • ANDREW PHASE II UPDATE MODELING middotAND ANALYSIS RECIPE
    • -ANDREW PHASE II CONDUCTOR RESISTANCE
    • ANDREW PHASE Il WA VE FORCE PROFILE
    • ANDREW PHASE II WA VE FORCE PROFILEmiddot
    • ANDREW PHASE II BRACE MODELING
    • ANDREW PHASE II FOUNDATIONS
    • ANDREW PHASE II middot middot JOINT MODELING
    • ANDREW PHASE Il JOINT MODELING OPTIONS DISCUSSION
    • ANDREW PHASE II JOINT MODEJJNG OPTIONS ILLUSTRATIONS
    • ANDREW PHASE II JOINT MODELING PHASE II METHOD
    • ANDREW PHASE II ANDREW STORM CONDmONS UPDATED
    • Platform Selection bullbullbull1111111111111
    • BasisCriteria for Selection bullbullbull1111111111111
    • CALIBRATION
    • WHERE TO INTRODUCE BIAS FACTOR (B
    • PHASE I -KEY CONCLUSIONS
    • bull CALIBRATION ISSUES PHASE II
    • PHASE II WORK SCOPE (CALIBRATION)
    • SOFTWARE
    • SENSITIVITY STUDIES POSTERIOR DISTRIBUTION OF B
    • WEIGHTING APPROACH bull
    • INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON
    • CATEGORIZATION OF ADDITIONAL PLATFORMS
    • UPDATING OF CALIBRATION RESULTS
    • APPLICATIONS OF BIAS FACTORS
    • EXAMPLE USE OF BIAS FACTORS
    • EXAMPLE APPLICATION bull PHASE I PLATFORM ST 151 K
    • EXAMPLE USE OF BIAS FACTORS -PHASE II middot
    • BENEFITS OF ANDREW JIP bull PHASE II
Page 71:  · . ' ANDREW PHASE II PROJECT OBJECTIVES . ' • ·UPDATE MODELING AND ANALYSIS RECIPE· JOINTS ' ' ·BRACES WAVE PROFILE FOUNDATIONS . ' ' CONDUCTORS· • ANALVZE PLATFORMS USING

INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON

bull WATER DEPTH

bull bull YEAR OF INSTALLATIONDESIGN

bull PLATFORM ORIENTATION

bull PLATFORM LATITUDE AND LONGITUDE

bull KEY SKETCHES OF PLATFORM STRUCTURAL FRAMING

Number of Legs Plies Leg Batter Number of Conductors bull Deck Elevations (Including sub-cellar deck If any) Jacket Framing Elevations Brace type In Vertical Frames Sizes of Primary Members (leg pile vertical frames braces primary horizontals) Joint Details (can gap etc) LegPiie Annulus Grouting Status

bull middotPOST ANDREW INSPECTION REPORT Significant Wave Height Hs (ft)

bull INFORMATION ON DAMAGES PRIOR TO ANDREW

bull RESULTS FROM IN-HOUSE STUDIES IF AVAILABLE

bull ~RIGINAL DESIGN CRITERIA INFORMATION IF AVAILABLE

bull SOIL SHEAR STRENGTH PROFILE IF AVAILABLE

Platforms Identified for Evaluating Therr Likelihood of Falling Under Unexpected Survival Category

OPlRATOR

middotADA

BLOClt

llmucnJllE

NAME

YEAll

INSTAUJD

WA1ER

DBPTll

DISTANCE

JlllOM SHOllE

(MILU)

Bl

(M)

AMOCO 111113

STl11

WDI075

A

bull D

St

17

120

1n

44

20

10 I

CllllVllQfl STlllO c 61 110 21 11 STIUO D Q 1110 21 11

STl134 p Q 137 29 11

ST0151 G 61 137 11

ST0151 I 3 121 11

ST0151 J Q 140 11

STl151 It 137 11

ST0151 PJl001 Q UT 11

ST0151 PROD-2 Q UT 11

ST017 A 3 140 35 11

WDIUT c 65 214 34 10

llXXON STI055 II 51 6T 14 I

STl15 A t3 34

WDI073

WD0073

Amiddotshy Q

1111 1111

22

22

I

I

WDI073 c 1n 22 I

- WDOCl93-shy II

65 1110 2T A 3 194 31 MOBU 111121

SS IOl7

A B

501

50

50

31

T

I

T 6

MtlllPllY SS114 B T 50 T T

PlllllIPS SS0149 A 5 55 33 T

SAMllDAN STln A 107 T T

SBllLL WDl103 A 65 223 2T WDl103 B 65 221 2T WD0104 c 65 221 2T WD0104 D 6T 20 2T

UNOCAL SS020I

SS 020I SSl20I

SS 0209

SS0209

II p

PURll

A

B

3

n 61

103

97

97

95

100

34

34

34

34 34

bullbull bullbullbull

- SS0209 D u 95 34 I

- SS 1209 SSl253

SS0253

G A

c

6T

2

100

165

1T5

34

54

54

bullbull I

I I

I I

SOUTH MARCH

~LAND

N

SOUTH PaTOI ENE ISWltD AREA

MMS Defined Limits ror Offshore Structures Requiring Inspections Following Andrew

t---Jbulln3 T25bull I bullTtl

bull1$1Htl1Abull bull1SIK1779bull

~

middot-DaTA AREA~T

CRANO rsLE

AREA bull1300

bullll4W

=shy

bullPath of Hurricane Andrew with Platforms Used In Calibration

Likelihood Function vs b All Success Case STISI K STIJOQ ST134W WD90A MCJll MCJ97

I 0000

0 9000

0 8000

07000

-Bo6000

tiiii 0 5000 -

04000

0 3000

0 2000

01000

0 0000

0 05 25

ri1rt middot-middot-middot-middot-middot-

I 5 2

b

---11 middot151K(blsu~ccsI

- bull - bull - fk 130Q(bfU~lCSS)

lie bull134W(blsuccc]

lk-WD90A[blsultCS]

- - - lk middotMC397(bfUtlCS]

--o-- lk-MCJll(bbuctcsJ

--- ll (bl 6 out of 6 Mlttcs case]

_

( Likelihood Functions middotSuccess Cases - p - 1 c I

CATEGORIZATION OF ADDITIONAL PLATFORMS

I I I

bull CATEGORIES

Unexpected Survivals

Expected Survivals

bull APPROACH

Identify Corresponding Platform(s) Evaluated In the JIP

Determine Max Wave Height at Platform Location

Evaluate Whether Wave Hit the Deck

middot - Evaluate Likelihood of R and S to Differ from the Base Cases

Review Joint Information

Assess Special Features

bull IDENTIFY NUMBER OF PLATFORMS IN EACH CATEGORY

I

UPDATING OF CALIBRATION RESULTS

bull INCLUDE EFFECT OF ADDITIONAL PLATFORMS ON bullBbull

Consider Increased 1Uncertalntles In R and S

Develop bullNeW Ukellhood Functions for Different Categories

middotLet bullxbull Number of Ukely Unexpected Survival Cases and y Number of Ukely Expected Survival Cases

Determine Combined Likelihood Function lk (b Iadditional platforms) middot

bull REVISED LIKELIHOOD FUNCTION

lk(b I all cases) = lk(b I 9 platforms) middot x lk(b Ix platforms) x lk(b I y platforms)

bull ESTABLISH REVISED BIAS FACTOR B

f8 (b) u f8 (b) lk (b I all cases)

II

t t ~

shy

APPLICATIONS OF BIAS FACTORS

ii APPLICABLE TO PARTICULAR LOAD AND RESISTANCE RECIPE

bull USE BY THE API COMMITTEES AND MMS

Criteria Revisions

Recommended Design Guidelines Revisions bull

bullbull REASSESSMENT STUDIES AND DECISION MAKING

middot

bull RISK ASSESSMENT STUDIES BY THE OPERATORS

Individual Platforms

Fleet of Platforms

middot IMR Program Implementation

bullbull NOT TO BE USED IN DESIGN

I I fl J

EXAMPLE USE OF BIAS FACTORS

bull PHASE I EXAMPLE APPLICATION

bull EVALUATION OF ANNUAL PROBABILITY OF FAILURE

Establish Annual Maximum Seastate Data

Determine Coefficients to Define Base Shear

Determine Ultimate Capacity of Platform

Determine Probablllty of Failure Versus bullbbull

-bull DETERMINE UPDATED PROBABILITY OF FAILURE

Pf - P [S gt R]

Pf I P [~ gt Rb I b] f 8 (b) db

Pr - J P [S gt Rb I b] f bull 8 (b) db

bull TOTAL PROBABILITIES OF FAILURE

Total Pf - 1 bull Il [1 middotPf] - l Pf bull

Total Pf - 1 bull Il [1middot Pf] - l Pfmiddot

Total Pr = 1 - n [1middot PfbullJ = l Pfshy

I I I

EXAMPLE APPLICATION bull PHASE I

PLATFORM ST 151 K

bull GULF OF MEXICO

bull WATER DEPTH = 137 FT

bull ANNUAL MAXIMUM HS MEDIAN 175 ft COV 031

bull ASSUMED STORM DURATION 3 hoursmiddot

bull ASSUMED NUMBER OF WAVES 800

bull MEDIAN HS FOR STORM HOURS 1 2 AND 3

bull MEDIAN CURRENT VALUE bull

bull DETERMINE ULTIMATE CAPACITY FOR THREE DIRECTIONS

bull DETERMINE ANNUAL FAILURE PROBABILITIES

Broadsidemiddot Diagonal End-on

0037 0025 0007

0071 0044 0018

0022 0016 0004

DETERMINE TOTAL ANNUAL PROBABILITY OF FAILURE

Pf - 007

Pf - 013

Pr - 004

bull

EXAMPLE USE OF BIAS FACTORS - PHASE II middot

bull DETERMINE PREDICTIVE PROBABILITY OF FAILURE

Case - 1

Pf _ S S Pf (b b2) fa1a2 (b b2) db db2

g bull -1 (b1 R I S) - 1

92 - (b2 R2 I S) - 1

Case -2

Pf s ~ [bJ f11 (b) db

Pf [bJ p [g lt 0)

g - bull 1 (b1 R1 I S) - 1

1

bull1) l

I

BENEFITS OF ANDREW JIP bull PHASE II

lt

bull IMPROVED LOAD AND RESISTANCE RECIPE

bull IMPROVED CALIBRATION PROCEDURE

bull MULTIPLE BIAS FACTORS

bull PROCEDURE TO INCLUDE EFFECT OF OTHER PLATFORMS

bull PROCEDURE TO USE MULTIPLE BIAS FACTORS

  • Structure Bookmarks
    • ANDREW PHASE II JOINT1 INDUSTRY PROJECT
    • PROGRESS MEETING NUMBER 1
    • ANDREW PHASE II JOINT INDUSTRY PROJECT PROGRESS MEETING NUMBER 1 NOVEMBER 16 AGENDA
    • ANDREW PHASE II UST OF PARTICIPANTS
    • ANDREW PHASE I OBJECTIVES AND RESULTS
    • ANDREW PHASE II PROJECT OBJECTIVES
    • SUBSEQUENT PROJECTS
    • ANDREW PHASE II PROJECT TEAM
    • ANDREW PHASE II UPDATE MODELING middotAND ANALYSIS RECIPE
    • -ANDREW PHASE II CONDUCTOR RESISTANCE
    • ANDREW PHASE Il WA VE FORCE PROFILE
    • ANDREW PHASE II WA VE FORCE PROFILEmiddot
    • ANDREW PHASE II BRACE MODELING
    • ANDREW PHASE II FOUNDATIONS
    • ANDREW PHASE II middot middot JOINT MODELING
    • ANDREW PHASE Il JOINT MODELING OPTIONS DISCUSSION
    • ANDREW PHASE II JOINT MODEJJNG OPTIONS ILLUSTRATIONS
    • ANDREW PHASE II JOINT MODELING PHASE II METHOD
    • ANDREW PHASE II ANDREW STORM CONDmONS UPDATED
    • Platform Selection bullbullbull1111111111111
    • BasisCriteria for Selection bullbullbull1111111111111
    • CALIBRATION
    • WHERE TO INTRODUCE BIAS FACTOR (B
    • PHASE I -KEY CONCLUSIONS
    • bull CALIBRATION ISSUES PHASE II
    • PHASE II WORK SCOPE (CALIBRATION)
    • SOFTWARE
    • SENSITIVITY STUDIES POSTERIOR DISTRIBUTION OF B
    • WEIGHTING APPROACH bull
    • INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON
    • CATEGORIZATION OF ADDITIONAL PLATFORMS
    • UPDATING OF CALIBRATION RESULTS
    • APPLICATIONS OF BIAS FACTORS
    • EXAMPLE USE OF BIAS FACTORS
    • EXAMPLE APPLICATION bull PHASE I PLATFORM ST 151 K
    • EXAMPLE USE OF BIAS FACTORS -PHASE II middot
    • BENEFITS OF ANDREW JIP bull PHASE II
Page 72:  · . ' ANDREW PHASE II PROJECT OBJECTIVES . ' • ·UPDATE MODELING AND ANALYSIS RECIPE· JOINTS ' ' ·BRACES WAVE PROFILE FOUNDATIONS . ' ' CONDUCTORS· • ANALVZE PLATFORMS USING

Platforms Identified for Evaluating Therr Likelihood of Falling Under Unexpected Survival Category

OPlRATOR

middotADA

BLOClt

llmucnJllE

NAME

YEAll

INSTAUJD

WA1ER

DBPTll

DISTANCE

JlllOM SHOllE

(MILU)

Bl

(M)

AMOCO 111113

STl11

WDI075

A

bull D

St

17

120

1n

44

20

10 I

CllllVllQfl STlllO c 61 110 21 11 STIUO D Q 1110 21 11

STl134 p Q 137 29 11

ST0151 G 61 137 11

ST0151 I 3 121 11

ST0151 J Q 140 11

STl151 It 137 11

ST0151 PJl001 Q UT 11

ST0151 PROD-2 Q UT 11

ST017 A 3 140 35 11

WDIUT c 65 214 34 10

llXXON STI055 II 51 6T 14 I

STl15 A t3 34

WDI073

WD0073

Amiddotshy Q

1111 1111

22

22

I

I

WDI073 c 1n 22 I

- WDOCl93-shy II

65 1110 2T A 3 194 31 MOBU 111121

SS IOl7

A B

501

50

50

31

T

I

T 6

MtlllPllY SS114 B T 50 T T

PlllllIPS SS0149 A 5 55 33 T

SAMllDAN STln A 107 T T

SBllLL WDl103 A 65 223 2T WDl103 B 65 221 2T WD0104 c 65 221 2T WD0104 D 6T 20 2T

UNOCAL SS020I

SS 020I SSl20I

SS 0209

SS0209

II p

PURll

A

B

3

n 61

103

97

97

95

100

34

34

34

34 34

bullbull bullbullbull

- SS0209 D u 95 34 I

- SS 1209 SSl253

SS0253

G A

c

6T

2

100

165

1T5

34

54

54

bullbull I

I I

I I

SOUTH MARCH

~LAND

N

SOUTH PaTOI ENE ISWltD AREA

MMS Defined Limits ror Offshore Structures Requiring Inspections Following Andrew

t---Jbulln3 T25bull I bullTtl

bull1$1Htl1Abull bull1SIK1779bull

~

middot-DaTA AREA~T

CRANO rsLE

AREA bull1300

bullll4W

=shy

bullPath of Hurricane Andrew with Platforms Used In Calibration

Likelihood Function vs b All Success Case STISI K STIJOQ ST134W WD90A MCJll MCJ97

I 0000

0 9000

0 8000

07000

-Bo6000

tiiii 0 5000 -

04000

0 3000

0 2000

01000

0 0000

0 05 25

ri1rt middot-middot-middot-middot-middot-

I 5 2

b

---11 middot151K(blsu~ccsI

- bull - bull - fk 130Q(bfU~lCSS)

lie bull134W(blsuccc]

lk-WD90A[blsultCS]

- - - lk middotMC397(bfUtlCS]

--o-- lk-MCJll(bbuctcsJ

--- ll (bl 6 out of 6 Mlttcs case]

_

( Likelihood Functions middotSuccess Cases - p - 1 c I

CATEGORIZATION OF ADDITIONAL PLATFORMS

I I I

bull CATEGORIES

Unexpected Survivals

Expected Survivals

bull APPROACH

Identify Corresponding Platform(s) Evaluated In the JIP

Determine Max Wave Height at Platform Location

Evaluate Whether Wave Hit the Deck

middot - Evaluate Likelihood of R and S to Differ from the Base Cases

Review Joint Information

Assess Special Features

bull IDENTIFY NUMBER OF PLATFORMS IN EACH CATEGORY

I

UPDATING OF CALIBRATION RESULTS

bull INCLUDE EFFECT OF ADDITIONAL PLATFORMS ON bullBbull

Consider Increased 1Uncertalntles In R and S

Develop bullNeW Ukellhood Functions for Different Categories

middotLet bullxbull Number of Ukely Unexpected Survival Cases and y Number of Ukely Expected Survival Cases

Determine Combined Likelihood Function lk (b Iadditional platforms) middot

bull REVISED LIKELIHOOD FUNCTION

lk(b I all cases) = lk(b I 9 platforms) middot x lk(b Ix platforms) x lk(b I y platforms)

bull ESTABLISH REVISED BIAS FACTOR B

f8 (b) u f8 (b) lk (b I all cases)

II

t t ~

shy

APPLICATIONS OF BIAS FACTORS

ii APPLICABLE TO PARTICULAR LOAD AND RESISTANCE RECIPE

bull USE BY THE API COMMITTEES AND MMS

Criteria Revisions

Recommended Design Guidelines Revisions bull

bullbull REASSESSMENT STUDIES AND DECISION MAKING

middot

bull RISK ASSESSMENT STUDIES BY THE OPERATORS

Individual Platforms

Fleet of Platforms

middot IMR Program Implementation

bullbull NOT TO BE USED IN DESIGN

I I fl J

EXAMPLE USE OF BIAS FACTORS

bull PHASE I EXAMPLE APPLICATION

bull EVALUATION OF ANNUAL PROBABILITY OF FAILURE

Establish Annual Maximum Seastate Data

Determine Coefficients to Define Base Shear

Determine Ultimate Capacity of Platform

Determine Probablllty of Failure Versus bullbbull

-bull DETERMINE UPDATED PROBABILITY OF FAILURE

Pf - P [S gt R]

Pf I P [~ gt Rb I b] f 8 (b) db

Pr - J P [S gt Rb I b] f bull 8 (b) db

bull TOTAL PROBABILITIES OF FAILURE

Total Pf - 1 bull Il [1 middotPf] - l Pf bull

Total Pf - 1 bull Il [1middot Pf] - l Pfmiddot

Total Pr = 1 - n [1middot PfbullJ = l Pfshy

I I I

EXAMPLE APPLICATION bull PHASE I

PLATFORM ST 151 K

bull GULF OF MEXICO

bull WATER DEPTH = 137 FT

bull ANNUAL MAXIMUM HS MEDIAN 175 ft COV 031

bull ASSUMED STORM DURATION 3 hoursmiddot

bull ASSUMED NUMBER OF WAVES 800

bull MEDIAN HS FOR STORM HOURS 1 2 AND 3

bull MEDIAN CURRENT VALUE bull

bull DETERMINE ULTIMATE CAPACITY FOR THREE DIRECTIONS

bull DETERMINE ANNUAL FAILURE PROBABILITIES

Broadsidemiddot Diagonal End-on

0037 0025 0007

0071 0044 0018

0022 0016 0004

DETERMINE TOTAL ANNUAL PROBABILITY OF FAILURE

Pf - 007

Pf - 013

Pr - 004

bull

EXAMPLE USE OF BIAS FACTORS - PHASE II middot

bull DETERMINE PREDICTIVE PROBABILITY OF FAILURE

Case - 1

Pf _ S S Pf (b b2) fa1a2 (b b2) db db2

g bull -1 (b1 R I S) - 1

92 - (b2 R2 I S) - 1

Case -2

Pf s ~ [bJ f11 (b) db

Pf [bJ p [g lt 0)

g - bull 1 (b1 R1 I S) - 1

1

bull1) l

I

BENEFITS OF ANDREW JIP bull PHASE II

lt

bull IMPROVED LOAD AND RESISTANCE RECIPE

bull IMPROVED CALIBRATION PROCEDURE

bull MULTIPLE BIAS FACTORS

bull PROCEDURE TO INCLUDE EFFECT OF OTHER PLATFORMS

bull PROCEDURE TO USE MULTIPLE BIAS FACTORS

  • Structure Bookmarks
    • ANDREW PHASE II JOINT1 INDUSTRY PROJECT
    • PROGRESS MEETING NUMBER 1
    • ANDREW PHASE II JOINT INDUSTRY PROJECT PROGRESS MEETING NUMBER 1 NOVEMBER 16 AGENDA
    • ANDREW PHASE II UST OF PARTICIPANTS
    • ANDREW PHASE I OBJECTIVES AND RESULTS
    • ANDREW PHASE II PROJECT OBJECTIVES
    • SUBSEQUENT PROJECTS
    • ANDREW PHASE II PROJECT TEAM
    • ANDREW PHASE II UPDATE MODELING middotAND ANALYSIS RECIPE
    • -ANDREW PHASE II CONDUCTOR RESISTANCE
    • ANDREW PHASE Il WA VE FORCE PROFILE
    • ANDREW PHASE II WA VE FORCE PROFILEmiddot
    • ANDREW PHASE II BRACE MODELING
    • ANDREW PHASE II FOUNDATIONS
    • ANDREW PHASE II middot middot JOINT MODELING
    • ANDREW PHASE Il JOINT MODELING OPTIONS DISCUSSION
    • ANDREW PHASE II JOINT MODEJJNG OPTIONS ILLUSTRATIONS
    • ANDREW PHASE II JOINT MODELING PHASE II METHOD
    • ANDREW PHASE II ANDREW STORM CONDmONS UPDATED
    • Platform Selection bullbullbull1111111111111
    • BasisCriteria for Selection bullbullbull1111111111111
    • CALIBRATION
    • WHERE TO INTRODUCE BIAS FACTOR (B
    • PHASE I -KEY CONCLUSIONS
    • bull CALIBRATION ISSUES PHASE II
    • PHASE II WORK SCOPE (CALIBRATION)
    • SOFTWARE
    • SENSITIVITY STUDIES POSTERIOR DISTRIBUTION OF B
    • WEIGHTING APPROACH bull
    • INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON
    • CATEGORIZATION OF ADDITIONAL PLATFORMS
    • UPDATING OF CALIBRATION RESULTS
    • APPLICATIONS OF BIAS FACTORS
    • EXAMPLE USE OF BIAS FACTORS
    • EXAMPLE APPLICATION bull PHASE I PLATFORM ST 151 K
    • EXAMPLE USE OF BIAS FACTORS -PHASE II middot
    • BENEFITS OF ANDREW JIP bull PHASE II
Page 73:  · . ' ANDREW PHASE II PROJECT OBJECTIVES . ' • ·UPDATE MODELING AND ANALYSIS RECIPE· JOINTS ' ' ·BRACES WAVE PROFILE FOUNDATIONS . ' ' CONDUCTORS· • ANALVZE PLATFORMS USING

I I

I I

SOUTH MARCH

~LAND

N

SOUTH PaTOI ENE ISWltD AREA

MMS Defined Limits ror Offshore Structures Requiring Inspections Following Andrew

t---Jbulln3 T25bull I bullTtl

bull1$1Htl1Abull bull1SIK1779bull

~

middot-DaTA AREA~T

CRANO rsLE

AREA bull1300

bullll4W

=shy

bullPath of Hurricane Andrew with Platforms Used In Calibration

Likelihood Function vs b All Success Case STISI K STIJOQ ST134W WD90A MCJll MCJ97

I 0000

0 9000

0 8000

07000

-Bo6000

tiiii 0 5000 -

04000

0 3000

0 2000

01000

0 0000

0 05 25

ri1rt middot-middot-middot-middot-middot-

I 5 2

b

---11 middot151K(blsu~ccsI

- bull - bull - fk 130Q(bfU~lCSS)

lie bull134W(blsuccc]

lk-WD90A[blsultCS]

- - - lk middotMC397(bfUtlCS]

--o-- lk-MCJll(bbuctcsJ

--- ll (bl 6 out of 6 Mlttcs case]

_

( Likelihood Functions middotSuccess Cases - p - 1 c I

CATEGORIZATION OF ADDITIONAL PLATFORMS

I I I

bull CATEGORIES

Unexpected Survivals

Expected Survivals

bull APPROACH

Identify Corresponding Platform(s) Evaluated In the JIP

Determine Max Wave Height at Platform Location

Evaluate Whether Wave Hit the Deck

middot - Evaluate Likelihood of R and S to Differ from the Base Cases

Review Joint Information

Assess Special Features

bull IDENTIFY NUMBER OF PLATFORMS IN EACH CATEGORY

I

UPDATING OF CALIBRATION RESULTS

bull INCLUDE EFFECT OF ADDITIONAL PLATFORMS ON bullBbull

Consider Increased 1Uncertalntles In R and S

Develop bullNeW Ukellhood Functions for Different Categories

middotLet bullxbull Number of Ukely Unexpected Survival Cases and y Number of Ukely Expected Survival Cases

Determine Combined Likelihood Function lk (b Iadditional platforms) middot

bull REVISED LIKELIHOOD FUNCTION

lk(b I all cases) = lk(b I 9 platforms) middot x lk(b Ix platforms) x lk(b I y platforms)

bull ESTABLISH REVISED BIAS FACTOR B

f8 (b) u f8 (b) lk (b I all cases)

II

t t ~

shy

APPLICATIONS OF BIAS FACTORS

ii APPLICABLE TO PARTICULAR LOAD AND RESISTANCE RECIPE

bull USE BY THE API COMMITTEES AND MMS

Criteria Revisions

Recommended Design Guidelines Revisions bull

bullbull REASSESSMENT STUDIES AND DECISION MAKING

middot

bull RISK ASSESSMENT STUDIES BY THE OPERATORS

Individual Platforms

Fleet of Platforms

middot IMR Program Implementation

bullbull NOT TO BE USED IN DESIGN

I I fl J

EXAMPLE USE OF BIAS FACTORS

bull PHASE I EXAMPLE APPLICATION

bull EVALUATION OF ANNUAL PROBABILITY OF FAILURE

Establish Annual Maximum Seastate Data

Determine Coefficients to Define Base Shear

Determine Ultimate Capacity of Platform

Determine Probablllty of Failure Versus bullbbull

-bull DETERMINE UPDATED PROBABILITY OF FAILURE

Pf - P [S gt R]

Pf I P [~ gt Rb I b] f 8 (b) db

Pr - J P [S gt Rb I b] f bull 8 (b) db

bull TOTAL PROBABILITIES OF FAILURE

Total Pf - 1 bull Il [1 middotPf] - l Pf bull

Total Pf - 1 bull Il [1middot Pf] - l Pfmiddot

Total Pr = 1 - n [1middot PfbullJ = l Pfshy

I I I

EXAMPLE APPLICATION bull PHASE I

PLATFORM ST 151 K

bull GULF OF MEXICO

bull WATER DEPTH = 137 FT

bull ANNUAL MAXIMUM HS MEDIAN 175 ft COV 031

bull ASSUMED STORM DURATION 3 hoursmiddot

bull ASSUMED NUMBER OF WAVES 800

bull MEDIAN HS FOR STORM HOURS 1 2 AND 3

bull MEDIAN CURRENT VALUE bull

bull DETERMINE ULTIMATE CAPACITY FOR THREE DIRECTIONS

bull DETERMINE ANNUAL FAILURE PROBABILITIES

Broadsidemiddot Diagonal End-on

0037 0025 0007

0071 0044 0018

0022 0016 0004

DETERMINE TOTAL ANNUAL PROBABILITY OF FAILURE

Pf - 007

Pf - 013

Pr - 004

bull

EXAMPLE USE OF BIAS FACTORS - PHASE II middot

bull DETERMINE PREDICTIVE PROBABILITY OF FAILURE

Case - 1

Pf _ S S Pf (b b2) fa1a2 (b b2) db db2

g bull -1 (b1 R I S) - 1

92 - (b2 R2 I S) - 1

Case -2

Pf s ~ [bJ f11 (b) db

Pf [bJ p [g lt 0)

g - bull 1 (b1 R1 I S) - 1

1

bull1) l

I

BENEFITS OF ANDREW JIP bull PHASE II

lt

bull IMPROVED LOAD AND RESISTANCE RECIPE

bull IMPROVED CALIBRATION PROCEDURE

bull MULTIPLE BIAS FACTORS

bull PROCEDURE TO INCLUDE EFFECT OF OTHER PLATFORMS

bull PROCEDURE TO USE MULTIPLE BIAS FACTORS

  • Structure Bookmarks
    • ANDREW PHASE II JOINT1 INDUSTRY PROJECT
    • PROGRESS MEETING NUMBER 1
    • ANDREW PHASE II JOINT INDUSTRY PROJECT PROGRESS MEETING NUMBER 1 NOVEMBER 16 AGENDA
    • ANDREW PHASE II UST OF PARTICIPANTS
    • ANDREW PHASE I OBJECTIVES AND RESULTS
    • ANDREW PHASE II PROJECT OBJECTIVES
    • SUBSEQUENT PROJECTS
    • ANDREW PHASE II PROJECT TEAM
    • ANDREW PHASE II UPDATE MODELING middotAND ANALYSIS RECIPE
    • -ANDREW PHASE II CONDUCTOR RESISTANCE
    • ANDREW PHASE Il WA VE FORCE PROFILE
    • ANDREW PHASE II WA VE FORCE PROFILEmiddot
    • ANDREW PHASE II BRACE MODELING
    • ANDREW PHASE II FOUNDATIONS
    • ANDREW PHASE II middot middot JOINT MODELING
    • ANDREW PHASE Il JOINT MODELING OPTIONS DISCUSSION
    • ANDREW PHASE II JOINT MODEJJNG OPTIONS ILLUSTRATIONS
    • ANDREW PHASE II JOINT MODELING PHASE II METHOD
    • ANDREW PHASE II ANDREW STORM CONDmONS UPDATED
    • Platform Selection bullbullbull1111111111111
    • BasisCriteria for Selection bullbullbull1111111111111
    • CALIBRATION
    • WHERE TO INTRODUCE BIAS FACTOR (B
    • PHASE I -KEY CONCLUSIONS
    • bull CALIBRATION ISSUES PHASE II
    • PHASE II WORK SCOPE (CALIBRATION)
    • SOFTWARE
    • SENSITIVITY STUDIES POSTERIOR DISTRIBUTION OF B
    • WEIGHTING APPROACH bull
    • INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON
    • CATEGORIZATION OF ADDITIONAL PLATFORMS
    • UPDATING OF CALIBRATION RESULTS
    • APPLICATIONS OF BIAS FACTORS
    • EXAMPLE USE OF BIAS FACTORS
    • EXAMPLE APPLICATION bull PHASE I PLATFORM ST 151 K
    • EXAMPLE USE OF BIAS FACTORS -PHASE II middot
    • BENEFITS OF ANDREW JIP bull PHASE II
Page 74:  · . ' ANDREW PHASE II PROJECT OBJECTIVES . ' • ·UPDATE MODELING AND ANALYSIS RECIPE· JOINTS ' ' ·BRACES WAVE PROFILE FOUNDATIONS . ' ' CONDUCTORS· • ANALVZE PLATFORMS USING

Likelihood Function vs b All Success Case STISI K STIJOQ ST134W WD90A MCJll MCJ97

I 0000

0 9000

0 8000

07000

-Bo6000

tiiii 0 5000 -

04000

0 3000

0 2000

01000

0 0000

0 05 25

ri1rt middot-middot-middot-middot-middot-

I 5 2

b

---11 middot151K(blsu~ccsI

- bull - bull - fk 130Q(bfU~lCSS)

lie bull134W(blsuccc]

lk-WD90A[blsultCS]

- - - lk middotMC397(bfUtlCS]

--o-- lk-MCJll(bbuctcsJ

--- ll (bl 6 out of 6 Mlttcs case]

_

( Likelihood Functions middotSuccess Cases - p - 1 c I

CATEGORIZATION OF ADDITIONAL PLATFORMS

I I I

bull CATEGORIES

Unexpected Survivals

Expected Survivals

bull APPROACH

Identify Corresponding Platform(s) Evaluated In the JIP

Determine Max Wave Height at Platform Location

Evaluate Whether Wave Hit the Deck

middot - Evaluate Likelihood of R and S to Differ from the Base Cases

Review Joint Information

Assess Special Features

bull IDENTIFY NUMBER OF PLATFORMS IN EACH CATEGORY

I

UPDATING OF CALIBRATION RESULTS

bull INCLUDE EFFECT OF ADDITIONAL PLATFORMS ON bullBbull

Consider Increased 1Uncertalntles In R and S

Develop bullNeW Ukellhood Functions for Different Categories

middotLet bullxbull Number of Ukely Unexpected Survival Cases and y Number of Ukely Expected Survival Cases

Determine Combined Likelihood Function lk (b Iadditional platforms) middot

bull REVISED LIKELIHOOD FUNCTION

lk(b I all cases) = lk(b I 9 platforms) middot x lk(b Ix platforms) x lk(b I y platforms)

bull ESTABLISH REVISED BIAS FACTOR B

f8 (b) u f8 (b) lk (b I all cases)

II

t t ~

shy

APPLICATIONS OF BIAS FACTORS

ii APPLICABLE TO PARTICULAR LOAD AND RESISTANCE RECIPE

bull USE BY THE API COMMITTEES AND MMS

Criteria Revisions

Recommended Design Guidelines Revisions bull

bullbull REASSESSMENT STUDIES AND DECISION MAKING

middot

bull RISK ASSESSMENT STUDIES BY THE OPERATORS

Individual Platforms

Fleet of Platforms

middot IMR Program Implementation

bullbull NOT TO BE USED IN DESIGN

I I fl J

EXAMPLE USE OF BIAS FACTORS

bull PHASE I EXAMPLE APPLICATION

bull EVALUATION OF ANNUAL PROBABILITY OF FAILURE

Establish Annual Maximum Seastate Data

Determine Coefficients to Define Base Shear

Determine Ultimate Capacity of Platform

Determine Probablllty of Failure Versus bullbbull

-bull DETERMINE UPDATED PROBABILITY OF FAILURE

Pf - P [S gt R]

Pf I P [~ gt Rb I b] f 8 (b) db

Pr - J P [S gt Rb I b] f bull 8 (b) db

bull TOTAL PROBABILITIES OF FAILURE

Total Pf - 1 bull Il [1 middotPf] - l Pf bull

Total Pf - 1 bull Il [1middot Pf] - l Pfmiddot

Total Pr = 1 - n [1middot PfbullJ = l Pfshy

I I I

EXAMPLE APPLICATION bull PHASE I

PLATFORM ST 151 K

bull GULF OF MEXICO

bull WATER DEPTH = 137 FT

bull ANNUAL MAXIMUM HS MEDIAN 175 ft COV 031

bull ASSUMED STORM DURATION 3 hoursmiddot

bull ASSUMED NUMBER OF WAVES 800

bull MEDIAN HS FOR STORM HOURS 1 2 AND 3

bull MEDIAN CURRENT VALUE bull

bull DETERMINE ULTIMATE CAPACITY FOR THREE DIRECTIONS

bull DETERMINE ANNUAL FAILURE PROBABILITIES

Broadsidemiddot Diagonal End-on

0037 0025 0007

0071 0044 0018

0022 0016 0004

DETERMINE TOTAL ANNUAL PROBABILITY OF FAILURE

Pf - 007

Pf - 013

Pr - 004

bull

EXAMPLE USE OF BIAS FACTORS - PHASE II middot

bull DETERMINE PREDICTIVE PROBABILITY OF FAILURE

Case - 1

Pf _ S S Pf (b b2) fa1a2 (b b2) db db2

g bull -1 (b1 R I S) - 1

92 - (b2 R2 I S) - 1

Case -2

Pf s ~ [bJ f11 (b) db

Pf [bJ p [g lt 0)

g - bull 1 (b1 R1 I S) - 1

1

bull1) l

I

BENEFITS OF ANDREW JIP bull PHASE II

lt

bull IMPROVED LOAD AND RESISTANCE RECIPE

bull IMPROVED CALIBRATION PROCEDURE

bull MULTIPLE BIAS FACTORS

bull PROCEDURE TO INCLUDE EFFECT OF OTHER PLATFORMS

bull PROCEDURE TO USE MULTIPLE BIAS FACTORS

  • Structure Bookmarks
    • ANDREW PHASE II JOINT1 INDUSTRY PROJECT
    • PROGRESS MEETING NUMBER 1
    • ANDREW PHASE II JOINT INDUSTRY PROJECT PROGRESS MEETING NUMBER 1 NOVEMBER 16 AGENDA
    • ANDREW PHASE II UST OF PARTICIPANTS
    • ANDREW PHASE I OBJECTIVES AND RESULTS
    • ANDREW PHASE II PROJECT OBJECTIVES
    • SUBSEQUENT PROJECTS
    • ANDREW PHASE II PROJECT TEAM
    • ANDREW PHASE II UPDATE MODELING middotAND ANALYSIS RECIPE
    • -ANDREW PHASE II CONDUCTOR RESISTANCE
    • ANDREW PHASE Il WA VE FORCE PROFILE
    • ANDREW PHASE II WA VE FORCE PROFILEmiddot
    • ANDREW PHASE II BRACE MODELING
    • ANDREW PHASE II FOUNDATIONS
    • ANDREW PHASE II middot middot JOINT MODELING
    • ANDREW PHASE Il JOINT MODELING OPTIONS DISCUSSION
    • ANDREW PHASE II JOINT MODEJJNG OPTIONS ILLUSTRATIONS
    • ANDREW PHASE II JOINT MODELING PHASE II METHOD
    • ANDREW PHASE II ANDREW STORM CONDmONS UPDATED
    • Platform Selection bullbullbull1111111111111
    • BasisCriteria for Selection bullbullbull1111111111111
    • CALIBRATION
    • WHERE TO INTRODUCE BIAS FACTOR (B
    • PHASE I -KEY CONCLUSIONS
    • bull CALIBRATION ISSUES PHASE II
    • PHASE II WORK SCOPE (CALIBRATION)
    • SOFTWARE
    • SENSITIVITY STUDIES POSTERIOR DISTRIBUTION OF B
    • WEIGHTING APPROACH bull
    • INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON
    • CATEGORIZATION OF ADDITIONAL PLATFORMS
    • UPDATING OF CALIBRATION RESULTS
    • APPLICATIONS OF BIAS FACTORS
    • EXAMPLE USE OF BIAS FACTORS
    • EXAMPLE APPLICATION bull PHASE I PLATFORM ST 151 K
    • EXAMPLE USE OF BIAS FACTORS -PHASE II middot
    • BENEFITS OF ANDREW JIP bull PHASE II
Page 75:  · . ' ANDREW PHASE II PROJECT OBJECTIVES . ' • ·UPDATE MODELING AND ANALYSIS RECIPE· JOINTS ' ' ·BRACES WAVE PROFILE FOUNDATIONS . ' ' CONDUCTORS· • ANALVZE PLATFORMS USING

CATEGORIZATION OF ADDITIONAL PLATFORMS

I I I

bull CATEGORIES

Unexpected Survivals

Expected Survivals

bull APPROACH

Identify Corresponding Platform(s) Evaluated In the JIP

Determine Max Wave Height at Platform Location

Evaluate Whether Wave Hit the Deck

middot - Evaluate Likelihood of R and S to Differ from the Base Cases

Review Joint Information

Assess Special Features

bull IDENTIFY NUMBER OF PLATFORMS IN EACH CATEGORY

I

UPDATING OF CALIBRATION RESULTS

bull INCLUDE EFFECT OF ADDITIONAL PLATFORMS ON bullBbull

Consider Increased 1Uncertalntles In R and S

Develop bullNeW Ukellhood Functions for Different Categories

middotLet bullxbull Number of Ukely Unexpected Survival Cases and y Number of Ukely Expected Survival Cases

Determine Combined Likelihood Function lk (b Iadditional platforms) middot

bull REVISED LIKELIHOOD FUNCTION

lk(b I all cases) = lk(b I 9 platforms) middot x lk(b Ix platforms) x lk(b I y platforms)

bull ESTABLISH REVISED BIAS FACTOR B

f8 (b) u f8 (b) lk (b I all cases)

II

t t ~

shy

APPLICATIONS OF BIAS FACTORS

ii APPLICABLE TO PARTICULAR LOAD AND RESISTANCE RECIPE

bull USE BY THE API COMMITTEES AND MMS

Criteria Revisions

Recommended Design Guidelines Revisions bull

bullbull REASSESSMENT STUDIES AND DECISION MAKING

middot

bull RISK ASSESSMENT STUDIES BY THE OPERATORS

Individual Platforms

Fleet of Platforms

middot IMR Program Implementation

bullbull NOT TO BE USED IN DESIGN

I I fl J

EXAMPLE USE OF BIAS FACTORS

bull PHASE I EXAMPLE APPLICATION

bull EVALUATION OF ANNUAL PROBABILITY OF FAILURE

Establish Annual Maximum Seastate Data

Determine Coefficients to Define Base Shear

Determine Ultimate Capacity of Platform

Determine Probablllty of Failure Versus bullbbull

-bull DETERMINE UPDATED PROBABILITY OF FAILURE

Pf - P [S gt R]

Pf I P [~ gt Rb I b] f 8 (b) db

Pr - J P [S gt Rb I b] f bull 8 (b) db

bull TOTAL PROBABILITIES OF FAILURE

Total Pf - 1 bull Il [1 middotPf] - l Pf bull

Total Pf - 1 bull Il [1middot Pf] - l Pfmiddot

Total Pr = 1 - n [1middot PfbullJ = l Pfshy

I I I

EXAMPLE APPLICATION bull PHASE I

PLATFORM ST 151 K

bull GULF OF MEXICO

bull WATER DEPTH = 137 FT

bull ANNUAL MAXIMUM HS MEDIAN 175 ft COV 031

bull ASSUMED STORM DURATION 3 hoursmiddot

bull ASSUMED NUMBER OF WAVES 800

bull MEDIAN HS FOR STORM HOURS 1 2 AND 3

bull MEDIAN CURRENT VALUE bull

bull DETERMINE ULTIMATE CAPACITY FOR THREE DIRECTIONS

bull DETERMINE ANNUAL FAILURE PROBABILITIES

Broadsidemiddot Diagonal End-on

0037 0025 0007

0071 0044 0018

0022 0016 0004

DETERMINE TOTAL ANNUAL PROBABILITY OF FAILURE

Pf - 007

Pf - 013

Pr - 004

bull

EXAMPLE USE OF BIAS FACTORS - PHASE II middot

bull DETERMINE PREDICTIVE PROBABILITY OF FAILURE

Case - 1

Pf _ S S Pf (b b2) fa1a2 (b b2) db db2

g bull -1 (b1 R I S) - 1

92 - (b2 R2 I S) - 1

Case -2

Pf s ~ [bJ f11 (b) db

Pf [bJ p [g lt 0)

g - bull 1 (b1 R1 I S) - 1

1

bull1) l

I

BENEFITS OF ANDREW JIP bull PHASE II

lt

bull IMPROVED LOAD AND RESISTANCE RECIPE

bull IMPROVED CALIBRATION PROCEDURE

bull MULTIPLE BIAS FACTORS

bull PROCEDURE TO INCLUDE EFFECT OF OTHER PLATFORMS

bull PROCEDURE TO USE MULTIPLE BIAS FACTORS

  • Structure Bookmarks
    • ANDREW PHASE II JOINT1 INDUSTRY PROJECT
    • PROGRESS MEETING NUMBER 1
    • ANDREW PHASE II JOINT INDUSTRY PROJECT PROGRESS MEETING NUMBER 1 NOVEMBER 16 AGENDA
    • ANDREW PHASE II UST OF PARTICIPANTS
    • ANDREW PHASE I OBJECTIVES AND RESULTS
    • ANDREW PHASE II PROJECT OBJECTIVES
    • SUBSEQUENT PROJECTS
    • ANDREW PHASE II PROJECT TEAM
    • ANDREW PHASE II UPDATE MODELING middotAND ANALYSIS RECIPE
    • -ANDREW PHASE II CONDUCTOR RESISTANCE
    • ANDREW PHASE Il WA VE FORCE PROFILE
    • ANDREW PHASE II WA VE FORCE PROFILEmiddot
    • ANDREW PHASE II BRACE MODELING
    • ANDREW PHASE II FOUNDATIONS
    • ANDREW PHASE II middot middot JOINT MODELING
    • ANDREW PHASE Il JOINT MODELING OPTIONS DISCUSSION
    • ANDREW PHASE II JOINT MODEJJNG OPTIONS ILLUSTRATIONS
    • ANDREW PHASE II JOINT MODELING PHASE II METHOD
    • ANDREW PHASE II ANDREW STORM CONDmONS UPDATED
    • Platform Selection bullbullbull1111111111111
    • BasisCriteria for Selection bullbullbull1111111111111
    • CALIBRATION
    • WHERE TO INTRODUCE BIAS FACTOR (B
    • PHASE I -KEY CONCLUSIONS
    • bull CALIBRATION ISSUES PHASE II
    • PHASE II WORK SCOPE (CALIBRATION)
    • SOFTWARE
    • SENSITIVITY STUDIES POSTERIOR DISTRIBUTION OF B
    • WEIGHTING APPROACH bull
    • INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON
    • CATEGORIZATION OF ADDITIONAL PLATFORMS
    • UPDATING OF CALIBRATION RESULTS
    • APPLICATIONS OF BIAS FACTORS
    • EXAMPLE USE OF BIAS FACTORS
    • EXAMPLE APPLICATION bull PHASE I PLATFORM ST 151 K
    • EXAMPLE USE OF BIAS FACTORS -PHASE II middot
    • BENEFITS OF ANDREW JIP bull PHASE II
Page 76:  · . ' ANDREW PHASE II PROJECT OBJECTIVES . ' • ·UPDATE MODELING AND ANALYSIS RECIPE· JOINTS ' ' ·BRACES WAVE PROFILE FOUNDATIONS . ' ' CONDUCTORS· • ANALVZE PLATFORMS USING

I

UPDATING OF CALIBRATION RESULTS

bull INCLUDE EFFECT OF ADDITIONAL PLATFORMS ON bullBbull

Consider Increased 1Uncertalntles In R and S

Develop bullNeW Ukellhood Functions for Different Categories

middotLet bullxbull Number of Ukely Unexpected Survival Cases and y Number of Ukely Expected Survival Cases

Determine Combined Likelihood Function lk (b Iadditional platforms) middot

bull REVISED LIKELIHOOD FUNCTION

lk(b I all cases) = lk(b I 9 platforms) middot x lk(b Ix platforms) x lk(b I y platforms)

bull ESTABLISH REVISED BIAS FACTOR B

f8 (b) u f8 (b) lk (b I all cases)

II

t t ~

shy

APPLICATIONS OF BIAS FACTORS

ii APPLICABLE TO PARTICULAR LOAD AND RESISTANCE RECIPE

bull USE BY THE API COMMITTEES AND MMS

Criteria Revisions

Recommended Design Guidelines Revisions bull

bullbull REASSESSMENT STUDIES AND DECISION MAKING

middot

bull RISK ASSESSMENT STUDIES BY THE OPERATORS

Individual Platforms

Fleet of Platforms

middot IMR Program Implementation

bullbull NOT TO BE USED IN DESIGN

I I fl J

EXAMPLE USE OF BIAS FACTORS

bull PHASE I EXAMPLE APPLICATION

bull EVALUATION OF ANNUAL PROBABILITY OF FAILURE

Establish Annual Maximum Seastate Data

Determine Coefficients to Define Base Shear

Determine Ultimate Capacity of Platform

Determine Probablllty of Failure Versus bullbbull

-bull DETERMINE UPDATED PROBABILITY OF FAILURE

Pf - P [S gt R]

Pf I P [~ gt Rb I b] f 8 (b) db

Pr - J P [S gt Rb I b] f bull 8 (b) db

bull TOTAL PROBABILITIES OF FAILURE

Total Pf - 1 bull Il [1 middotPf] - l Pf bull

Total Pf - 1 bull Il [1middot Pf] - l Pfmiddot

Total Pr = 1 - n [1middot PfbullJ = l Pfshy

I I I

EXAMPLE APPLICATION bull PHASE I

PLATFORM ST 151 K

bull GULF OF MEXICO

bull WATER DEPTH = 137 FT

bull ANNUAL MAXIMUM HS MEDIAN 175 ft COV 031

bull ASSUMED STORM DURATION 3 hoursmiddot

bull ASSUMED NUMBER OF WAVES 800

bull MEDIAN HS FOR STORM HOURS 1 2 AND 3

bull MEDIAN CURRENT VALUE bull

bull DETERMINE ULTIMATE CAPACITY FOR THREE DIRECTIONS

bull DETERMINE ANNUAL FAILURE PROBABILITIES

Broadsidemiddot Diagonal End-on

0037 0025 0007

0071 0044 0018

0022 0016 0004

DETERMINE TOTAL ANNUAL PROBABILITY OF FAILURE

Pf - 007

Pf - 013

Pr - 004

bull

EXAMPLE USE OF BIAS FACTORS - PHASE II middot

bull DETERMINE PREDICTIVE PROBABILITY OF FAILURE

Case - 1

Pf _ S S Pf (b b2) fa1a2 (b b2) db db2

g bull -1 (b1 R I S) - 1

92 - (b2 R2 I S) - 1

Case -2

Pf s ~ [bJ f11 (b) db

Pf [bJ p [g lt 0)

g - bull 1 (b1 R1 I S) - 1

1

bull1) l

I

BENEFITS OF ANDREW JIP bull PHASE II

lt

bull IMPROVED LOAD AND RESISTANCE RECIPE

bull IMPROVED CALIBRATION PROCEDURE

bull MULTIPLE BIAS FACTORS

bull PROCEDURE TO INCLUDE EFFECT OF OTHER PLATFORMS

bull PROCEDURE TO USE MULTIPLE BIAS FACTORS

  • Structure Bookmarks
    • ANDREW PHASE II JOINT1 INDUSTRY PROJECT
    • PROGRESS MEETING NUMBER 1
    • ANDREW PHASE II JOINT INDUSTRY PROJECT PROGRESS MEETING NUMBER 1 NOVEMBER 16 AGENDA
    • ANDREW PHASE II UST OF PARTICIPANTS
    • ANDREW PHASE I OBJECTIVES AND RESULTS
    • ANDREW PHASE II PROJECT OBJECTIVES
    • SUBSEQUENT PROJECTS
    • ANDREW PHASE II PROJECT TEAM
    • ANDREW PHASE II UPDATE MODELING middotAND ANALYSIS RECIPE
    • -ANDREW PHASE II CONDUCTOR RESISTANCE
    • ANDREW PHASE Il WA VE FORCE PROFILE
    • ANDREW PHASE II WA VE FORCE PROFILEmiddot
    • ANDREW PHASE II BRACE MODELING
    • ANDREW PHASE II FOUNDATIONS
    • ANDREW PHASE II middot middot JOINT MODELING
    • ANDREW PHASE Il JOINT MODELING OPTIONS DISCUSSION
    • ANDREW PHASE II JOINT MODEJJNG OPTIONS ILLUSTRATIONS
    • ANDREW PHASE II JOINT MODELING PHASE II METHOD
    • ANDREW PHASE II ANDREW STORM CONDmONS UPDATED
    • Platform Selection bullbullbull1111111111111
    • BasisCriteria for Selection bullbullbull1111111111111
    • CALIBRATION
    • WHERE TO INTRODUCE BIAS FACTOR (B
    • PHASE I -KEY CONCLUSIONS
    • bull CALIBRATION ISSUES PHASE II
    • PHASE II WORK SCOPE (CALIBRATION)
    • SOFTWARE
    • SENSITIVITY STUDIES POSTERIOR DISTRIBUTION OF B
    • WEIGHTING APPROACH bull
    • INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON
    • CATEGORIZATION OF ADDITIONAL PLATFORMS
    • UPDATING OF CALIBRATION RESULTS
    • APPLICATIONS OF BIAS FACTORS
    • EXAMPLE USE OF BIAS FACTORS
    • EXAMPLE APPLICATION bull PHASE I PLATFORM ST 151 K
    • EXAMPLE USE OF BIAS FACTORS -PHASE II middot
    • BENEFITS OF ANDREW JIP bull PHASE II
Page 77:  · . ' ANDREW PHASE II PROJECT OBJECTIVES . ' • ·UPDATE MODELING AND ANALYSIS RECIPE· JOINTS ' ' ·BRACES WAVE PROFILE FOUNDATIONS . ' ' CONDUCTORS· • ANALVZE PLATFORMS USING

II

t t ~

shy

APPLICATIONS OF BIAS FACTORS

ii APPLICABLE TO PARTICULAR LOAD AND RESISTANCE RECIPE

bull USE BY THE API COMMITTEES AND MMS

Criteria Revisions

Recommended Design Guidelines Revisions bull

bullbull REASSESSMENT STUDIES AND DECISION MAKING

middot

bull RISK ASSESSMENT STUDIES BY THE OPERATORS

Individual Platforms

Fleet of Platforms

middot IMR Program Implementation

bullbull NOT TO BE USED IN DESIGN

I I fl J

EXAMPLE USE OF BIAS FACTORS

bull PHASE I EXAMPLE APPLICATION

bull EVALUATION OF ANNUAL PROBABILITY OF FAILURE

Establish Annual Maximum Seastate Data

Determine Coefficients to Define Base Shear

Determine Ultimate Capacity of Platform

Determine Probablllty of Failure Versus bullbbull

-bull DETERMINE UPDATED PROBABILITY OF FAILURE

Pf - P [S gt R]

Pf I P [~ gt Rb I b] f 8 (b) db

Pr - J P [S gt Rb I b] f bull 8 (b) db

bull TOTAL PROBABILITIES OF FAILURE

Total Pf - 1 bull Il [1 middotPf] - l Pf bull

Total Pf - 1 bull Il [1middot Pf] - l Pfmiddot

Total Pr = 1 - n [1middot PfbullJ = l Pfshy

I I I

EXAMPLE APPLICATION bull PHASE I

PLATFORM ST 151 K

bull GULF OF MEXICO

bull WATER DEPTH = 137 FT

bull ANNUAL MAXIMUM HS MEDIAN 175 ft COV 031

bull ASSUMED STORM DURATION 3 hoursmiddot

bull ASSUMED NUMBER OF WAVES 800

bull MEDIAN HS FOR STORM HOURS 1 2 AND 3

bull MEDIAN CURRENT VALUE bull

bull DETERMINE ULTIMATE CAPACITY FOR THREE DIRECTIONS

bull DETERMINE ANNUAL FAILURE PROBABILITIES

Broadsidemiddot Diagonal End-on

0037 0025 0007

0071 0044 0018

0022 0016 0004

DETERMINE TOTAL ANNUAL PROBABILITY OF FAILURE

Pf - 007

Pf - 013

Pr - 004

bull

EXAMPLE USE OF BIAS FACTORS - PHASE II middot

bull DETERMINE PREDICTIVE PROBABILITY OF FAILURE

Case - 1

Pf _ S S Pf (b b2) fa1a2 (b b2) db db2

g bull -1 (b1 R I S) - 1

92 - (b2 R2 I S) - 1

Case -2

Pf s ~ [bJ f11 (b) db

Pf [bJ p [g lt 0)

g - bull 1 (b1 R1 I S) - 1

1

bull1) l

I

BENEFITS OF ANDREW JIP bull PHASE II

lt

bull IMPROVED LOAD AND RESISTANCE RECIPE

bull IMPROVED CALIBRATION PROCEDURE

bull MULTIPLE BIAS FACTORS

bull PROCEDURE TO INCLUDE EFFECT OF OTHER PLATFORMS

bull PROCEDURE TO USE MULTIPLE BIAS FACTORS

  • Structure Bookmarks
    • ANDREW PHASE II JOINT1 INDUSTRY PROJECT
    • PROGRESS MEETING NUMBER 1
    • ANDREW PHASE II JOINT INDUSTRY PROJECT PROGRESS MEETING NUMBER 1 NOVEMBER 16 AGENDA
    • ANDREW PHASE II UST OF PARTICIPANTS
    • ANDREW PHASE I OBJECTIVES AND RESULTS
    • ANDREW PHASE II PROJECT OBJECTIVES
    • SUBSEQUENT PROJECTS
    • ANDREW PHASE II PROJECT TEAM
    • ANDREW PHASE II UPDATE MODELING middotAND ANALYSIS RECIPE
    • -ANDREW PHASE II CONDUCTOR RESISTANCE
    • ANDREW PHASE Il WA VE FORCE PROFILE
    • ANDREW PHASE II WA VE FORCE PROFILEmiddot
    • ANDREW PHASE II BRACE MODELING
    • ANDREW PHASE II FOUNDATIONS
    • ANDREW PHASE II middot middot JOINT MODELING
    • ANDREW PHASE Il JOINT MODELING OPTIONS DISCUSSION
    • ANDREW PHASE II JOINT MODEJJNG OPTIONS ILLUSTRATIONS
    • ANDREW PHASE II JOINT MODELING PHASE II METHOD
    • ANDREW PHASE II ANDREW STORM CONDmONS UPDATED
    • Platform Selection bullbullbull1111111111111
    • BasisCriteria for Selection bullbullbull1111111111111
    • CALIBRATION
    • WHERE TO INTRODUCE BIAS FACTOR (B
    • PHASE I -KEY CONCLUSIONS
    • bull CALIBRATION ISSUES PHASE II
    • PHASE II WORK SCOPE (CALIBRATION)
    • SOFTWARE
    • SENSITIVITY STUDIES POSTERIOR DISTRIBUTION OF B
    • WEIGHTING APPROACH bull
    • INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON
    • CATEGORIZATION OF ADDITIONAL PLATFORMS
    • UPDATING OF CALIBRATION RESULTS
    • APPLICATIONS OF BIAS FACTORS
    • EXAMPLE USE OF BIAS FACTORS
    • EXAMPLE APPLICATION bull PHASE I PLATFORM ST 151 K
    • EXAMPLE USE OF BIAS FACTORS -PHASE II middot
    • BENEFITS OF ANDREW JIP bull PHASE II
Page 78:  · . ' ANDREW PHASE II PROJECT OBJECTIVES . ' • ·UPDATE MODELING AND ANALYSIS RECIPE· JOINTS ' ' ·BRACES WAVE PROFILE FOUNDATIONS . ' ' CONDUCTORS· • ANALVZE PLATFORMS USING

I I fl J

EXAMPLE USE OF BIAS FACTORS

bull PHASE I EXAMPLE APPLICATION

bull EVALUATION OF ANNUAL PROBABILITY OF FAILURE

Establish Annual Maximum Seastate Data

Determine Coefficients to Define Base Shear

Determine Ultimate Capacity of Platform

Determine Probablllty of Failure Versus bullbbull

-bull DETERMINE UPDATED PROBABILITY OF FAILURE

Pf - P [S gt R]

Pf I P [~ gt Rb I b] f 8 (b) db

Pr - J P [S gt Rb I b] f bull 8 (b) db

bull TOTAL PROBABILITIES OF FAILURE

Total Pf - 1 bull Il [1 middotPf] - l Pf bull

Total Pf - 1 bull Il [1middot Pf] - l Pfmiddot

Total Pr = 1 - n [1middot PfbullJ = l Pfshy

I I I

EXAMPLE APPLICATION bull PHASE I

PLATFORM ST 151 K

bull GULF OF MEXICO

bull WATER DEPTH = 137 FT

bull ANNUAL MAXIMUM HS MEDIAN 175 ft COV 031

bull ASSUMED STORM DURATION 3 hoursmiddot

bull ASSUMED NUMBER OF WAVES 800

bull MEDIAN HS FOR STORM HOURS 1 2 AND 3

bull MEDIAN CURRENT VALUE bull

bull DETERMINE ULTIMATE CAPACITY FOR THREE DIRECTIONS

bull DETERMINE ANNUAL FAILURE PROBABILITIES

Broadsidemiddot Diagonal End-on

0037 0025 0007

0071 0044 0018

0022 0016 0004

DETERMINE TOTAL ANNUAL PROBABILITY OF FAILURE

Pf - 007

Pf - 013

Pr - 004

bull

EXAMPLE USE OF BIAS FACTORS - PHASE II middot

bull DETERMINE PREDICTIVE PROBABILITY OF FAILURE

Case - 1

Pf _ S S Pf (b b2) fa1a2 (b b2) db db2

g bull -1 (b1 R I S) - 1

92 - (b2 R2 I S) - 1

Case -2

Pf s ~ [bJ f11 (b) db

Pf [bJ p [g lt 0)

g - bull 1 (b1 R1 I S) - 1

1

bull1) l

I

BENEFITS OF ANDREW JIP bull PHASE II

lt

bull IMPROVED LOAD AND RESISTANCE RECIPE

bull IMPROVED CALIBRATION PROCEDURE

bull MULTIPLE BIAS FACTORS

bull PROCEDURE TO INCLUDE EFFECT OF OTHER PLATFORMS

bull PROCEDURE TO USE MULTIPLE BIAS FACTORS

  • Structure Bookmarks
    • ANDREW PHASE II JOINT1 INDUSTRY PROJECT
    • PROGRESS MEETING NUMBER 1
    • ANDREW PHASE II JOINT INDUSTRY PROJECT PROGRESS MEETING NUMBER 1 NOVEMBER 16 AGENDA
    • ANDREW PHASE II UST OF PARTICIPANTS
    • ANDREW PHASE I OBJECTIVES AND RESULTS
    • ANDREW PHASE II PROJECT OBJECTIVES
    • SUBSEQUENT PROJECTS
    • ANDREW PHASE II PROJECT TEAM
    • ANDREW PHASE II UPDATE MODELING middotAND ANALYSIS RECIPE
    • -ANDREW PHASE II CONDUCTOR RESISTANCE
    • ANDREW PHASE Il WA VE FORCE PROFILE
    • ANDREW PHASE II WA VE FORCE PROFILEmiddot
    • ANDREW PHASE II BRACE MODELING
    • ANDREW PHASE II FOUNDATIONS
    • ANDREW PHASE II middot middot JOINT MODELING
    • ANDREW PHASE Il JOINT MODELING OPTIONS DISCUSSION
    • ANDREW PHASE II JOINT MODEJJNG OPTIONS ILLUSTRATIONS
    • ANDREW PHASE II JOINT MODELING PHASE II METHOD
    • ANDREW PHASE II ANDREW STORM CONDmONS UPDATED
    • Platform Selection bullbullbull1111111111111
    • BasisCriteria for Selection bullbullbull1111111111111
    • CALIBRATION
    • WHERE TO INTRODUCE BIAS FACTOR (B
    • PHASE I -KEY CONCLUSIONS
    • bull CALIBRATION ISSUES PHASE II
    • PHASE II WORK SCOPE (CALIBRATION)
    • SOFTWARE
    • SENSITIVITY STUDIES POSTERIOR DISTRIBUTION OF B
    • WEIGHTING APPROACH bull
    • INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON
    • CATEGORIZATION OF ADDITIONAL PLATFORMS
    • UPDATING OF CALIBRATION RESULTS
    • APPLICATIONS OF BIAS FACTORS
    • EXAMPLE USE OF BIAS FACTORS
    • EXAMPLE APPLICATION bull PHASE I PLATFORM ST 151 K
    • EXAMPLE USE OF BIAS FACTORS -PHASE II middot
    • BENEFITS OF ANDREW JIP bull PHASE II
Page 79:  · . ' ANDREW PHASE II PROJECT OBJECTIVES . ' • ·UPDATE MODELING AND ANALYSIS RECIPE· JOINTS ' ' ·BRACES WAVE PROFILE FOUNDATIONS . ' ' CONDUCTORS· • ANALVZE PLATFORMS USING

I I I

EXAMPLE APPLICATION bull PHASE I

PLATFORM ST 151 K

bull GULF OF MEXICO

bull WATER DEPTH = 137 FT

bull ANNUAL MAXIMUM HS MEDIAN 175 ft COV 031

bull ASSUMED STORM DURATION 3 hoursmiddot

bull ASSUMED NUMBER OF WAVES 800

bull MEDIAN HS FOR STORM HOURS 1 2 AND 3

bull MEDIAN CURRENT VALUE bull

bull DETERMINE ULTIMATE CAPACITY FOR THREE DIRECTIONS

bull DETERMINE ANNUAL FAILURE PROBABILITIES

Broadsidemiddot Diagonal End-on

0037 0025 0007

0071 0044 0018

0022 0016 0004

DETERMINE TOTAL ANNUAL PROBABILITY OF FAILURE

Pf - 007

Pf - 013

Pr - 004

bull

EXAMPLE USE OF BIAS FACTORS - PHASE II middot

bull DETERMINE PREDICTIVE PROBABILITY OF FAILURE

Case - 1

Pf _ S S Pf (b b2) fa1a2 (b b2) db db2

g bull -1 (b1 R I S) - 1

92 - (b2 R2 I S) - 1

Case -2

Pf s ~ [bJ f11 (b) db

Pf [bJ p [g lt 0)

g - bull 1 (b1 R1 I S) - 1

1

bull1) l

I

BENEFITS OF ANDREW JIP bull PHASE II

lt

bull IMPROVED LOAD AND RESISTANCE RECIPE

bull IMPROVED CALIBRATION PROCEDURE

bull MULTIPLE BIAS FACTORS

bull PROCEDURE TO INCLUDE EFFECT OF OTHER PLATFORMS

bull PROCEDURE TO USE MULTIPLE BIAS FACTORS

  • Structure Bookmarks
    • ANDREW PHASE II JOINT1 INDUSTRY PROJECT
    • PROGRESS MEETING NUMBER 1
    • ANDREW PHASE II JOINT INDUSTRY PROJECT PROGRESS MEETING NUMBER 1 NOVEMBER 16 AGENDA
    • ANDREW PHASE II UST OF PARTICIPANTS
    • ANDREW PHASE I OBJECTIVES AND RESULTS
    • ANDREW PHASE II PROJECT OBJECTIVES
    • SUBSEQUENT PROJECTS
    • ANDREW PHASE II PROJECT TEAM
    • ANDREW PHASE II UPDATE MODELING middotAND ANALYSIS RECIPE
    • -ANDREW PHASE II CONDUCTOR RESISTANCE
    • ANDREW PHASE Il WA VE FORCE PROFILE
    • ANDREW PHASE II WA VE FORCE PROFILEmiddot
    • ANDREW PHASE II BRACE MODELING
    • ANDREW PHASE II FOUNDATIONS
    • ANDREW PHASE II middot middot JOINT MODELING
    • ANDREW PHASE Il JOINT MODELING OPTIONS DISCUSSION
    • ANDREW PHASE II JOINT MODEJJNG OPTIONS ILLUSTRATIONS
    • ANDREW PHASE II JOINT MODELING PHASE II METHOD
    • ANDREW PHASE II ANDREW STORM CONDmONS UPDATED
    • Platform Selection bullbullbull1111111111111
    • BasisCriteria for Selection bullbullbull1111111111111
    • CALIBRATION
    • WHERE TO INTRODUCE BIAS FACTOR (B
    • PHASE I -KEY CONCLUSIONS
    • bull CALIBRATION ISSUES PHASE II
    • PHASE II WORK SCOPE (CALIBRATION)
    • SOFTWARE
    • SENSITIVITY STUDIES POSTERIOR DISTRIBUTION OF B
    • WEIGHTING APPROACH bull
    • INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON
    • CATEGORIZATION OF ADDITIONAL PLATFORMS
    • UPDATING OF CALIBRATION RESULTS
    • APPLICATIONS OF BIAS FACTORS
    • EXAMPLE USE OF BIAS FACTORS
    • EXAMPLE APPLICATION bull PHASE I PLATFORM ST 151 K
    • EXAMPLE USE OF BIAS FACTORS -PHASE II middot
    • BENEFITS OF ANDREW JIP bull PHASE II
Page 80:  · . ' ANDREW PHASE II PROJECT OBJECTIVES . ' • ·UPDATE MODELING AND ANALYSIS RECIPE· JOINTS ' ' ·BRACES WAVE PROFILE FOUNDATIONS . ' ' CONDUCTORS· • ANALVZE PLATFORMS USING

EXAMPLE USE OF BIAS FACTORS - PHASE II middot

bull DETERMINE PREDICTIVE PROBABILITY OF FAILURE

Case - 1

Pf _ S S Pf (b b2) fa1a2 (b b2) db db2

g bull -1 (b1 R I S) - 1

92 - (b2 R2 I S) - 1

Case -2

Pf s ~ [bJ f11 (b) db

Pf [bJ p [g lt 0)

g - bull 1 (b1 R1 I S) - 1

1

bull1) l

I

BENEFITS OF ANDREW JIP bull PHASE II

lt

bull IMPROVED LOAD AND RESISTANCE RECIPE

bull IMPROVED CALIBRATION PROCEDURE

bull MULTIPLE BIAS FACTORS

bull PROCEDURE TO INCLUDE EFFECT OF OTHER PLATFORMS

bull PROCEDURE TO USE MULTIPLE BIAS FACTORS

  • Structure Bookmarks
    • ANDREW PHASE II JOINT1 INDUSTRY PROJECT
    • PROGRESS MEETING NUMBER 1
    • ANDREW PHASE II JOINT INDUSTRY PROJECT PROGRESS MEETING NUMBER 1 NOVEMBER 16 AGENDA
    • ANDREW PHASE II UST OF PARTICIPANTS
    • ANDREW PHASE I OBJECTIVES AND RESULTS
    • ANDREW PHASE II PROJECT OBJECTIVES
    • SUBSEQUENT PROJECTS
    • ANDREW PHASE II PROJECT TEAM
    • ANDREW PHASE II UPDATE MODELING middotAND ANALYSIS RECIPE
    • -ANDREW PHASE II CONDUCTOR RESISTANCE
    • ANDREW PHASE Il WA VE FORCE PROFILE
    • ANDREW PHASE II WA VE FORCE PROFILEmiddot
    • ANDREW PHASE II BRACE MODELING
    • ANDREW PHASE II FOUNDATIONS
    • ANDREW PHASE II middot middot JOINT MODELING
    • ANDREW PHASE Il JOINT MODELING OPTIONS DISCUSSION
    • ANDREW PHASE II JOINT MODEJJNG OPTIONS ILLUSTRATIONS
    • ANDREW PHASE II JOINT MODELING PHASE II METHOD
    • ANDREW PHASE II ANDREW STORM CONDmONS UPDATED
    • Platform Selection bullbullbull1111111111111
    • BasisCriteria for Selection bullbullbull1111111111111
    • CALIBRATION
    • WHERE TO INTRODUCE BIAS FACTOR (B
    • PHASE I -KEY CONCLUSIONS
    • bull CALIBRATION ISSUES PHASE II
    • PHASE II WORK SCOPE (CALIBRATION)
    • SOFTWARE
    • SENSITIVITY STUDIES POSTERIOR DISTRIBUTION OF B
    • WEIGHTING APPROACH bull
    • INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON
    • CATEGORIZATION OF ADDITIONAL PLATFORMS
    • UPDATING OF CALIBRATION RESULTS
    • APPLICATIONS OF BIAS FACTORS
    • EXAMPLE USE OF BIAS FACTORS
    • EXAMPLE APPLICATION bull PHASE I PLATFORM ST 151 K
    • EXAMPLE USE OF BIAS FACTORS -PHASE II middot
    • BENEFITS OF ANDREW JIP bull PHASE II
Page 81:  · . ' ANDREW PHASE II PROJECT OBJECTIVES . ' • ·UPDATE MODELING AND ANALYSIS RECIPE· JOINTS ' ' ·BRACES WAVE PROFILE FOUNDATIONS . ' ' CONDUCTORS· • ANALVZE PLATFORMS USING

1

bull1) l

I

BENEFITS OF ANDREW JIP bull PHASE II

lt

bull IMPROVED LOAD AND RESISTANCE RECIPE

bull IMPROVED CALIBRATION PROCEDURE

bull MULTIPLE BIAS FACTORS

bull PROCEDURE TO INCLUDE EFFECT OF OTHER PLATFORMS

bull PROCEDURE TO USE MULTIPLE BIAS FACTORS

  • Structure Bookmarks
    • ANDREW PHASE II JOINT1 INDUSTRY PROJECT
    • PROGRESS MEETING NUMBER 1
    • ANDREW PHASE II JOINT INDUSTRY PROJECT PROGRESS MEETING NUMBER 1 NOVEMBER 16 AGENDA
    • ANDREW PHASE II UST OF PARTICIPANTS
    • ANDREW PHASE I OBJECTIVES AND RESULTS
    • ANDREW PHASE II PROJECT OBJECTIVES
    • SUBSEQUENT PROJECTS
    • ANDREW PHASE II PROJECT TEAM
    • ANDREW PHASE II UPDATE MODELING middotAND ANALYSIS RECIPE
    • -ANDREW PHASE II CONDUCTOR RESISTANCE
    • ANDREW PHASE Il WA VE FORCE PROFILE
    • ANDREW PHASE II WA VE FORCE PROFILEmiddot
    • ANDREW PHASE II BRACE MODELING
    • ANDREW PHASE II FOUNDATIONS
    • ANDREW PHASE II middot middot JOINT MODELING
    • ANDREW PHASE Il JOINT MODELING OPTIONS DISCUSSION
    • ANDREW PHASE II JOINT MODEJJNG OPTIONS ILLUSTRATIONS
    • ANDREW PHASE II JOINT MODELING PHASE II METHOD
    • ANDREW PHASE II ANDREW STORM CONDmONS UPDATED
    • Platform Selection bullbullbull1111111111111
    • BasisCriteria for Selection bullbullbull1111111111111
    • CALIBRATION
    • WHERE TO INTRODUCE BIAS FACTOR (B
    • PHASE I -KEY CONCLUSIONS
    • bull CALIBRATION ISSUES PHASE II
    • PHASE II WORK SCOPE (CALIBRATION)
    • SOFTWARE
    • SENSITIVITY STUDIES POSTERIOR DISTRIBUTION OF B
    • WEIGHTING APPROACH bull
    • INFORMATION REQUIRED (PRELIMINARY) FOR COMPARISON
    • CATEGORIZATION OF ADDITIONAL PLATFORMS
    • UPDATING OF CALIBRATION RESULTS
    • APPLICATIONS OF BIAS FACTORS
    • EXAMPLE USE OF BIAS FACTORS
    • EXAMPLE APPLICATION bull PHASE I PLATFORM ST 151 K
    • EXAMPLE USE OF BIAS FACTORS -PHASE II middot
    • BENEFITS OF ANDREW JIP bull PHASE II

Braces en bandages

Braces en bandages

Documents
BS 215-1 Aluminum Conductors & Aluminum Conductors Steel-Reinforced for Overhead Power Transmission (Aluminum Stranded Conductors)

BS 215-1 Aluminum Conductors & Aluminum Conductors Steel-Reinforced for Overhead Power Transmission (Aluminum Stranded Conductors)

Documents
BUCKLING RESTRAINED BRACES

BUCKLING RESTRAINED BRACES

Engineering
Buckling-Restrained Braces and Applications1 Chapter 1: Composition and history of Buckling‐restrained Braces 1 Buckling‐restrained Braces and Applications Buckling-Restrained

Buckling-Restrained Braces and Applications1 Chapter 1: Composition and history of Buckling‐restrained Braces 1 Buckling‐restrained Braces and Applications Buckling-Restrained

Documents
Aspen Vista Lower Spine Braces...Aspen Vista ™ Lower Spine Braces Vista Lower Spine braces are a part of Aspen’s Acute Restriction line. These braces offer comfort, fit, performance

Aspen Vista Lower Spine Braces...Aspen Vista ™ Lower Spine Braces Vista Lower Spine braces are a part of Aspen’s Acute Restriction line. These braces offer comfort, fit, performance

Documents
Invisible Braces Treatment | Orthodontic Braces bangalote

Invisible Braces Treatment | Orthodontic Braces bangalote

Healthcare
New Braces

New Braces

Technology
About Invisalign Braces and Cost of Clear Braces

About Invisalign Braces and Cost of Clear Braces

Health & Medicine
Snake braceS

Snake braceS

Documents
Electrical Stimulation Braces

Electrical Stimulation Braces

Documents
Braces las vegas

Braces las vegas

Business
Braces summerlin

Braces summerlin

Documents
BRACES Pictures

BRACES Pictures

Documents
Acknowledgments Buckling Restrained Braces in Buckling ... - CIE-500D Seminar 10-8.pdfBuckling Restrained Braces in Buckling Restrained Braces in Structural Fuse ApplicationStructural

Acknowledgments Buckling Restrained Braces in Buckling ... - CIE-500D Seminar 10-8.pdfBuckling Restrained Braces in Buckling Restrained Braces in Structural Fuse ApplicationStructural

Documents
Electricity Magnetism Lecture 9: Conductors and Capacitance Lecture 10 - Conductors and... · Electricity & Magnetism Lecture 9: Conductors and Capacitance ... The two conductors

Electricity Magnetism Lecture 9: Conductors and Capacitance Lecture 10 - Conductors and... · Electricity & Magnetism Lecture 9: Conductors and Capacitance ... The two conductors

Documents
ONiA, Adelaide—Lingual Braces

ONiA, Adelaide—Lingual Braces

Design
Invisible Braces Bangalore

Invisible Braces Bangalore

Healthcare
What are dental braces

What are dental braces

Healthcare
Conductors, Semi Conductors and Insulators

Conductors, Semi Conductors and Insulators

Documents
Conductors and Insulators Conductors - Appalachian State

Conductors and Insulators Conductors - Appalachian State

Documents
Truss...of truss (example ladder truss to triangular truss) BRACES Outside braces add additional stability to a truss structure. BRACES Inside braces con be used to increase structural

Truss...of truss (example ladder truss to triangular truss) BRACES Outside braces add additional stability to a truss structure. BRACES Inside braces con be used to increase structural

Documents
Conductors and non-conductors Conductors are materials that conduct electricity. Non-conductors or insulators are materials that do not conduct electricity

Conductors and non-conductors Conductors are materials that conduct electricity. Non-conductors or insulators are materials that do not conduct electricity

Documents
Braces vs. Aligners

Braces vs. Aligners

Health & Medicine
11.4 Conductors and Insulators - Weebly...11.4 Conductors and Insulators Scientists have categorized most materials into two categories: conductors and insulators (Figure 1). Conductors

11.4 Conductors and Insulators - Weebly...11.4 Conductors and Insulators Scientists have categorized most materials into two categories: conductors and insulators (Figure 1). Conductors

Documents
INCORPORATING BUCKLING RESTRAINED BRACES (BRB) AS …BUCKLING RESTRAINED BRACES Buckling restrained braces are generally constructed of a cruciform or rectangular steel core surrounded

INCORPORATING BUCKLING RESTRAINED BRACES (BRB) AS …BUCKLING RESTRAINED BRACES Buckling restrained braces are generally constructed of a cruciform or rectangular steel core surrounded

Documents
Dentallecurehospital.com/Dental.pdf · B – Lingual Braces C – Ceramic Braces a) CERAMIC NORMAL b) CERAMIC SELF-LIGATING D – Steel/Metal Braces . a) STEEL NORMAL ... patients

Dentallecurehospital.com/Dental.pdf · B – Lingual Braces C – Ceramic Braces a) CERAMIC NORMAL b) CERAMIC SELF-LIGATING D – Steel/Metal Braces . a) STEEL NORMAL ... patients

Documents
Cast and Braces

Cast and Braces

Documents
Dental braces slideshare

Dental braces slideshare

Health & Medicine
Inside Braces

Inside Braces

Health & Medicine
UK PETROCHEMICAL INDUSTRY BRACES FOR VARIOUS BREXIT …€¦ · UK PETROCHEMICAL INDUSTRY BRACES FOR ... BY YANA PALAGACHEVA NOVEMBER 2018 BREXIT UK PETROCHEMICAL INDUSTRY BRACES

UK PETROCHEMICAL INDUSTRY BRACES FOR VARIOUS BREXIT …€¦ · UK PETROCHEMICAL INDUSTRY BRACES FOR ... BY YANA PALAGACHEVA NOVEMBER 2018 BREXIT UK PETROCHEMICAL INDUSTRY BRACES

Documents