Analyzing & ImplementingDelayed Deceleration Approaches

DISTRIBUTION STATEMENT A. Approved for public release: distribution unlimited. © 2017 Massachusetts Institute of Technology

12th USA/Europe ATM Research & Development Seminar, Seattle, WAJune 26-30, 2017

Tom G. Reynolds, Emily Clemons& Lanie Sandberg

R. John Hansman & Jacquie Thomas

DDA - 2TGR 5/5/17

Aircraft Operations Environment Assessment

• Initial scoping study to identify & evaluate operational techniques to reduce fuel burn and environmental impacts in the near/mid-term with minimal implementation barriers*

• Objectives: – Identification, evaluation

and prioritization of 60+options

– Detailed analysis of benefits &barriers of promising options

• Cruise Altitude and SpeedOptimization

• Delayed Deceleration Approaches

– Promote deployment of best practice operations• Socialization with stakeholders

• Integration strategies for current & future operations

Surface (S):16 mitigations

Departure (D):11 mitigations

Approach (A)& Landing (L)9 mitigations

Cruise (C):14 mitigations

Miscellaneous (M): 11 mitigations

ORIGINAIRPORT

DESTINATIONAIRPORT

*Marais, K., et al., “Evaluation of Potential Near-term Operational Changes to Mitigate Environmental Impacts of Aviation”, Journal of Aerospace Engineering, Vol. 227, No. 8.

DDA - 3TGR 5/5/17

• Keep aircraft “clean” for longer on approach when appropriate without impacting terminal area entry or final approach stabilization criteria– Between these

speed gates, opportunity for encouraging moreefficient approachspeed profiles

Delayed Deceleration Approach (DDA) Concept

Distance to touchdown

AirspeedTypicalConventional

Terminal areaentry speed

Final approachspeed

Sample flap 1

Sample flap 2

Runway

Delayed Decel.=> Low Power/

Low Drag

“Clean” configuration

“Dirty”configuration

230-250kts IAS

160-180kts IAS

Airspeed

Distance to Touchdown ≈10 NM≈30 NM

DDA - 4TGR 5/5/17

• Delayed Deceleration Approach (DDA) Concept

• DDA fuel burn and emissions reduction potential

• Analyzing speed profiles and barriers at US airports

• Noise analysis

• Implementing DDA via RNAV procedures

• Conclusions & Recommendations

Outline

DDA - 5TGR 5/5/17

DDA Benefits Potential

A320 performanceprofiles

European A320 Flight

Data Recorder Analysis

(similar results for B757 & B777)

30-50% fuel burn reduction potential from DDAs from 10,000 ft to touchdown

30-50% fuel burn reduction potential from DDAs from 10,000 ft to touchdown

• Lowest fuel burn flights (green profiles) associated with delayed deceleration

(NM)

(NM)

(NM)

(NM)

DDA - 6TGR 5/5/17

System-Wide DDA Benefit Potential

• Estimated system-wide fuel saving benefits potential of increased DDA utilization

y = 0.7465x + 89.752R² = 0.847

0

50

100

150

200

250

300

350

400

0 50 100 150 200 250 300 350 400

Maximum Gross Take-Off Weight (metric tons)

DD

A F

ue

l S

avin

gR

ela

tive

to

Ave

rag

e (

lbs

)

A320

B757

B777

RJ Small NB

Large NB Two Engine WB

Four Engine

WB

Based on FDR analysis

Aircraft Weight Class

Example Aircraft Types

DDASaving

per Approach

Approx # Flights

perDay

FuelReduc-

tion Benefit

Pool (gal/yr)

$ Annual Savings per 1% Inc. in DDAUse*

RJ CRJERJ 120 7500 49m $1.5m

Small NB A320 B737 146 14,400 115m $3.5m

Large NB B757 183 1,800 18m $0.5m

Two Engine

WB

A330 B777 276 3,900 59m $1.8m

Four Engine

WB

A340 B747 375 2,400 49m $1.5m

Totals 30,000 290m $8.8m

* FAA investment analysis recommended fuel price*FAA investment analysis recommended fuel price of $3.02/gal

DDA - 7TGR 5/5/17

• Delayed Deceleration Approach (DDA) Concept

• DDA fuel burn and emissions reduction potential

• Analyzing speed profiles and barriers at US airports

• Noise analysis

• Implementing DDA via RNAV procedures

• Conclusions & Recommendations

Outline

DDA - 8TGR 5/5/17

Fuel Burn Correlations with Other Observable Metrics

• Need to find correlations between fuel burn and states observable in US radar track data

• Correlation analysis identified Airspeed and Time with Flap1 as main drivers of fuel burn– Flap1 speed

180-210 kts for most “large” aircraft

Time flown below 180kts used as proxy for fuel burn for large weight category aircraft

DDA - 9TGR 5/5/17

• Analyzed speed profiles at range of US airports– Capacity-constrained standalone airports (ATL, LAX, BOS, CLT)– New York metroplex airports (EWR, JFK, LGA)– Washington DC metroplex airports (DCA, IAD, BWI)– Capacity-unconstrained standalone airports (STL, RIC, DFW)

• 9 months of radar archives from 2011 and 2015• Ground speed converted to airspeed using NARR wind data

Quantifying Speed Profiles at US Airports

 

Airspeed Estimation

DDA Analysis Metrics

Fuel Efficiency:Time Flown

Below

Throughput Efficiency: Minimum Approach Spacing

DDA Opportunity Evaluation

WindData

Radar Archives

DDA - 10TGR 5/5/17

ATL Approach Speed Analysis

Airspeeds█ <= 210KIAS █ <= 180KIAS

20 NM

ATL

“Large” aircraft arrivals from ZDC,4 sample days

DDA - 11TGR 5/5/17

ATL Approach Speed Analysis

• Time flown below 180 kts correlated most strongly with fuel burn for “large” aircraft type

• Cumulative curves of time flown below 180 kts used as key airport speed metric

Arrivals from ZDCJAN – SEP 2011

DDA - 12TGR 5/5/17

NYC Metroplex Approach Speed Analysis

• Earlier decelerations generally observed under IMC compared to VMC

• LGA under IMC has earliest decelerations

EWRJFK

LGA

EWR█ <= 210KIAS█ <= 180KIAS

LGA█ <= 210KIAS█ <= 180KIAS

JFK█ <= 210KIAS█ <= 180KIAS

15 mi

Arrivals from ZDC, 8 sample days 1 3 5 7 9 11 13 150

20

40

60

80

100

Time Flown Below 180 kt (min)

Perc

enta

ge o

f All

Land

ings

LGA VMC n = 10,530LGA IMC n = 1,600JFK VMC n = 9,070JFK IMC n = 1,048EWR VMC n = 4,435EWR IMC n = 775

Arrivals from ZDCJAN – SEP 2011

DDA - 13TGR 5/5/17

Airport Approach Speed Profile Comparison

 

0

0.5

1

1.5

2

2.5

3

3.5

4

4.5

5

DCA LGA EWR BOS IAD JFK BWI LAX ATL RIC STL

VMC IMC Weighted

Time F

lown

Belo

w 18

0 kts

for 5

0% o

f Flig

hts

(min

s) Arrivals from ZDCJAN – SEP 2011

Unconstrained standalone

Constrained metroplex

DDA - 14TGR 5/5/17

• Need to understand causes of differing speed profiles to identify opportunities for increased DDA operations– If primary drivers to encourage greater DDA usage are easily

modifiable (e.g., ATC training or procedures) => good target airports– More difficult to increase DDA-type procedures at airports where

primary drivers are elements such as airspace or airport constraints

• Created decision trees to find combinations of independent variables correlated with time flown below 180kts

• Independent Variables– Weather (VMC or IMC)– Hourly Airport Acceptance Rate (AAR)– Total arrival demand (15 min bins)

Analyzing Drivers of Speed Profiles

– Airport configuration– Airline

DDA - 15TGR 5/5/17

Classification Tree Approach

• Technique identified key drivers of approach speed profiles

1. Weather conditions (VMC vs. IMC)

2. Dominant operator3. Airport configuration4. Higher capacities

(AAR)

• Relative importance varies by airport

• Dominant carrier influence suggests airline procedural effect

Wx=VMC

Config=West flow

AAR >= 107.5

ArrDem >= 18.5

AAR >= 113 AAR >= 114.5

AAR >= 118.5

Airline=Delta

Condition

4

4

4

3

3 2

2

1

1

1: t < 1222: 122 <= t < 1693: 169 <= t < 2674: t >= 267

Time Flown Below 180kts (s)

ATL Example

AAR = Airport Acceptance Rate (per hr)ArrDem = Arrival Demand (per 15 mins)

DDA - 16TGR 5/5/17

• Delayed Deceleration Approach (DDA) Concept

• DDA fuel burn and emissions reduction potential

• Analyzing speed profiles and barriers at US airports

• Noise analysis

• Implementing DDA via RNAV procedures

• Conclusions & Recommendations

Outline

DDA - 17TGR 5/5/17

• Airframe and engine noise both important during approach & landing

DDA Noise Impacts

Landing gearnoiseEngine

noise

Airframenoise

Noise“hot spots”

Empirical & modeling studies conducted to understand DDA noise impacts

DDA - 18TGR 5/5/17

BOS Noise Measurement CampaignMeasurement Period: 11/13/15 – 1/25/16

Airport Selection For DDA Feasibility

DDA Opportunity Analysis

Noise Monitor Site Selection

Noise Measurement Campaign

Noise & Aircraft Track Analysis

Research Outputs:• Noise modeling validations• Noise as f(speed, configuration,

approach procedures, aircraft type)

Airline Standard Operating

Procedures, Flap Schedules, etc.

Radar Flight Track Data

50 100 150 200 250 300 350 400 450 500

50

100

150

200

250

300

350

400

450

500

BOS

Runway 22L/22R ArrivalsJan-Mar and Jun-Aug 2014

5 nmi 0

100

200

300

400

500

600

700

800

900

1000

No

. o

f R

ad

ar

Su

rve

illa

nc

e H

its

0

0.5

1

1.5

2

2.5

3

3.5

4

4.5

5

DCA LGA EWR BOS IAD JFK BWI LAX ATL RIC STL

VMC IMC Weighted

Tim

e Fl

own

Belo

w 1

80 k

ts fo

r 50%

of F

light

s (m

in)

0

0.5

1

1.5

2

2.5

3

3.5

4

4.5

5

DCA LGA EWR BOS IAD JFK BWI LAX ATL RIC STL

VMC IMC Weighted

Tim

e Fl

own

Belo

w 1

80 k

ts fo

r 50%

of F

light

s (m

in)

DDA - 19TGR 5/5/17

• Slight negative trends indicate marginal decrease in noise with increasing airspeed on average

• Similar results for Lmax

Noise Measurement Results

Linear fit slope: -0.01 Linear fit slope: -0.03 Linear fit slope: -0.06

Linear fit slope: 0.00 Linear fit slope: -0.02 Linear fit slope: -0.05

Linear fit slope: -0.03 Linear fit slope: -0.02 Linear fit slope: -0.01

A320(n=135)

Monitor A

B737(n=163)

E190(n=58)

A320(n=269)

B737(n=295)

E190(n=112)

Monitor B Monitor CA320(n=365)

B737(n=352)

E190(n=153)

• 10-15 dBA variability in data– Not removed when correction for energy

change rate applied– Attributed to atmospheric differences

DDA - 20TGR 5/5/17

TASOPT, BADA4 & ANOPPModel Integration

DDA - 21TGR 5/5/17

• Modeled range of approaches with different aircraft types & profiles

Modeling of EnhancedDDA Operations

AirframeOnly

EngineOnly

Total Airframe& Engine

A320

Overall, negligible effect of DDA speed profiles on noise impacts on ground

DDA - 22TGR 5/5/17

• DDA fuel burn and emissions reduction potential

• Analyzing speed profiles and barriers at US airports

• Noise analysis

• Implementing DDA via RNAV procedures– Feasibility of RNAV procedures for DDA implementation– Design considerations for RNAV DDAs– Assessing RNAV Procedure Targets

• Conclusions & Recommendations

Outline

DDA - 23TGR 5/5/17

Assessing Feasibility of RNAV DDAs

• Biggest uncertainty in effective implementation of DDA: track distance to touchdown– PBN (RNAV/RNP) approach

procedures reduce uncertainty

• Simulated approaches with:– Same programmed

lateral/vertical route as published RNAV arrivals

– Different speed constraints– Different aircraft weights

• Utilized Lincoln FMS analysis capability– Pegasus FMS flight code– Integrated to B757-200

simulation

-84.4 -84.2 -84 -83.833.6

33.7

33.8

33.9

34

34.1

Longitude (deg)

Latit

ude

(deg

) 200K

200K

185K

175K

ATLZELOW20NM

BAMMM27NM

HAARY40NM

DIRTY49NM

150K

BAMBU10NM

160K 200K

230K

250K

250KATL DIRTY RNAV Arrival to Runway 26R

Late decel.profileEarly decel. profile

DDA - 24TGR 5/5/17

0

200

400

600

800

1000

1200

1400

1600

1800

Scenario 1: Late decel.

Scenario 2: Medium decel.

Scenario 3: Early decel.

Aircraft Weight

Fuel

Bur

n In

side

TR

AC

ON

40 N

M to

50

ft A

GL

(lbs)

Light Medium Heavy

Simulated B757 ATL RNAV DDA Arrivals

• Early deceleration shown to use 54% more fuel in TRACON compared to late

• Late deceleration fuel burn shows the least sensitivity to aircraft weight

Deceleration Profile Late Medium Early

TRACON Duration 12 mins 14 mins 15 mins

TRACON Fuel Use 1030 lbs 1347 lbs 1588 lbs

TRACON Fuel Use Relative to Lowest - +31% +54%

DDA - 25TGR 5/5/17

Design Considerationsfor RNAV DDAs

• For RNAV DDA procedures to be practically useful, need to minimally modify existing procedures and be flyable by different types

• Evaluated speed envelope for range of representative profiles and aircraft types

• Significant differences in deceleration profiles observed between types

Based on BADA 4 analysis

Need to account for aircraft deceleration capabilities in RNAV DDA design

DDA - 26TGR 5/5/17

• Modeled fuel burn on BOS ROBUC2 RNAV arrival with four different speed profiles

• Significant difference between “B757 latest” and “a/c type latest”– Trade-off between procedure simplicity and fuel saving

Comparison of Fuel Saving with Different Speed Profiles for Different Types

0

200

400

600

800

1000

1200

A320 B737 B757 B777

Empirical earlyROBUC2 publishedB757 latestA/c type latest

23%-1% -15%

Ref19%

0%-23%

Ref

53%

-8%-8%Ref

35%

-24%-34%

Ref

Fuel

Bur

n R

OB

UC

to

Tou

chdo

wn

(kg)

Based on BADA 4 analysis

DDA - 27TGR 5/5/17

• Compared speed profiles of range of current RNAV arrivals compliant with– “As published” speed targets– Latest speed profile flyable by B757 reference aircraft

Assessment of Current RNAV Arrival Speed Targets

Distance to Touchdown (NM)

Indi

cate

d Ai

rspe

ed (k

ts)

BOS/ROBUC/4R/as publishedBOS/ROBUC/4R/B757 latestBOS/JFUND/15R/as publishedBOS/JFUND/15R/B757 latestCLT/FILPZ/36L/as publishedCLT/FILPZ/36L/B757 latestCLT/FILPZ/18R/as publishedCLT/FILPZ/18R/B757 latestRNAV procedure speed targets

Airport/RNAV proc/rwy/speed profile

DDA - 28TGR 5/5/17

• Area between lines indicates how close published procedure speed targets are to B757-optimized speed profile– Area comparison methodology proposed as a procedure efficiency

screening tool

Assessment of Current RNAV Arrival Speed Targets

Area

Bet

wee

n R

NAV

Spe

ed T

arge

t Com

plia

nt&

B75

7 La

test

Spe

ed P

rofil

e C

urve

s (N

M.k

ts)

0

200

400

600

800

1000

1200

1400

1600

1800

2000

Incr

easi

ng R

NA

VFu

el E

ffici

ency

Dec

reas

eing

RN

AV

Fuel

Effi

cien

cy

DDA - 29TGR 5/5/17

• Delayed Deceleration Approach (DDA) Concept

• DDA fuel burn and emissions reduction potential

• Analyzing speed profiles and barriers at US airports

• Noise analysis

• Implementing DDA via RNAV procedures

• Conclusions & Recommendations

Outline

DDA - 30TGR 5/5/17

• DDA is a promising concept for reducing approach fuel & emissions

• Radar analysis and classification tree approach provides insight into airports with highest benefit potential and lowest barriers

• Noise analysis showed negligible effects of DDA on noise impacts on ground

• RNAV arrival procedures offer a promising implementation path

• Tools presented for identifying RNAV arrival procedure candidates for modified speed targets to gain DDA benefits

Conclusions

DDA - 31TGR 5/5/17

• Further promote DDA concept to relevant stakeholders– ATC facilities, Airlines/flight crews, Procedure designers

• Use proposed methodologies and tools to undertake:– Wider screening of existing procedures– Identify candidates for re-design process– Redesign, analyze and deploy appropriate modified procedures to

realize DDA benefits in operational system

• Assess human factors implications of DDA on ATC and flight crews

• Explore how new automation (e.g., TSS) could be leveraged to promote DDA

Recommended Next Steps

TSS = Terminal Sequencing and Spacing

DDA - 32TGR 5/5/17

Acknowledgments & Disclaimer

Many thanks for Chris Dorbian, Aniel Jardines, Stephen Merlin &James Hileman of the FAA Office of Environment & Energy forsupporting this work.

DISTRIBUTION STATEMENT A. Approved for public release: distribution unlimited. Thismaterial is based upon work supported by the Federal Aviation Administration under Air ForceContract No. FA8721-05-C-0002 and/or FA8702-15-D-0001. Any opinions, findings, conclusionsor recommendations expressed in this material are those of the author(s) and do notnecessarily reflect the views of the FAA.

© 2017 Massachusetts Institute of Technology.

Delivered to the U.S. Government with Unlimited Rights, as defined in DFARS Part 252.227-7013 or 7014 (Feb 2014). Notwithstanding any copyright notice, U.S. Government rights in thiswork are defined by DFARS 252.227-7013 or DFARS 252.227-7014 as detailed above. Use ofthis work other than as specifically authorized by the U.S. Government may violate anycopyrights that exist in this work.