ACRP 4-09 �Risk Assessment Method to Support Modification of Airfield Separation Standard

AIRPORT DESIGN

1. ACRP 4-09
Risk Assessment Method to
Support Modification of
Airfield Separation Standards
Developed by:
Applied Research Associates, Inc.
Robert E. David & Associates
University of Oklahoma
Period:
Jun 2009 to Feb 2011ADG V
1.E-09
1.E-08
1.E-07
1.E-06
1.E-05
1.E-04
1.E-03
222 224 226 228 230 232 234 236
Taxiway/Taxiway Centerline to Centerline Separation (ft)
Risk of Collision per Operation
ADG V Standard = 267 ft ADG V
1.E-09
1.E-08
1.E-07
1.E-06
1.E-05
1.E-04
1.E-03
222 224 226 228 230 232 234 236
Taxiway/Taxiway Centerline to Centerline Separation (ft)
Risk of Collision per Operation
ADG V Standard = 267 ft ADG V
1.E-09
1.E-08
1.E-07
1.E-06
1.E-05
1.E-04
1.E-03
222 224 226 228 230 232 234 236
Taxiway/Taxiway Centerline to Centerline Separation (ft)
Risk of Collision per Operation
ADG V Standard = 267 ft

2. Project Panel
Chair
Ms. Laurie Cullen –HNTB Corporation
ACRP Staff Representatives
Ms. Marci A. Greenberger –Program Officer
Mr. Joseph J. Brown-Snell–Program Associate
Members
Mr. Gary C. Cathey -California Department of Transportation
Mr. Chad A. Gunderson -TKDA
Mr. Paul Herrera -Los Angeles World Airports
Mr. Scott McMahon-Morristown Municipal Airport
Jorge E. Panteli-MacFarland-Johnson
Liaison Representatives
Mr. John Dermody -Federal Aviation Administration
Mr. Chris Oswald -Airports Council International -North America
Christine Gerencher–Transportation Research Board

3. Project Team
Principal Investigator
Jim Hall–Applied Research Associates
Co-Principal Investigator
Richard Speir–Applied Research Associates
Project Manager
Manuel Ayres–Applied Research Associates
Team Members
Hamid Shirazi–Applied Research Associates
Robert E. David–RED & Associates
Yih-Ru Huang –University of Oklahoma
Regis Carvalho –Applied Research Associates
Arun Rao–Consultant
Samuel Cardoso–Applied Research Associates
Edith Arambula–Applied Research Associates

4. Briefing Outline
Background
Study Objectives
Project Tasks
Rationale of Airfield Separations
Accident and Incident Data Collected
Basis of Approach Used
Risk-Based Analysis Methodology
Case Studies and Validation
Plan to Gain Industry Support
Limitations and Conclusions

5. Background
Many airports were built before current standards were set
There is a need to increase airport and aviation capacity,
and operation of larger aircraft may be required in existing
airfields
In many cases there are physical and environmental
restrictions to increase existing separations
Available analysis alternatives are prescriptive and not
based on risk
Approximately 20% of ground (commercial aviation)
accidents in the U.S. are collisions during taxiing or
parking
More than 50% of fatal accidents occur during landing and
takeoff operations

6. Modification of Standards (MOS)
AC 150/5300-13 (FAA, 1989)
Modification to standards means any change to FAA
design standards other than dimensional standards for
runway safety areas.
Unique local conditions may require modification to
airport design standards for a specific airport.
The request for MOS should show that the
modification will provide an acceptable level of safety,
economy, durability, and workmanship.

7. Study Objectives
Develop simple and easy to use methodology to
evaluate risk of collisions associated with non-
standard airfield separations.
Obtain quantitative assessment for decision making
when standard cannot be met.
The methodology should serve as a screening tool to
evaluate the feasibility of submitting to the FAA a
request for Modification of Standards.

8. Project Tasks
1.Literature review and rationale of airfield separations
2.Collection of veer-off accident and incident data
3.Modification of Standards (MOS) survey
4.Develop proposed risk assessment methodology
5.Perform airport survey for selected MOS cases
6.Develop risk assessment methodology
7.Develop plan to gain industry support
8.Prepare project report

9. Rationale for Standards -FAA
Taxiways and Taxilanes: probability distribution of lateral
deviations plus a safety buffer of 10 ft
TWY/TWY: 1.2 x WS + 10 ft (between centerlines)
TWY/OBJ: 0.7 x WS + 10 ft (axis to object)
TXL/TXL: 1.1 x WS + 10 ft (between centerlines)
TXL/OBJ: 0.6 x WS + 10 ft (axis to object)
Runways: probability distributions of lateral and vertical
deviations during final approach and initial climb, as well
as probability of veer-offs during landing and takeoff
Indication that standards were developed based on best
engineering judgment and experience from WW II

10. Rationale for Standards -ICAO
Taxiway/Taxiway and Taxiway/Object:
Wingtip Clearance = clearance (C) between the outer
main gear wheel and the taxiway edge plus safety
buffer (Z).
Runway/Taxiway
Distance to accommodate potential veer-offs and provide
sterile area free of obstacles for aircraft executing a
missed approach or balked landing maneuver.

11. Veer-off Data Collection
Veer-off accidents and incidents occurring in several
countries from 1980 to 2009
Taxiway/Taxilane veer-offs
Identified 300 incidents in straight segments of
taxiways
Only 6 relevant incidents were identified in taxilanes
Identified 679 runway veer-off accidents and incidents
during landing and takeoff

12. Taxiway Veer-offs –Some
Conclusions
Taxiing airplanes are at lower speeds (normal 20
knots, max 30 knots) when compared to runway
operations.
The edge of the paved area is a discontinuity and the
pilot is able to stop as soon as the aircraft departs the
taxiway.
The model for lateral deviation can be truncated for
taxiways outside the ramp area.
The collisions occurred in curves or when other aircraft
and equipment were inside the taxiway/taxilane OFA.

13. Taxiway Veer-offs –More Conclusions
Taxiway veer-offs in straight segments occured due to
poor visibility or low surface friction (e.g. Icing
conditions).
Two-part models based on frequency and location
were not appropriate for the methodology.
Only two fatal accidents due to taxiway veer-offs were
identified; neither was relevant to this study.

14. Basis of Approach Used
Probability distributions of lateral and vertical
deviations during operations
Boeing/FAA Taxiway Deviation Studies at ANC and
JFK (Scholz, 2003 and 2005)
Airborne risk during landing derived from Collision
Risk Model (CRM) runs
Ground roll risk of veer-off derived from models
developed in this project (landing and takeoff)

15. Taxiways and Taxilanes Separation
Probability Distribution of Lateral Deviations
X
d= wingtip separation
centerline separation (CS)
WS1
WS2
d= CS –(WS1 + WS2) / 2

16. Taxiway or Taxilane to Object Separation
Probability Distribution of Lateral Deviations
0 X
obstacle
wingtip lateral
deviation probability
distribution
aircraft semi wingspan

17. RWY/TWY Separation
Risk of collision during airborne phase
Landing
Final Approach
Missed Approach
Rejected Landing
Takeoff –Initial Climb
Risk of collision during ground roll
Landing
Takeoff

18. Deviations in Airborne PhaseNominal Flight Path
(x = 0, y = 0)
h
y
Obstacle
x
Nominal Flight Path
(x = 0, y = 0)
h
y
Obstacle
x Nominal Flight Path
h
y
Obstacle
x
Y
1
= NFPh-h
X
1
= XO –WS/2
Nominal Flight Path
h
y
Obstacle
x
Y
1
= NFPh-h
X
1
= XO –WS/2

19. Runway Veer-off
x
Landing (or Takeoff)
1
2 3

20. Risk-Based Analysis Methodology
Taxiway to Taxiway or Taxilane
Taxiway to Object
Taxilane to Taxilane
Taxilane to Object
Runway to Taxiway/Taxilane/Object
Landing
Airborne phase
Ground rolling phase
Takeoff
Ground rolling phase

21. Taxiway Lateral Deviation Studies
FAA/Boeing (Scholz, 2003 and 2005)
Collision risk models were developed by Boeing/FAA
based on B-747 taxiway deviation studies at ANC and
JFK
The objective was to evaluate the risk of collision for
B-747-800 operations
Data was collected during one year
In both cases, lateral deviation data was collected in
straight segments with taxiway centerline lights

22. Assumptions
Lateral deviation for smaller aircraft are similar or
smaller than those of the B-747
The taxiway or taxilane centerline is conspicuous and
visible to the pilot under any operational conditions
The FAA separation standards for taxiways and
taxilanes are based on similar probability of aircraft
departing the lane during taxiing operations
The risk estimated with the CRM is more restrictive
compared to the risk under visual conditions

23. ACRP 4-09 Methodology
Example of Risk Plot for Taxiway/Taxiway Separation –ADG
I ADG I
1.E-09
1.E-08
1.E-07
1.E-06
1.E-05
62 64 66 68 70 72
Taxiway/Taxiway Centerline Separation (ft)
Risk of Collision per Operation
ADG I Standard = 69 ft
8.0E-7

24. Lateral Deviation Models for
Taxilanes
Wingtip
Separation
ADG -Distances in ft
I IIIIIIV V VI
Taxiway/Object20 26 34 44 53 62
Taxilane/Object 15 18 22 27 31 36
Ratio 0.750.690.650.610.580.58
Taxilane
Taxiway
Similar Probability

25. Analysis Procedure
Taxiways/ Taxilanes/Objects
Identify the type of separation
Identify the ADG or aircraft types involved
Characterize the separation (between centerlines,
between centerline and object, or wingtip clearance)
Identify the appropriate risk plot to use
Use the centerline or wingtip clearance to estimate risk
of collision

26. Example -Taxiway/Taxiway
SeparationTaxiway/Taxiway Separation - ADG V
1.E-09
1.E-08
1.E-07
1.E-06
1.E-05
226 228 230 232 234 236
Taxiway/Taxiway Centerline to Centerline Separation (ft)
Risk of Collision per Operation
ADG V Standard = 267 ft
2.3E-08

27. Risk Analysis during Landing
Airborne Phase
Ground Roll Phase

28. Collision Risk Model (CRM) Runs

29. Development of Risk Curves
Airborne Phase ADG III - CAT I
1.0E-11
1.0E-10
1.0E-09
1.0E-08
200 250 300 350 400 450 500
Runway/Taxiway Centerline Separation (ft)
Risk of Collision per Operation.
ADG III Approach Cat C
Standard = 400 ft ADG III - CAT I
1.0E-15
1.0E-12
1.0E-09
1.0E-06
0 100 200 300 400 500 600 700
Aircraft Distance from Runway Centerline (ft)
Risk per Operation
-328 ft
0 ft
1500 ft
3000 ft
4500 ft

30. Runway Veer-off Incident Rates (U.S.)
(1980-2009)
Type of
Incident
Number
of
Incidents
Incident Rate
per Operation
Incident Rate in
Operations per
Incident
LDVO 512 1.195E-06 837,000
TOVO 111 2.590E-07 3,861,000

31. Location Model –Landing Veer-offProb=exp((-.02568)*Y**(.803946))
R2=99.5%
0 200 400 600 800 1000
Distance Y from Runway Edge (ft)
0.0
0.2
0.4
0.6
0.8
1.0
Probability of Stopping Beyond Y

32. Analysis Procedure –Runway/Taxiway
Identify the ADG
Identify type of approach (Cat I or Cat II)
Characterize the separation between the runway and
taxiway axes
Identify plots for specific ADG (landing)
Airborne phase (lateral and vertical deviations)
Ground roll phase (frequency and location)
Use axes separation to estimate risk of collision for
each phase
Repeat process for takeoffs

33. Risk Criteria –FAA Risk Matrix
Risk estimated is compared to risk criteria to check for acceptability

34. Case Studies and Validation
Airp.ADG Type of MOS Risk
Level
Expected #
Yrs
Risk <
1.0E-7
Risk <
1.0E-09
Credible
Severity
FAA Risk
Classification
Acceptable
PHL III, IVTaxilane/Taxilane<1.0E-9 N/A Yes Yes Major Low Yes
ANC VI Taxiway/Object<1.0E-9 N/A Yes Yes Major Low Yes
ADS III Runway/Taxiway 1.0E-7 > 100 Yes No CatastrophicMedium Yes
BDR II Runway/Taxiway 1.1E-7 > 100 No No CatastrophicMedium Yes
MFV II Runway/Object 5.9E-8 > 100 Yes No CatastrophicMedium Yes
N07 I Taxilane/Object1.2E-9 - Yes No Major Low Yes
JFK VI Taxiway/Taxiway<1.0E-9 - Yes Yes Major Low Yes
EWR V Taxiway/Taxiway
Taxilane/Object
<1.0E-9
<1.0E-9
N/A
N/A
Yes
Yes
Yes
Yes
Major
Major
Low
Low
Yes
Yes
MSP IV Taxiway/Taxiway<1.0E-9 N/A Yes Yes Major Low Yes
ORD V Taxiway/Object<1.0E-9 N/A Yes Yes Major Low Yes
ORD V Taxiway/Taxiway<1.0E-9 N/A Yes Yes Major Low Yes
HYA III Runway/Taxiway 8.8E-8 > 100 Yes No CatastrophicMedium Yes
LCI III Runway/Taxiway 2.0E-7 > 100 No No CatastrophicMedium Yes
SEA VI Runway/Taxiway 1.6E-6 N/A No No Catastrophic High* No*
SEA VI Taxiway/Taxilane<1.0E-9 N/A Yes Yes Major Low Yes
ASE III Runway/Taxiway 9.0E-8 > 100 Yes No CatastrophicMedium Yes
ACK III Taxiway/Taxiway<1.0E-9 N/A Yes Yes Major Low Yes
ILG IV Taxiway/Object2.8E-8 - Yes No Major Low Yes
JYO II Runway/Taxiway 1.2E-7 > 100 No No CatastrophicMedium Yes
TAN II Runway/Taxiway 8.0E-8 > 100 Yes No CatastrophicMedium Yes

35. Plan to Gain Industry Support
Research Product
Risk assessment methodology to evaluate airfield
separations and intended to serve as a screening tool to
support the submittal of MOS for FAA approval
Audience
Civil aviation agencies like the FAA, ICAO, military
aviation organizations, and civil aviation stakeholders
Main obstacle for implementation
Will require FAA support
Implementation
Actions to present the product in airport conferences
and aviation safety meetings (TRB, AAAE, ACC, ACI)
Presentation to the FAA Office of Airports

36. Limitations
Can only be used to assess risk for straight parallel
segments of taxiways and taxilanes.
Taxiway deviations for smaller aircraft were assumed
to be equal or smaller than deviations for the Boeing
747 aircraft.
Application of the models for taxiway and taxilane
deviations assume the centerline is conspicuous under
any weather and light conditions.
Veer-off models were developed based on incidents
and accidents of aircraft with MTOW larger than 5,600
lbs.
Assumed the lateral and vertical deviation probability
distributions provided by the Collision Risk Model is
conservative when considering visual conditions.

37. Conclusions
The methodology developed in this research study
provides a practical and simple guide to help airports
quantify and evaluate risk associated with non-
standard airfield separations.
The risk assessment obtained can be helpful to
examine the feasibility of and to support MOS requests
to the FAA.
The methodology is based on lateral and vertical
deviation studies and models developed in this
research as well as in previous studies conducted by
the FAA, Boeing, and ICAO.
The methodology was validated using twenty MOS
cases approved by the FAA.