Jochen Schwarz & Holger MaiwaldBauhaus-Universität Weimar, Earthquake Damage Analysis Center
Damage and loss prediction model Damage and loss prediction model based on the vulnerability of building typesbased on the vulnerability of building types
4th International Symposium on Flood Defence:
Managing Flood Risk, Reliability and Vulnerability, Toronto, Ontario, Canada, May 6-8, 2008
Damage and loss prediction model based on the vulnerability of bDamage and loss prediction model based on the vulnerability of building typesuilding types
MotivationMotivation
Residential buildings
1.706 Mio. €
Forestry and agriculture79 Mio. €
Infrastructure other sectors
111 Mio. €Emergency
management136 Mio. €
Content529 Mio. €
Governmentalinfrastructure
928 Mio. €
Communalinfrastructure1.287 Mio. €
Commercial sector
1.420 Mio. €
Reasons for the study Reasons for the study
Damage and loss prediction model based on the vulnerability of bDamage and loss prediction model based on the vulnerability of building typesuilding types
Distribution of losses in Saxony after the flood 2002*
* after Sächsische Staatskanzlei (2002)
Reasons for the study Reasons for the study
Damage and loss prediction model based on the vulnerability of bDamage and loss prediction model based on the vulnerability of building typesuilding types
Building type: masonryDamage grade: D2
Primary influence factor:Inundation level
Building type: masonryDamage grade: D2-3 (slight settlements)
Primary influence factor:Inundation level
Damage and loss prediction model based on the vulnerability of bDamage and loss prediction model based on the vulnerability of building typesuilding types
Reasons for the studyReasons for the study
Damage and loss prediction model based on the vulnerability of bDamage and loss prediction model based on the vulnerability of building typesuilding types
In addition to Flood Action Side, the impact of structural parameters (Resistance Side) has to be considered.
Can the procedures developed for the Risk Analysis of Earthquakes be adapted to Flood Risk Analysis procedures?
If “yes”: Development of a Damage and Loss Prediction Model
based on an Engineering Evaluation System of Buildings
Reasons for the study Reasons for the study
Damage and loss prediction model based on the vulnerability of bDamage and loss prediction model based on the vulnerability of building typesuilding types
Building type: clayDamage grade: D5
Special influence factor:Flow velocity and circulations
Building type: masonry with clay mortarDamage grade: D5
Primary influence factor:Inundation levelFlow velocity
„Step 1“
Reasons for the study Reasons for the study
Damage and loss prediction model based on the vulnerability of bDamage and loss prediction model based on the vulnerability of building typesuilding types
Inundation level Flow velocity DurationBuilding type
Building type: clay Building type: masonry
Inundation levelFlow velocityDuration Building type
„Step 1 + 1/2“
Reasons for the study Reasons for the study
Damage and loss prediction model based on the vulnerability of bDamage and loss prediction model based on the vulnerability of building typesuilding types
Inundation level Flow velocityDuration Building type
Inundation levelFlow velocity Duration Building type
Building type: masonry (clay mortar)Building type: masonry
Basic elementsBasic elements
Damage and loss prediction model based on the vulnerability of bDamage and loss prediction model based on the vulnerability of building typesuilding types
0 20 40 60 80 100
Durchfeuchtungsgrad df [%]
Dam
age
Gra
de D
i
S1
1
2
3
4
5S8S7
S6
S10 S9
S5
S2S3
S4
Bauweise Überflu-tungshöhe
VerweilzeitFließ-geschwin-digkeit
Einflussparameter
Inundation level [%]
Building type
Inundation level
Duration Flow velocity
Influence parameter
Influence: high moderate low
First approachClassification of different damage cases (samples S1 - S10)
Basics of the StudyBasics of the StudyEngineering Evaluation System of BuildingsEngineering Evaluation System of Buildings
Damage and loss prediction model based on the vulnerability of bDamage and loss prediction model based on the vulnerability of building typesuilding types
Data baseData base
Damage and loss prediction model based on the vulnerability of bDamage and loss prediction model based on the vulnerability of building typesuilding types
Dataset 1: EDAC – questionnaireSaxony August 2002Baden Württemberg 1978 - 1994
Bavaria September 2005Saxony April 2006
Dataset 2: MEDIS – telephone interview
Basic elementsBasic elements
Damage and loss prediction model based on the vulnerability of bDamage and loss prediction model based on the vulnerability of building typesuilding types
×Demolition of building required
××Replacement of structural elements
×××Replacement of non-structural elements
×××Replacement of extension elements
××××Cleaning of penetrated elements
Rehabilitation measures
Collapse of major parts of the building×Partial failure of primary structural elements×××Settlements
××Major cracks / deformations in structural walls and slabs
××××Slight cracks in structural elements
××××Impressed doors and windows
Structural damage
××××Contamination (oil, chemicals)
×××××Pollution (mud, excrements …)Chemical damage
××××Penetration of supporting and non-structural walls and the floor slabs
Building physical damage
D5D4D3D2D1
Damage grade DiObserved effect / measureCriteria
„New Step“
Basic elements: Definition of Damage Grades Basic elements: Definition of Damage Grades
Damage and loss prediction model based on the vulnerability of bDamage and loss prediction model based on the vulnerability of building typesuilding types
Damage grade D1
Drawing Example
No structural damageSlight non-structural damage
Only penetration and pollution
Damage and loss prediction model based on the vulnerability of bDamage and loss prediction model based on the vulnerability of building typesuilding types
Damage grade D2
Basic elements: Definition of Damage Grades Basic elements: Definition of Damage Grades
Drawing Example
No to slight structural damageModerate non-structural damage
Slight cracks in supporting elementsImpressed doors and windowsContaminationReplacement of extension elements
Damage and loss prediction model based on the vulnerability of bDamage and loss prediction model based on the vulnerability of building typesuilding types
Damage grade D3
Basic elements: Definition of Damage Grades Basic elements: Definition of Damage Grades
Drawing Example
Moderate structural damageHeavy non-structural damage
Major cracks and/or deformations in supporting walls and slabsSettlementsReplacement of non-structural elements
Damage and loss prediction model based on the vulnerability of bDamage and loss prediction model based on the vulnerability of building typesuilding types
Damage grade D4
Basic elements: Definition of Damage Grades Basic elements: Definition of Damage Grades
Drawing Example
Heavy structural damageVery heavy non-structural damage
Structural collapse of supporting walls, slabsReplacement of sructural elements
Damage and loss prediction model based on the vulnerability of bDamage and loss prediction model based on the vulnerability of building typesuilding types
Drawing Example
Damage grade D5
Basic elements: Definition of Damage Grades Basic elements: Definition of Damage Grades
Very heavy structural damageVery heavy non-structural damage
Collapse of the building or of major parts of the buildingDemolition of building required
Damage and loss prediction model based on the vulnerability of bDamage and loss prediction model based on the vulnerability of building typesuilding types
Basic elements Basic elements
1
2
3
4
5
-3 -2 -1 0 1 2 3 4 5
Inundation level hgf [m]
Dam
age
grad
e D
i
1
2
3
4
5
-3 -2 -1 0 1 2 3 4 5
Inundation level hgf [m]
Dam
age
grad
e D
i
Dataset 1 „EDAC“ Dataset 2 „MEDIS“
Main building type (masonry) Other building types
Damage and loss prediction model based on the vulnerability of bDamage and loss prediction model based on the vulnerability of building typesuilding types
Damage and loss prediction model based on the vulnerability of bDamage and loss prediction model based on the vulnerability of building typesuilding types
Basic elements Basic elements
Damage and loss prediction model based on the vulnerability of bDamage and loss prediction model based on the vulnerability of building typesuilding types
Dataset „EDAC“Dataset „EDAC + MEDIS“
Masonry buildings
Damage and loss prediction model based on the vulnerability of bDamage and loss prediction model based on the vulnerability of building typesuilding types
Basic elements Basic elements
Damage and loss prediction model based on the vulnerability of bDamage and loss prediction model based on the vulnerability of building typesuilding types
Classification of building types in vulnerability classes
Most likely vulnerability classProbable range of scatterRange of less probable, exceptional cases
Classification of building type Flood vulnerability class HW-VC
Main building type short A B C D E
Clay Clay
Prefabricated PF
Framework FW
Masonry MW
Reinforced concrete RC
Flood resistant designed buildings FRD
Damage and loss prediction model based on the vulnerability of bDamage and loss prediction model based on the vulnerability of building typesuilding types
Basic elements Basic elements
Damage and loss prediction model based on the vulnerability of bDamage and loss prediction model based on the vulnerability of building typesuilding types
Flood vulnerability class HW-E
Bundesministerium für Verkehr, Bau- und Wohnungswesen:Planen und Bauen von Gebäuden in hochwasser-gefährdeten Gebieten
Bauhaus-Universität WeimarInstitute for Foundations and Soil Mechanics
Damage and loss prediction model based on the vulnerability of bDamage and loss prediction model based on the vulnerability of building typesuilding types
Basic elements Basic elements
Building types Flood vulnerability classes
Damage and loss prediction model based on the vulnerability of bDamage and loss prediction model based on the vulnerability of building typesuilding types
Specific Vulnerability Functions of type: Dm = f (hgf)
1
2
3
4
5
-3 -2 -1 0 1 2 3 4 5
Inundation level hgf [m]
Dam
age
grad
e D
m
PF
FW
Clay
MW
RC
1
2
3
4
5
-3 -2 -1 0 1 2 3 4 5
Inundation level hgf [m]
Dam
age
grad
e D
m
HW-A
HW-B
HW-C
HW-D
Dm - Mean damage gradehgf - Inundation level over ground floor
Damage and loss prediction model based on the vulnerability of bDamage and loss prediction model based on the vulnerability of building typesuilding types
Basic elements Basic elements
Damage grades Dm Buildings (without specification)
Damage and loss prediction model based on the vulnerability of bDamage and loss prediction model based on the vulnerability of building typesuilding types
Specific Vulnerability Functions of type: Dm = f (hgl, vfl)
1
2
3
4
5
0 1 2 3 4 5Specific energy height H [m]
Dam
age
grad
e D
m
1
2
3
4
5
0 1 2 3 4 5Specific energy height H [m]
Dam
age
grad
e D
m
H = hgl + (vfl²/2g) hgl - Inundation level over ground level vfl - Flow velocity
Damage and loss prediction model based on the vulnerability of bDamage and loss prediction model based on the vulnerability of building typesuilding types
Basic elements Basic elements
Damage and loss prediction model based on the vulnerability of bDamage and loss prediction model based on the vulnerability of building typesuilding types
0
25
50
75
100
125
0 1 2 3 4 5Inundation level hgl [m]
Loss
[%]
Losses (MDR) and impact parameters
H = hgl + (vfl²/2g) hgl
0
25
50
75
100
125
0 1 2 3 4 5Specific energy height H [m]
Loss
[%]
Inundation level hgf
Loss
[%]
Damage and loss prediction model based on the vulnerability of bDamage and loss prediction model based on the vulnerability of building typesuilding types
Basic elements Basic elements
SDF 1b: Example for HW-C, with cellar
SDF 2: Example for floor number n, with cellar (under preparation)
Damage and loss prediction model based on the vulnerability of bDamage and loss prediction model based on the vulnerability of building typesuilding types
Specific Damage Functions (SDF i)
Inundation level hgf
Loss
[%]
1 floor
2 floors
3 floors
4 floors
D5
D4
D3
D2
D1
Case studiesCase studies
Damage and loss prediction model based on the vulnerability of bDamage and loss prediction model based on the vulnerability of building typesuilding types
1
23
Case studies
Damage and loss prediction model based on the vulnerability of bDamage and loss prediction model based on the vulnerability of building typesuilding types
Study areas in Saxony
1 – Eilenburg2 – Döbeln3 – Grimma
Damage and loss prediction model based on the vulnerability of bDamage and loss prediction model based on the vulnerability of building typesuilding types
Case studies Case studies
Observed damage grades Re-Interpretation
Damage and loss prediction model based on the vulnerability of bDamage and loss prediction model based on the vulnerability of building typesuilding types
1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0
no damage observed no damage data available
1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0
no damage observed no damage data available
Mean damage grades in the ATKIS- land use areas (MDm):
Study area 1: Eilenburg
Damage and loss prediction model based on the vulnerability of bDamage and loss prediction model based on the vulnerability of building typesuilding types
Case studies Case studies
Single Buildings ATKIS-land use areas
Damage and loss prediction model based on the vulnerability of bDamage and loss prediction model based on the vulnerability of building typesuilding types
Mean Damage Ratio(MDR) in [%]
Study area 2: Döbeln
0 10 20 30 40 50 60 70 80 90 100 >100
no damage observed
0 10 20 30 40 50 60 70 80 90 100 >100
no damage observed
Damage and loss prediction model based on the vulnerability of bDamage and loss prediction model based on the vulnerability of building typesuilding types
Case studies Case studies
Consideration of the inundation level; Dm = f (hgf)
Consideration of inundation level and flow velocity; Dm = f (hgl, vfl)
Damage and loss prediction model based on the vulnerability of bDamage and loss prediction model based on the vulnerability of building typesuilding types
1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0
no damage observed no damage data available
1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0
no damage observed no damage data available
Mean damage grades in the ATKIS- land use areas (MDm):
Study area 3: Grimma
Damage and loss prediction model based on the vulnerability of bDamage and loss prediction model based on the vulnerability of building typesuilding types
Case studies Case studies
Damage and loss prediction model based on the vulnerability of bDamage and loss prediction model based on the vulnerability of building typesuilding types
62.258.5Residential building stockMesoscale3
149.4145.0Total building stock
71.861.9Residential building stockMicroscale2
166.3146.0Total building stock
89.983.3Residential building stockMicroscale1
EDAC –loss model
ReportedLosses in [Mio. €]ReferenceLevelStudy area
ConclusionsConclusions
Damage and loss prediction model based on the vulnerability of bDamage and loss prediction model based on the vulnerability of building typesuilding types
Damage and loss prediction model based on the vulnerability of bDamage and loss prediction model based on the vulnerability of building typesuilding types
Conclusion (I) Conclusion (I)
Damage and loss prediction model based on the vulnerability of bDamage and loss prediction model based on the vulnerability of building typesuilding types
Result: Damage and loss prediction model based on an engineering evaluation system of buildings including:
- a unified definition of global structural Damage Grades (Di) - Specific Flood Vulnerability Classes (HW-A to HW-E)- Specific Vulnerability Functions (SVF) of type: Dm = f (hgf) Dm = f (hgl, vfl)
- Specific Damage Functions (SDF) of type:Loss [%] = f (hgf, HW-VC, n, poc)Loss [%] = f (hgf, Di,)
Damage and loss prediction model based on the vulnerability of bDamage and loss prediction model based on the vulnerability of building typesuilding types
Conclusion (II) Conclusion (II)
Damage and loss prediction model based on the vulnerability of bDamage and loss prediction model based on the vulnerability of building typesuilding types
The Results and developed Tools:
- enable the reinterpretation of damage and loss parameters, are suited for damage and loss prediction
- enable the identification of critical zonesare suited for disaster management decisions(short- and long-term)
Damage and loss prediction model based on the vulnerability of bDamage and loss prediction model based on the vulnerability of building typesuilding typesDamage and loss prediction model based on the vulnerability of bDamage and loss prediction model based on the vulnerability of building typesuilding types
Thanks for
Contact:http://www.edac.biz
Project ManagementFunding Coordination
RIMAX-MEDIS