ANALYSIS OF AIRCRAFT SURFACE MOTION
AT
BOSTON LOGAN INTERNATIONAL AIRPORT
by
Robert Elias Alhanatis
B.S. Aerospace Engineering
Boston University, 1992
Submitted to the Department of Aeronautics and Astronautics
in Partial Fulfillment of the Requirements for the
Degree of
MASTER OF SCIENCE
at the
Massachusetts Institute of Technology
September 1994
© 1994 Robert Elias Alhanatis
All Rights Reserved
The author hereby grants to MIT permission to reproduce and to
distribute publicly paper and electronic copies of this thesis
document in whole or in part.
. -x........................................
Signature of the Author..............
Department of Aeronautics and Astronautics
July 13, 1994
Certified by ............................... ...........................
...................................
Professor Robert W. Simpson
Director, Flight Transportation Laboratory
\
Thesis Supervisor
Accepted by ............................................
......
f
...
...............................
Professor Harold Y. Wachman
MASChairman, Department Graduate Committee
S ISES- INSTITUrTE
SEP 2 11994
Llorurrur-Q
Aero
ANALYSIS OF AIRCRAFT SURFACE MOTION
AT
BOSTON LOGAN INTERNATIONAL AIRPORT
by
ROBERT ELIAS ALHANATIS
Submitted to the Department of Aeronautics and Astronautics
in Partial Fulfillment of the Requirements for the Degree of Master of Science in
Aeronautics and Astronautics
ABSTRACT
The purpose of this thesis is to examine the nature of aircraft surface motion on
the airport surface during normal operations. Twelve hours of radar data, gathered by
MIT Lincoln Laboratories from Logan airport in Boston, were made available for this
study. Specifically, the data included target position reports from the ASDE-3 surface
surveillance radar and the ASR-9 radar from the near terminal airspace information. This
data covers a variety of runway configurations, weather conditions, traffic levels and high
or low visibility conditions.
The study is divided into three sections. The first one focuses on the runway, and
examines occupancy times, exit velocities, exit usage and velocity profiles of the final
approach and landing phase. The second section, analyzes fourteen runway-taxiway
intersections. Results are presented for the crossing times and usage of these
intersections. The analysis also focuses on relating crossing times and usage to crossing
direction, runway configuration and aircraft size. Finally, average taxiway velocities and
the overall taxiway usage is measured. Additionally, the role that the location of the
taxiway segment as well as its length, plays in the variation of these velocities are
examined. Where possible, this study includes means, standard deviations and sample
sizes of the variables in question.
Thesis Supervisor :
Dr. Robert W. Simpson
Director, Flight Transportation Laboratory
Acknowledgments
I would like to thank several people for their time and support throughout the process of
writing this thesis. First of all, I would like to extend my sincere gratitude to Professor
Robert Simpson, my thesis supervisor, for his constant support and guidance. I am truly
grateful for the opportunity to have worked with him.
I would also like to thank MIT Lincoln Labs for providing the data for this thesis
and in particular Jerry Welch, Ted Morin and Jim Eggert for assisting me in getting the
data, answering my questions and providing me with many useful suggestions.
Additionally, I would like to thank John Pararas, Dennis Mathaisel and Husni Idris for
their help in developing the data processing software and the GMS interface.
I would also like to thank Professor Amadeo Odoni for his insightful remarks and
Professor Peter Belobaba for teaching the Airline Management and Airline Econimics
courses during my studies. I would also like to express my thanks to all my fellow
classmates. My stay at the Flight Transportation Lab has been a rewarding experience
thanks to them.
Finally, and most importantly, I would like to express my deepest gratitude to my parents
for their continues love and support.
R. A.
Cambridge, Massachusetts
July, 1994
Table of Contents
A bstract ....................................................................................................................
Acknowledgments ....................................................................................................
Table of Contents...............................................................................................
List of Tables.......................................................................................................
L ist of Figures..........................................................................................................
3
5
7
9
10
Chapter 1:
Introduction................................................
1.1
Background....................................................................
1.2
Motivation...........................................................
..............
1.3
Scope..................................................
14
14
16
16
Chapter 2:
The Measurement Task....................................................................
18
2.1
2.2
2.3
2.4
18
19
24
25
C hapter 3:
Data Analysis...........................................................................
3.1
3.2
3.3
3.4
3.5
3.6
Chapter 4:
Introduction ......................................
The Runway and Taxiway System.......................................
Data Collection............................
Available Data................................. .....
.............
29
Introduction.................................................................
29
Preliminary Data Processing...........................
....... 30
Data Irregularities .........................................................................
35
Runway Analysis ........................................
......... 36
3.4.1 Runway Occupancy Time During Landing......... 36
3.4.2 Exit Use................................... .....
............ 50
3.4.3 Exit Velocities........................... ..
........... 58
3.4.4 Velocity Profiles.......................................................
70
3.4.4.1
Landing Velocity Profiles................ 70
3.4.4.2
Final Approach Velocity Profiles..... 83
....... 90
Intersection Analysis ......................................
90
3.5.1 Intersection Crossing Times...........................
103
3.5.2 Intersection Use..................................
106
Taxiway Analysis.............................
106
3.6.1 Taxiway Average Velocity......................
110
3.6.2 Taxiway Use.............................................................
Conclusions ......................................................................................
121
4.1
4.2
4.3
4.4
4.5
Introduction...................................................................
Runways.................................
Intersections ..........................
Taxiways ................................
Directions for Future Research.............................
121
122
123
124
124
References................................................................................................................
126
List of Tables
Chapter 2
2.2
Table
2.4
Table
Table
Table
Table
Table
Table
Table
Table
Table
Table
Table
Table
Table
Table
Table
Table
Table
Table
Table
Table
The Runway and Taxiway System
Runway Configurations at Logan Airport......................... 21
2.2
Available Data
2.4.1
Runway Configuration for Data Block-1...........................
Weather information for Data Block- .............................
2.4.2
2.4.3
Runway Configuration for Data Block-2...........................
2.4.4
Weather information for Data Block-2..............................
2.4.5
Runway Configuration for Data Block-3...........................
2.4.6
Weather information for Data Block-3.............................
2.4.7
Runway Configuration for Data Block-4...........................
Weather information for Data Block-4.............................
2.4.8
2.4.9
Runway Configuration for Data Block-5...........................
Weather information for Data Block-5.............................
2.4.10
2.4.11
Runway Configuration for Data Block-6...........................
Weather information for Data Block-6.............................
2.4.12
2.4.13
Runway Configuration for Data Block-7...........................
Weather information for Data Block-7.............................
2.4.14
Runway Configuration for Data Block-8...........................
2.4.15
Weather information for Data Block-8.............................
2.4.16
Runway Configuration for Data Block-9...........................
2.4.17
Weather information for Data Block-9..............................
2.4.18
Runway Configuration for Data Block-10.........................
2.4.19
Weather information for Data Block-10...........................
2.4.20
Chapter 3
Preliminary Data Processing
3.2
Typical Sample Information about a Target.....................
Table 3.2.1
Series of Nodes that Define the Airport Run. & Tax........
Table 3.2.2
3.4.1 Runway Occupancy Time During Landing
Average Occupancy Time During Landing...................
Table 3.4.1.1 :
Exits Links Runways 27, 9......................................
Table 3.4.1.2 :
Exits Links Runways 33L, 15 R................................... .
Table 3.4.1.3 :
.
Exits Links Runways 4L, 22R.................................
Table 3.4.1.4 :
.
22L...............................
Exits Links Runways 4R,
Table 3.4.1.5 :
3.4.2 Exit Use
.........
Classification of Aircraft.............................
Table 3.4.2
3.5.1 Intersection Crossing Times
Series of Links that Corresponds to Each Intersection.........
Table 3.5.1
3.5.2 Intersection Use
Number of Aircraft Using Each Intersection....................
Table 3.5.2
30
32
36
37
37
40
40
50
92
103
List of Figures
Chapter 2
2.2
The Runway
Figure 2.2.1
Figure 2.2.2
Figure 2.2.3
and Taxiway System
Boston Logan International Airport - Location Map........ 19
Runway and Taxiway Configuration - Logan Airport......... 20
Taxiway System Around the Terminal Area..................... 23
Chapter 3
3.2
Figure
Figure
3.4
Preliminary Data Processing
3.2.1
Logan Airport - GMS Format.......................................
Node-Link Network.................................... ..................
3.2.2
Runway Analysis
3.4.1 Runway Occupancy Time During Landing
Figure 3.4.1.1 :
Runway 33L/15R...................................................
Runway 27/9............................................................
Figure 3.4.1.2 :
Runways 4R/22L, 4L/22R and 15L/33R........................
Figure 3.4.1.3 :
Aver. Occup. Time During Landing Run 4R Block 1........
Figure 3.4.1.4 :
Run 4R Block 6........
Figure 3.4.1.5 :
Run 4R Block 9........
Figure 3.4.1.6:
Run 22L Block 8......
Figure 3.4.1.7 :
Run 22L Block 10....
Figure 3.4.1.8:
Run 27 Block 2.........
Figure 3.4.1.9:
Run 27 Block 3.........
Figure 3.4.1.10:
Run 27 Block 8.........
Figure 3.4.1.11:
Run 27 Block 10.......
Figure 3.4.1.12:
Run 33L Block 2......
Figure 3.4.1.13:
Run 33L Block 5......
Figure 3.4.1.14:
Run 15R Block 4......
Figure 3.4.1.15:
Run 15R Block 7......
Figure 3.4.1.16:
3.4.2 Exit Use
Run 4R Block 1........
Exit Use per Aircraft Class
Figure 3.4.2.1 :
Run 4R Block 6........
Exit Use per Aircraft Class
Figure 3.4.2.2 :
Run 4R Block 9........
Exit Use All Classes
Figure 3.4.2.3 :
Run 22L Block 8......
Exit Use All Classes
Figure 3.4.2.4 :
Run 22L Block 10....
Exit Use per Aircraft Class
Figure 3.4.2.5 :
Run 27 Block 2.........
Exit Use per Aircraft Class
Figure 3.4.2.6 :
Run 27 Block 3.........
Exit Use All Classes
Figure 3.4.2.7 :
Run 27 Block 8.........
Exit Use All Classes
Figure 2.4.2.8 :
Run 27 Block 10.......
Exit Use per Aircraft Class
Figure 3.4.2.9 :
Run 33L Block 2......
Exit Use per Aircraft Class
Figure 3.4.2.10:
Run 33L Block 5......
Exit Use per Aircraft Class
Figure 3.4.2.11:
Run 15R Block 4......
Exit Use per Aircraft Class
Figure 3.4.2.12:
Run 15R Block 7......
Exit Use per Aircraft Class
Figure 3.4.2.13:
33
34
38
39
41
43
43
44
45
45
46
46
47
47
48
48
49
49
51
51
52
53
53
54
54
55
55
56
56
57
57
3.4.3 Exit Velocities
Figure 3.4.3.1 :
Average Exit Vel All Classes
Run 4R Block 1........
Figure 3.4.3.2 :
Run 4R Block 6........
Figure 3.4.3.3 :
Run 4R Block 9........
Figure 3.4.3.4 :
Run 22L Block 8......
Figure 3.4.3.5 :
Run 22L Block 10....
Figure 3.4.3.6 :
Run 27 Block 2.........
Figure 3.4.3.7 :
Run 27 Block 3.........
Figure 3.4.3.8 :
Run 27 Block 8.........
Figure 3.4.3.9 :
Run 27 Block 10.......
Figure 3.4.3. 10:
Run 33L Block 2......
Figure 3.4.3.11:
Run 33L Block 5......
Figure 3.4.3.12:
Run 15R Block 4......
Figure 3.4.3.13:
Run 15R Block 7......
Figure 3.4.3.14:
Average Exit Vel per Aircraft Clas s Run 4R Block 1........
Run 4R Block 6........
Figure 3.4.3.15:
Run 22L Block 10....
Figure 3.4.3.16:
Figure 3.4.3.17:
Run 27 Block 2 .........
Figure 3.4.3.18:
Run 27 Block 10.......
Run 33L Block 2......
Figure 3.4.3.19:
Run 33L Block 5......
Figure 3.4.3.20:
Run 15R Block 4......
Figure 3.4.3.21:
Run-15R Block 7 ......
Figure 3.4.3.22:
3.4.4 Velocity Profiles
Landing Velocity Profiles
3.4.4.1
Figure 3.4.4.1.1:
Landing Velocity Profiles
Run 4R Block 1........
Run 4R Block 6........
Figure 3.4.4.1.2:
Run 4R Block 9........
Figure 3.4.4.1.3:
Figure 3.4.4.1.4:
Run 22L Block 8......
Run 22L Block 10....
Figure 3.4.4.1.5:
Run 27 Block 2.........
Figure 3.4.4.1.6:
Run 27 Block 3.........
Figure 3.4.4.1.7:
Run 27 Block 8.........
Figure 3.4.4.1.8:
Run
27 Block 10.......
Figure 3.4.4.1.9:
Run
33L Block 2......
Figure 3.4.4.1.10:
Run 33L Block 5......
Figure 3.4.4.1.11:
Run 15R Block 4......
Figure 3.4.4.1.12:
Run 15R Block 7......
Figure 3.4.4.1.13:
3.4.4.2
Final Approach Velocity Iprofiles
Final
Approach
Velocity Profiles
Run 4R Block 1........
Figure 3.4.4.2.1:
Run 4R Block 6........
Figure 3.4.4.2.2:
Run 4R Block 9........
Figure 3.4.4.2.3:
Run 22L Block 8......
Figure 3.4.4.2.4:
Run 22L Block 10....
Figure 3.4.4.2.5:
Run 27 Block 2 ........
Figure 3.4.4.2.6:
Run 27 Block 3.........
3.4.4.2.7:
Figure
27 Block 8.........
Run
3.4.4.2.8:
Figure
Run
27 Block 10.......
Figure 3.4.4.2.9:
Run
33L Block 2......
Figure 3.4.4.2.10:
Run
33L Block 5......
Figure 3.4.4.2.11:
Run 15R Block 4......
Figure 3.4.4.2.12:
Run 15R Block 7......
Figure 3.4.4.2.13:
3.5
Intersection Analysis
58
59
59
60
60
61
61
62
62
63
63
64
64
65
66
66
67
67
68
68
69
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
84
85
85
86
86
87
87
88
88
89
89
3.5.1
Intersection Crossing Times
Typical Intersection............................
Figure 3.5.1.1 :
..................
Logan's Intersections...........................................................
Figure 3.5.1.2 :
Average Intersection Crossing Times
Figure 3.5.1.3 :
Block 1.........
Figure 3.5.1.4 :
Block 2.........
Figure 3.5.1.5 :
Block 3.........
Figure 3.5.1.6 :
Block 4.........
Figure 3.5.1.7 :
Block 5.........
Figure 3.5.1.8 :
Block 6.........
Figure 3.5.1.9 :
Block 7.........
Block 8.........
Figure 3.5.1.10:
Figure 3.5.1.11:
Block 9.........
Block 10.......
Figure 3.5.1.12:
Figure 3.5.1.13: Avg. Inters. Cros. T. In Dir
Int. : 10, 11, 12, 13, 14......
Figure. 3.5.1.14: Avg. Inters. Cros. T. Out Dir
Int. : 10, 11, 12, 13, 14......
Int. 6...............................
Figure 3.5.1.15: Avg. Inters. Cros. T. Int Dir
Int. 6...............................
Figure. 3.5.1.16: Avg. Inters. Cros. T. Out Dir
Figure 3.5.1.17: Avg. Int.. Cros. T. per Aircraft Class
Block 1.....
..........
Figure 3.5.1.18: Avg. Int.. Cros. T. per Aircraft Class
Block 2..................
Figure. 3.5.1.19: Avg. Int.. Cros. T. per Aircraft Class
Block 4..................
Figure 3.5.1.20: Avg. Int.. Cros. T. per Aircraft Class
Block 5..................
3.5.2 Intersection Use
Number of Aircraft Using Each Intersection
Figure 3.5.2.1 :
Arr: 4R, 4L Dep :9, 4L.................
Configuration :
Figure 3.5.2.2 :
Configuration :
Arr : 33L
Dep :33L, 22R.........
Arr : 27, 22L Dep : 22R, 22L.........
Configuration :
Figure 3.5.2.3 :
Arr: 15R
Figure 3.5.2.4 :
Configuration :
Dep :9, 15R...........
3.6
Taxiwa3 Analysis
3.6.1 1'axiway Average Velocity
Figure 3.6.1.1
Average Velocity in Taxiw ays.............................................
Figure 3.6.1.2
Various Lengths.......
Average vel ocity in Taxiways
Various Locations .....
Figure 3.6.1.3
Average Vel ocity in Taxiways
Average Vel ocity in Taxiways
Figure 3.6.1.4
Various Loc.&Leng..
3.6.2 Taxiway Use
Block 1........
Figure 3.6.2.1
Taxiway use Inner and Outer Taxilane
Block 2.........
Figure 3.6.2.2
Block 3.........
Figure 3.6.2.3
Block 4.........
Figure 3.6.2.4
Block 5 ........
Figure 3.6.2.5
Block 6.........
Figure 3.6.2.6
Block 7.........
Figure 3.6.2.7
Block 8.........
Figure 3.6.2.8
Block 9.........
Figure 3.6.2.9
Block 10.......
Figure 3.6.2.1(
Taxiway use Supporting Taxiways
Block 1.........
Figure 3.6.2.11I:Z:
Block 2.........
Figure 3.6.2.1
Block 3.........
Figure 3.6.2.1 3:
Block 4.........
Figure 3.6.2.1 4:
Block 5.........
Figure 3.6.2.1 5:
Block 6.........
Figure 3.6.2.1(5:
Block 7.........
Figure 3.6.2.1' 7:
3:
Block 8.........
3.6.2.1
Figure
Block 9.........
Figure 3.6.2.15
Block 10.......
3.6.2.2(
Figure
90
91
94
94
95
95
96
96
97
97
98
98
99
99
100
100
101
101
102
102
104
104
105
105
107
108
109
109
111
111
112
112
113
113
114
114
115
115
116
116
117
117
118
118
119
119
120
120
13
Chapter 1
Introduction
1.1
Background
After forty years of regulation by the Civil Aeronautics Board, in 1978 Congress enacted
the Airline Deregulation Act, which phased out economic regulation of the industry. In
the years following deregulation many new carriers entered the airline industry. The old
and the new airlines soon started servicing new city-pair markets, offering expanded
services and competitive fares. These developments resulted in a significant increase in
the overall traffic levels. In order to provide higher schedule frequencies and more
efficient use of their fleet, the airlines soon abandoned the point-to-point route networks
and adopted hub and spoke network systems that concentrated traffic around hub airports.
The increased passenger traffic coupled with the concentration of this traffic around, led
to congestion within the available airspace and subsequent delays in these hub airports.
Due to the aforementioned reasons, the need to develop means for greater
efficiencies in aircraft operations became apparent. Major efforts are undertaken today,
focusing on the use of advanced technologies for airborne and ground traffic control
systems in a concentrated effort to decrease the unused airspace and increase airport
capacity while simultaneously maintaining or even increasing safety levels. One such
area of focus is the airport surface, where especially in periods of low visibility, aircraft
experience significant delays on their way to the gate or departing runway. During the
last decade, various systems have been conceptualized and are currently under
development which deal with problems controllers and pilots face every day on the
airport surface.
The major objective of these surface traffic systems is to enhance the safety,
capacity, and productivity of these airports, while at the same time reducing delays and
the workload of both controller and pilot. This is accomplished via the development of
advanced communications, surveillance and automation techniques for use in the control
towers of major airports. Various subsystems address isolated problems such as runway
incursions, taxiway guidance and surface traffic surveillance.
Airport safety is intrinsically linked to capacity. The spacing between aircraft
necessarily reduces with increasing capacity, and safety suffers unless the reduction in
spacing is done carefully. The suggested long term solution is a surface traffic
management system that will address all these subsets of problems in an integrated
manner and safely control the airport surface area. Such a system must address the
capacity issues of ground congestion and effective departure sequencing through the
implementation of efficient routing and sequencing of aircraft on the surface, thereby the
system would decrease delays and increase airport safety. In order for such a system to be
successfully developed and implemented, information about the nature of aircraft motion
on the airport surface must be detailed.
1.2
Motivation
Few studies have been conducted to date on aircraft motion on the airport surface during
normal operations. In 1960 the Airborne Instruments Laboratory at Cornell University
published a series of reports about velocities and accelerations of aircraft at Kennedy
Airport in New York. Later, in 1972 the Flight Transportation Laboratory at MIT studied
the air-side activity of Boston Logan and Atlanta airports. Measurements were taken for
runway occupancy times, velocity profiles along the runways, taxiway speeds and
intersection delays. Unfortunately, most of the aircraft that operated during those years
are not in service today. Additionally, the data was gathered solely in periods of good
visibility and therefore the data of these reports is of little value today. It is therefore of
vital interest to measure the surface movements of today's aircraft as completely and
effectively as possible.
1.3
Scope
An aircraft engages in a series of non-uniform and complex maneuvers on its way
to the gate or the departing runway. A departing aircraft for example, after getting the
clearance to push back from the gate, has to follow a taxi route that will lead it to the
takeoff runway. This path varies depending on the layout of the taxiway system, the
current runway configuration, and the location of the gate. It might be short or long and
might involve a considerable number of turns, stops, taxiway and runway crossings, and
varying length segments of straight taxing. Along this route, the pilot must be constantly
be aware of the position, not only of his own aircraft, but also of nearby aircraft, ground
vehicles (or even terminal buildings) in order to taxi safely and avoid any collisions. The
ability of the pilot to successfully taxi along the path depends on various factors. These
include the type (size) of aircraft that the pilot operates, the amount of traffic at that
particular instant at the airport, the surface visibility, the weather and surface conditions,
and the familiarity that the pilot might have with the specific taxiway system. We must
remember also, that the pilot during his taxi, is usually assisted by the ground controller
who directs him along his taxi route and provides him with information about
surrounding obstacles. It is important to note though, that the pilot is the one who makes
the final decisions and may override the controllers directions. For example, a controller's
request for a landing aircraft to use the first available exit can be ignored, or the pilot may
insist on taxiing to the starting end of a runway rather than start from an intermediate
point.
Such factors as the human element cannot easily be quantified and often introduce
variance into the events that we want to measure, and therefore must be taken in to
account in the final analysis. Among many surface motion variables that can be
measured, those of interest are: the approach speed of a landing aircraft, its landing speed
profile during roll-out, the runway occupancy time, the exit used, the exit velocity, the
time required to cross runway intersections, and the taxiing velocities on different
segments of the taxiway system. The analysis of these variables in conjunction with the
major factors that affect them will be the focus of this thesis.
Chapter 2
The Measurement Task
2.1
Introduction
The first section of this chapter describes the existing runway and taxiway system
at Logan airport in Boston so the reader can get a better understanding of the airport
layout and better relate the measured variables. The second section, discusses the main
elements of the data collection method that was employed. Finally, the last section
provides information about the different days that the data was collected. Included in this
information, is a description of the weather and surface conditions as well as any
particular events that occurred during the collection period and which might be of interest
in the later stages of the analysis.
2.2
The Runway and Taxiway System
Boston Logan International Airport lies at the edge of Boston harbor, surrounded
by water in the majority of its perimeter. The commercial and residential area of East
Boston is adjacent to it while the Winthrop area lies across the harbor (Figure 2.2.1).
Figure 2.2.1
Logan is the dominant airport ( 70 % of the passenger traffic 1 ) in a regional airport
system that also includes airports serving Hartford, Manchester, Worcester, Hyannis,
Portland, and Providence. It has five runways with four of them (4L, 4R, 33R and 27)
1
Boston Logan International Airport Capacity Enhancement Plan, October 1992 published by the FAA.
capable of handling large transport aircraft. Three of these runways (4R, 33R and 27)
have instrument landing capability. The configuration of the runways is rather complex
(Figure 2.2.2), as they intersect six times with each other.
t I,--
I
-
Figure 2.2.2
Typically, peak hour demand is 100 operations per hour. The serving capability
depends on the runway operating configuration, and can vary from 46 operations per hour
during the most restrictive IFR conditions to 120 operations per hour during good VFR
weather 2 . This fluctuation is primarily due to the lack of parallel runways within
adequate spacing between them for simultaneous IFR approaches under certain weather
conditions. Consequently, at certain times all landings must be sequenced into a single
arrival stream, thus lowering the airport serving capability. The high proportion of
commuter aircraft operations at Logan further deteriorates the airport effective capacity,
as larger separations maybe needed under certain runway configurations to safely
accommodate these smaller sized aircraft due to the wake turbulence considerations
during mixed (in terms of size) operations. In addition, in order to keep the noise levels
that the nearby communities experience within reasonable levels, the Massachusetts Port
Authority has imposed certain regulations that further complicate aircraft operations.
Specifically, only certain runway configurations can be used at night and airlines are
required to conduct a specific portion of their Logan operations in Stage 3 equipment 3 .
The configurations 4 that are used most often at Logan are:
Table 2.2
Configuration
IFR
VFR
Arrivals
Departures
Arrivals
Departures
1
4L & 4R
4L, 4R & 9
4R
4L, 4R & 9
2
22L & 27*
22R & 22L
22L
22R & 22L
3
33L & 33R
27 & 33L
33L
33L
15R & 9
9, 15R & 15L*
4
* These configurations employ hold-short procedures.
15R
15R & 9
2,4 Boston Logan International Airport Capacity Enhancement Plan, October 1992 published by the FAA.
3
Summary of Logan's Noise Abatement Rules and Regulations published by Massport.
Due to the complexity of the runway system, various procedures for intersection
departures and hold-short arrival are often used.
The taxiway system (Figure 2.2.3) consists of two main circumferential taxi lanes
(inner & outer) around the perimeter of the terminal building area with smaller taxiway
segments supporting the traffic towards the gate area. Longer taxiways also exist to feed
the outbound traffic to the departure runways, and the incoming traffic to the terminal
area. Runways 4R, 33L/15R and 27 have additional high speed exits conveniently located
so that the landing aircraft can vacate the runway as soon as possible, and then there are
various common taxi paths from these exits to the gate areas. For example, the high speed
exit most commonly used for runway 4R is exit 12 (link A29-A56) , and crosses runway
4L (used only for turboprop landings and takeoffs in this case) before joining taxiway N
(link A74-A75) to return to the terminal area.
Rotate)
Zomn
Ou
Zoom
Full View)
Redraw)
Display
Quit
Edit
Zooming to full airport view...
,IC=
A97
A9
80
~9
8 1
85
AGO
C'LL
9
e
D
A
]A
819
f**
APR
bythe ation Laboratory, IT,Cambridge, assachusett.
Developed
Fliht Tran
Dfeveloped by the Flight Transportation Laboratory, MIT, Cambridge, Massachusetts.
2.3
Data Collection
In the past, the techniques used to study the aircraft motion on the surface of
airports fell into two major categories : those that involved direct observation of the
traffic through a number of observers out in the airfield (the MIT study) and those
involving indirect observation through the use of radar or other types of monitoring
equipment.
Each one of the two methods has its own advantages and disadvantages. The
indirect radar method is more complex and requires expensive equipment but is fairly
accurate, imposes no interference in the traffic, and once operational can be employed for
long time periods. On the other hand, the other method (direct observation) is less
complex but requires a large number of observers, often in coordination with each other,
which involves intense manual effort and as one might suspect, and provides changing
levels of accuracy. Nevertheless, both methods require the authorization and cooperation
of the local FAA and airport authorities.
Luckily, in our case the MIT Lincoln Laboratories had installed an experimental
ground surveillance system that gathered data from Logan airport in Boston. Specifically,
the data included target position reports from the ASDE-3 surface surveillance radar and
the ASR-9 radar from the near terminal airspace traffic information. These two outputs of
the surveillance sensors were integrated by a combined tracking system that also provided
derived information about the velocity, heading and acceleration of the targets 1 . A
simultaneous interface with the ARTS computer was often used and then information
about the aircraft type and flight number was made available.
2.4
Available Data
As mentioned earlier, Lincoln Laboratory had installed a surface traffic data
gathering system at Logan airport in Boston in 1993 for the development and testing of a
runway status lights network (ASTA-1) to help prevent runway incursions. As much as
ninety hours of traffic data were collected for this purpose. Approximately twelve of
these ninety hours were preprocessed by Lincoln Labs, and made available to this study
for further processing and analysis of the aircraft surface movements. These twelve hours
came in the form of 10 separate blocks of data, each corresponding to an individual data
gathering session. These blocks cover a variety of runway configurations, weather
conditions, traffic levels, and high or low visibility conditions. A brief description of the
available blocks of data follows :
Block-i
Day:
Thursday, April 1, 1993
Time: 16:00-17:15 Local
Table 2.4.1
Runway Configuration
Arrivals
4R
Weather / ATIS
Departures
Switched to
9, 4L
Arrivals
Departures
4R
4R
Table 2.4.2
Temp Ceiling
Visibility
Wind
Comments
n/a
500ft ovc
2 miles
50*@ 15knots
Rain & Fog
n/a
1800ft ovc
2 miles
40*@24knots
Rain & Fog --> Thunderstorms
n/a
1100ft ovc
2 miles
400@ 20knots Rain & Fog --> Thunderstorms
Block-2
Day: Friday, March 26, 1993
Time: 10:35-11:35 Local
Table 2.4.3
Runway Configuration
Arrivals
Departures Switched to Arrivals
4
9
33L, 27
Table 2.4.4
Weather / ATIS
Temp Ceiling
Visibility Wind
55 * F 800ft. scat 7 miles
Departures
33L, 22R
Comments
160"@7knots Heavy Traffic, All Taxiways OK.
Block-3
Day: Wednesday, April 21,
Time: 17:35-18:35 Local
Table 2.4.5
Runway Configuration
Arrivals
Departures
27
22R
Table 2.4.6
Weather / ATIS
Con iments
Visit iility Wind
Temp Ceiling
a65 *F
j
15mi les
2500ft ovc
180"@ l7knots Healvy Traffic.
Block-4
Day: Tuesday, March 26, 1993
Time: 13:15-14:15 Local
Table 2.4.7
Runway Configuration
Arrivals
Departures
15R
9
Table 2.4.8
Weather / ATIS
Comments
Temp Ceiling
Visibility
Wind
50 * F
12 miles
140*@7knots Busy Traffic, later quieting down
Sunny
Blc~-5
Day: Thursday, March 11, 1993
Time: 14:50-16:10 Local
Runway Configuration
Table 2.4.9
Arrivals
33L
Departures
22R, 33L
Table 2.4.10
Weather / ATIS
Temp Ceiling
Visibility
Wind
Comments
40 * F 5500ft
15 miles
3000@ 15knots Snow in the morning
Block-6
Day: Wednesday, March 31, 1993
Time: 19:15-20:15 Local
Runway Configuration
Table 2.4.11
Arrivals
4R; 4L
Weather / ATIS
Departures
9, 4L
Table 2.4.12
Temp Ceiling
Visibility Wind
Comments
42 "F
15 miles
Snow in the morning
6500ft
110"@ 8knots
Block-7
Day: Saturday, March 13, 1993
Time: 09:45-10:45 Local
Table 2.4.13
Runway Configuration
Arrivals
15R
Weather / ATIS
Departures
9, 15R
Table 2.4.14
Temp Ceiling
Visibility Wind
Comments
31 * F
5000 ft
5 miles
127" @ 8knots ILS approaches 15R
33 * F
3800 ft
12 miles
110"@ l7knots ILS approaches 15R, light snow
32 * F
1500ftovc
1 mile
110"@ 15knots ILS approaches 4R, light snow
Day:
Wednesday, April 21, 1993
Time: 19:50-20:50
Local
Table 2.4.15
Runway Configuration
Arrivals
Departures
22L, 27
22R, 22L
Table 2.4.16
Weather / ATIS
Temp Ceiling
59 F
2500ft scat
Comments
Visibility
Wind
15 miles
225@11llknots n/a
Day: Tuesday, March 30, 1993
Time: 07:45-08:45 Local
Table 2.4.17
Runway Configuration
Arrivals
Departures
4
9
Table 2.4.18
Weather / ATIS
Temp Ceiling
Visibility
Wind
Comments
43 * F 700 ft
2 miles
40*@ 12knots
Light drizzle & Fog
Block-10
Day: Wednesday, April 21, 1993
Time: 09:00-10:02 Local
Runway Configuration
Table 2.4.19
IArrivals
Departures
22R, 27
22L, 27
Table 2.4.20
Weather / ATIS
Comments
Temp Ceiling
Visibility
Wind
60 * F 2500ftovc
15 miles
225"@ l6knots n/a
Chapter 3
Data Analysis
3.1
Introduction
The first section of this chapter describes the preliminary data processing that was
undertaken along with various problems that were encountered due to several data
irregularities. The second section, provides a detailed runway analysis that includes
information about occupancy times, exit velocities, exit use, and landing velocities
profiles. The next section analyzes the intersection crossing times and the particular level
of use of each intersection. Finally, an analysis of the taxiway system is presented.
3.2
Preliminary Data Processing
As soon as the ten blocks of collected data were received, all the possible ways to
process and analyze the available information were considered. Each block of data
consisted of information about all the targets that were picked up by the ASDE-3 and
ASR-9 radar during each gathering session. Every target had its own ASDE and target ID
and contained among other things, position information in terms of x and y coordinates
with respect to the radar location, its derived velocity, acceleration, and heading, and the
corresponding time stamp for specific data items, measured in seconds from 0:00 GMT.
In addition, some targets included information about the aircraft type and airline flight
number (Table 3.2.1).
Tt:11584
Length: 27
Start time: 57995.3
-
Target
ID
11584
11584
11584
11584
11584
11584
11584
11584
11584
ASDE
ID
5575
5575
5575
5575
5575
5575
5575
5575
5575
Time
57995.375
57997.126
57998.878
58000.629
58002.382
58004.134
58005.885
58007.637
58009.388
End time: 58588.6
Stamp State ASF
ID
1
TAX 993
2
TAX g93
3
TAX
3
4
TAX ,93
5
TAX 76
6
TAX 76
7
TAX g76
8
TAX g76
TAX g76
9
States: DEP TAX STP
(deg)
-.
Position
North: -82.7.07
North: -827.22
North: -802.69
North: -793.17
North: -783.10
North: -769.83
North: -756.49
North: -742.40
North: -732.36
East: 242.13
East: 249.34
East: 252.97
East: 256.41
East: 258.79
East: 263.59
East: 267.36
East: 271.39
East: 276.04
(knots)
Heading Speed
10.7
31.3
14.8
14.9
12.3
17.1
16.9
16.4
22
13.9
10.1
17.9
15.1
12.5
14
15.2
16.4
13.9
(g's)
Accel.
0.0874
-0.0552
0.113
0.0166
-0.042
0.0116
0.0215
0.0215
-0.0348
Flight
Num
SR188
SR188
SR188
SR188
SR188
SR188
SR188
SR188
SR188
Type
B747
B747
B747
B747
B747
B747
B747
B747
B747
Table 3.2.1 : Typical sample information about a target inside a block of data.
The individual position reports for every target, constituted a very large amount of
information, and in order to be useful, had to be related again to the surface layout of the
airport. A graphical replay of the information of the available data was needed since it
would enable us to visualize the actual aircraft motion, check the analysis output, and
explain any possible counterintuitive findings.
Recently, the Flight Transportation Laboratory at MIT had designed and
developed an aircraft Ground Motion Simulator (GMS) to realistically simulate airport
ground activity. The GMS simulates the environment at any arbitrary airport and provides
high quality graphic views, in color on UNIX workstations. This system has an internal
aircraft position generator that provides the simulation with motion updates. It was
decided to use the GMS system for visualization purposes, after bypassing its position
generator function and writing the necessary code to provide it with the actual aircraft
motion information from the Lincoln Laboratory data.
As a second step, the Logan airport geometrical layout along with its features
(terminal buildings, hangars, etc.) had to be inputted in the GMS system (Figure 3.2.1).
Next, the underlying network of nodes and links had to be inserted in GMS format data
files in order to define the runways and taxiways of the airport (Figure 3.2.2). Table 3.2.2
lists the series of nodes that define every taxiway. The next step was to write the
computer code that will associate every aircraft position with an airport link in order to be
able to automate the data reduction process. Various computer subroutines were also
written to perform other preliminary analyses of the recorded data. As a result, computed
values were obtained for approach speeds, exit velocities, intersections crossing times,
and various taxiway velocities. A more complete discussion of these values, and their
significance will start in the next chapter. During the analysis process various routines
had to be modified in order to overcome some irregularities in the collected data.
Series
Taxiway
U
B08
A92
V
B09
A93
X
A50
A20
A42
A35
A36
of
Nodes
Y
A55
A56
A29
INNER
A97
A98
A80
A99
BOO
B07
B08
B09
A95
A96
A97
A79
A81
B10
AOO
A10
A80
B10
A20
B11
A71
A72
A20
A40
N2
A37
B14
A38
B04
A60
A46
A33
A32
A31
A18
A19
A73
A59
OUTER
2
1
2
A
B06
A10
NB
A63
A64
2A
A32
A26
B03
A70
A21
A18
B14
B13
A49
A48
A48
A14
A98
A81
A79
A77
BOO
A83
A02
A03
01
A
B
00
01
B01
B02
B03
B04
B05
B06
A82
A83
A84
A85
A86
A70
A60
A50
A91
A92
A93
A94
A95
A96
A39
A37
A31
A25
A24
A18
A17
A16
A12
A01
A02
A04
A07
A90
A06
A43
A44
A09
A08
A47
A48
A23
A22
A75
A74
A73
A71
A69
A66
A63
A62
A51
A52
A43
A35
B13
A34
A33
A27
A23
A49
A24
A78
A44
A45
A04
A05
B02
A86
A66
A65
A99
A82
B05
AOO
B06
A50
B02
A86
A85
A53
22L
A61
A62
A64
A65
A57
A29
A14
A45
A09
A28
A27
A26
A25
A87
4R
A54
A53
A51
A47
A46
A43
A21
A22
A14
A52
A53
A75
A74
A13
A89
A17
15L
A08
A28
A36
A34
Runway
4R
4L
15R
15L
22R
A68
A67
A72
A59
A42
A88
Nl
A39
4L
33L
All
A12
A13
A01
B11
A78
A77
A76
15R
33R
27
A58
A15
A57
A06
A55
A05
A54
A03
A38
B14
N1
N2
9
Table 3.2.2
Q
*
co
m
Rotate)
Zoom In)
Zoom Out)
Full View)
Redraw
Display
)
Qut)
Starting simulation...
JI.._LJ,J
Developed by the Flight Transportation Laboratory, MIT. Cambridge, Massachusetts.
J
3.3
Data Irregularities
Due to the performance of the radar tracking system on the surface of the airport,
frequently during the gathering session, an aircraft target was dropped and then picked up
later on by the radar. The result was that in the data file, two different targets with
separate IDs (identification numbers) could in fact have been the same aircraft, and the
intermediate information about the aircraft movement between the time that the aircraft
was dropped from the radar and then picked up again was not available. Another
irregularity was the fact that not all targets had information about the aircraft type or
flight number. This limited the classification of results according to aircraft size to only
those targets where that information was available. In addition, this prevented us from
identifying targets that were not aircraft but rather other ground vehicles moving on the
airport surface and therefore might have infected our results if they were on the runways
or taxiways. Indeed, Blocks 3, 8 and 9 did not include any information about aircraft
types and flight number because the required computer tap was not in service during the
collection period.
3.4 Runway Analysis
3.4.1 Runway Occupancy Time During Landing
Runway occupancy time is the time over which a runway is effectively blocked
(occupied) to any other traffic by a single landing or departing aircraft. As such, it
potentially affects the traffic capacity of that runway. In the case when the runway is used
only for landings, the runway occupancy time and its potential variations currently do not
significantly affect the overall runway capacity as the inter-arrival radar separation
standards of the approaching aircraft cause spacing which is almost always greater than
the occupancy time. On the other hand, if the runway is used for mixed arrivals and
departures, the landing occupancy time becomes more critical. In that case, the shorter the
landing occupancy time, the more the time allowed to insert a takeoff between landings.
This results in higher runway operational capacity, and smaller delays for the departing
aircraft.
Average Occupancy Time During Landing
Data Block
1,6,9
Figure
1, 2, 3
4L
n/a
n/a
22R
22L
27
9
33R
33L
15R
15L
n/a
8, 10
2,3,8,10
n/a
n/a
2, 5
4, 7
n/a
n/a
4, 5
6,7,8,9
n/a
n/a
10, 11
12, 13
n/a
Runway
4R
II
Table 3.4.1.1
The following tables (3.4.1.2 to 5) correlate each exit link of every runway from
the GMS airport layout (Figure 3.4.1.1 to 3) to an exit number for the graphs that follow.
Link
Exit Number
Runway 9
12
Runway 27
1
2
3
i
4
5
6
7
I
A06-A90
11
10
II
A05-A04
A03-A02
I
9
11
A13-A01
JI
8
7
6
11
A17-A18
A08-A09
A09-A28
8
1F 5 11
9
4
10
3
B14-A37
2
N1-A39
1
11
II
II
1
12
A36-A35
A34-B 13
N2-A40
Table 3.4.1.2
Link
Exit Number
Runway 33L Runway 15R
14
ii
1
13
2
12
3
11
4
5
6
7
9
Jj
10
I
A22-A23
IL
9
8
1
A14-A23
A14-A48
1
8
II
12
13
14
7
6
10
11
A12-A16
A13-A17
A01-A19
A21-A18
I
II
II
A52-A51
II
A53-A51
5
A53-A85
4
A 75-A83
3
B11-B10
2
1
1
II
Table 3.4.1.3
A78-A81
A77-A79
Rotat)
Zoom n)
ZoomOut)
FullView)
Redraw)
Display r)
Edit
t)
Zooming to full airport view...
29
A
00
t
12 Al
A7
.A97.~
.J
e
Developed by the Flight Transportation Laboratory. MIT Cambridge, Massachusetts.
Rotate)
C]
We
Zoomn)
ZoomOut)
FullView)
Redraw)
Display r)
Edit r-
it
Zooming out by 1.50...
-Q
J4
II ..
T
LT
AD
Developed by the Flight Transportation Laboratory, MIT, Cambridge. Massachusetts.
19
J
-J
Exit Number
Link
Runway 4L
Runway 22R
1
2
3
14
13
12
A39-N2
N1-N2
A42-A20
4
11
A43-A10
5
6
10
9
A43-A00
A46-A60
7
8
8
7
A51-A86
9
6
A53-A75
10
11
5
4
A53-A85
A54-A74
12
3
13
2
A72-A71
14
1
A67-69
I
I
A47-A70
A59-A73
Table 3.4.1.4
Exit Number
Link
Runway 4R
Runway 22L
1
16
A12-A16
2
15
A26-A32
3
4
5
6
14
13
12
11
A27-A33
A28-A36
A09-A44
A45-A44
7
10
A24-A46
8
9
A49-A48
9
8
10
7
11
6
12
13
5
4
II
14
II 3
15
16
2
1
]
1
A23-A48
A14-A48
A14-A52
I
A29-A56
A57-A55
L
Table 3.4.1.5
A65-A66
A64-A63
A62-A63
9
Rotate
Zoom
Zoom Out
Full View
Rd
Display
Rotating airport by 30 degrees..
Ca
Ue
SQ
t
Developed by the Flight Transportation Laboratory. MIT. Cambridge. Massachusetts.
The occupancy time during landing is measured from the moment the aircraft is
over the runway threshold until the time it has turned in the exit and its tail has crossed
the runway edge. Since our aircraft motion data was in the form of a series of discrete
radar hits (approximately every 1.7 sec on the surface), the time between the first hit
inside the first runway link and the first hit inside the exit link was used. In this way the
size of the error was minimized. As expected before the analysis of the results, occupancy
time tends to increase with the distance of the exit location from the runway threshold.
This is normally true except in some cases ( Figure 3.4.1.4 Exit 12 in Runway 4R and
Figure 3.4.1.7 Exit 11 in Runway 22L) where the particular angle of these exits allow
aircraft to exit with higher speeds, and therefore maintain a higher average landing
velocity resulting in occupancy times similar to exits that are located much closer to the
threshold.
Figures 3.4.1.4 through 3.4.1.16 are graphs of the average occupancy time during
landing for all aircraft types over a single runway and exit for every block of collected
data. It seems there exits a relationship between aircraft weight and occupancy time. We
observe that usually, the standard deviation of the occupancy times are quite small (5-10
seconds) for aircraft using the first exits, unlike for those using exits that are located
further down the runway. Runway 27 under configuration 2 (arrivals 27 and departures
22R) displayed the lowest occupancy time (35 seconds) for aircraft exiting at high speed
from exit 6 (Figures 3.4.1.9, 10, 12). In data block 8 (Figure 3.4.1.11), with similar weather
conditions but at night (20:00-21:00), most aircraft used exit number 8 (low speed) and
the occupancy times were significantly larger (53 seconds). The heavier aircraft tend to
land with higher velocities and require longer landing distances and therefore exit further
down the runway resulting in higher occupancy times. However aircraft using a given
exit have similar occupancy times, independent of aircraft size.
Rauway 4R
Average Occupacy Time During8 Landing
Block 1
80
Stdev.
l
70
20
18
Avg.Oc.Tim.
m
-- 16
N
-
60
-
0
14
-
__
--12 4
50
50
10'S
8
40
u
-
P 30
6
Landing Direction
20
20--
4<
2
0
10
0
t- w
ON
-
-
%
Exit
Figure 3.4.1.4
Average Occupancy Time Dring Landing
Runway 4R
Block 6
80
Avg.Oc.Tim.
20
70
Stdev.
18
16
1
N
14
12
60
-
j50
10 C.
, 40 -
8
-30
20
.'
6
Landing Direction
4
10
2
0
0
Exit
Figure 3.4.1.5
Average Occupancy Time During Landing
Rinway 4R
Block 9
80
I
70
-.
20
18
IAvg. Oc. Tim.
Stdev.
16
60
0
S
O
ep
14
N
12
50
40
- 10 or
30
6
8=
Landing Direction
20
4
10
2
O
1
*I
Cl
I
.1
tf
A0--
.
\o
r-
X0 a-
0-
Exit
Figure 3.4.1.6
44
C
t\
ziiSy~n~n
-- L-~1-*r-I~-.~m~ .1--l~Y1~4
r~~h~L"4~----
-
_~,
~__
.,.~.;1.-_1._,~.,__~,~__,_~_;____
_
Average Occupancy Time During Landing
Runway 22L
Block 8
80
70 -
A.O.T.
D
L
-
*Stdev.
-N
60 h
50
03.
-
40 30
-
20 Landing Direction
rL'Ur .1
10 II
-Ii
0.
*
a*V)s
~
rCt
I
P.M
c
-
I
L,.u
IM
--
IM
a-
IM-
I
e-
Exit
Figure 3.4.1.7
Ave ge Occupancy Time During Landing
Runway 22L
Block 10
A.O.T.
20
80
S
60
PZ
50
- 18
Stdev.
70
-
.16
1
14
12
N
Landing Direction
S
10
40
8
S30
6
0
20
Exit
Figure 3.4.1.8
.
.
_
.
l~---~-rr~Fnl
Average Occupancy Time During La ding
Runway 27
Block 2
80
Avg.Oc.Tim.
-20
- 18
Stdev.
- 16
N
- 14
-* 12
1
70
e
60
U
t..
50
+
70
- -
40 -
,<
Oc.Ii
IAvgI
- 10
mIi.
30 Landing Direction
20 10 -
rlrlrNo
-2
I
I
0.
7
6
5
4
3
2
1
: *L
-8
-6
.4
8
9
10
11
0
12
Exit
Figure 3.4.1.9
Average Occupancy Time During Landing
Block 3
K
80 70
U
SaM.
-20
- 18
Avg.Oc.Tim.
-
- 16
- 14
- 12
Stdev.
60-
0
Ramway 27
N
-U
50 -
-10
-8
40
30 -
t_
20
10 0-
-6
-4
Landing Direction
-
--
- S1
*
-
~-~-~-~-~
I 2
3
.
-
I
]
I
4
5
sJ
-
,I
-2
-
7
6
-0
8
Exit
Figure 3.4.1.10
9
10
11
12
-4 a sm
L
U
h
Average Occupancy Time During Landing
Runway 27
Block 8
80
-
70
Avg.Oc.Tim.
20
Stdev.
18
16
14
N60
CL-
50
12
t 10
40
e
5
I
3
8
30
30
6
Landing Direction
20
%
4-2
10
0
0
1
2
4
3
7
6
5
8
9
10
11
12
Exit
Figure 3.4.1.11
Average Occupancy Time During Landing
Runway 27
Block 10
80 -
Avg.Oc.Tim.
20
70 "
Stdev.
50
18
16 .
14
12
40
10
S60--
6 - -
86
Landing Direction
30
<
N
20
4-
10
2
0
0
1
2
3
4
5
7
6
8
Exit
Figure 3.4.1.12
9
10
11
12
t
2a
-~Fpl)(-? 4~------
-^c"ir;iic~anaaharrr;h-cu*ru4paaL~ia3~1
Average Occupancy Time During Landing
Raway 33L
Block 2
80
Avg.Oc.Tim.
20
70
Stdev.
18
16
14
12
N
,
50
tZ
10
S 40
30
Landing Direction
20
-
8
6
-4
2
10
Z
b
0
0 Cl
en
0
N00
-O
0
CqT
Exit
Figure 3.4.1.13
Average Occupancy Time During Landing
Ranway 33L
Block 5
Avg.Oc.Tim.
20
80
Sdoccup
70
Landing Direction
E
60
50
18
16
16
14
12
40
o-08
30
10
20
4
10
- 2
6
mC 4
I
\O
-
00
Exit
Figure 3.4.1.14
0
C
enTExit
z
co
Average Occupancy Time During Landing
Rnway 15R
Block 4
111888l Avg.Oc.Tim.
"-
-20
80
70
60 -
a
a
t-
U:E
v
-18
Stdev.
-
-i--&--
-16
- 14
N
--
N
50 -
- 12
40
, 10
-
8
30 -
D
Landing Direction
20 -
I IIIhhhI
10 -
0*
-6
-4
I
I
m
M
f
T-u.
I.
r\
-
m
m
a
~
-I
]
I
0%
a
00
Pm
r-
i
lr
I-
Srv
I-
z
3 .8
3
-2
U.
1. .
aC w
I.-
I
m
B
B
0
-
Exit
Figure 3.4.1.15
Average Occupancy Time DTring Landing
Block 7
Ruaway
S15R
U
1.1.
v
U
z
ela
4J
I-
Landing Direction
S(
0
0N
r-
00
Exit
Figure 3.4.1.16
49
CExit
*x
-r~
eruu;srr;
ll(
.~_~......_
- -^-iiC;;i-:-"-
l~-r.rr~ic c,~c~r
rl
_ _
3.4.2 Exit Use
Figures 3.4.2.1 through 3.4.2.13 represent graphically the exit use both for all
aircraft that landed in every runway and for each aircraft (weight) class in every data
block that information about the aircraft types were available. If information was
available during data collection, the exit use per particular aircraft class is presented.
Each aircraft is classified according to its weight into one of the following three classes :
Class
Weight Range ('000 LB)
Heavy
300-900
Large
Small
12.6-299
0-12.5
Table 3.4.2
The major observations are that, as expected, the probability of exit is related to
the aircraft size (weight class). Hence, heavier aircraft tended to use exits that are located
further away from the runway threshold while smaller sized ones required shorter landing
distances and exited earlier.
A second factor that affected exit use was the specific turning angle of the exit.
This angle (which could have been acute, right, or obtuse) provides a measure of the
difficulty of using each exit. As a result, independent of runway, most aircraft tended to
prefer the use of obtuse angled (high speed) exits.
I.,,,,.n nrr~.-
-_-.p~-__I ~*TYC~~L~
i.
l?~_i~DC;~EF-~-Q-~L-il
I---LI IIC--s~_I~=~
IP
Figure 3.4.2.1
Exit Use per Aircraft Class
Runway 4R Block 6 23 Landings
35
30
25
* Small
20
O
Large
Landing Direction
Heavy
15
10
'i
I
SI
-
|
e
4Sa
•
00O
0
i
F..
-
Ci
e
I
1
-f
Exit
Figure 3.4.2.2
a
7
\O
Exit Use Runway 4R Block 9 29 Landings All Classes
35
30
25-
I
All Classes
S20
S15
10 .
Landing Direction
5
0o
I
II-I
I
I
-I--l-
Exit
Figure 3.4.2.3
I
1..7 L--..i~YV~33~-
..:^.....,,II~~,
.~l-r_--r~u..L~
~.r-~;~I~LFP~I~_~E ~
(^i~
~-~9~_,?LII
Figure 3.4.2.4
Runway 22L Block 10 16 Landings
Exit Use per Aircraft Class
40
8 Small
35 30
O Large
-
25
* Heavy
20
15
Landing Direction
10
n
-.
a
I
\O
I
I
II
I
'r
I
'i
1
1
I
1
U •
-
cc
-
-
-
0
-
-
1
O\
00
Exit
Figure 3.4.2.5
53
r-
L
I
I3
*I
C
J
J
M
n
NT
M
e,
-."
_i ___
___~I.I
__ ~)__I____Yil____IU__I-R-C---41_9~.--.r- ... __ - I_; ..----
Figure 3.4.2.6
Exit Use Runway 27 Block 3 8 Landings All Classes
100 90 80 -
2
70 -
All Classes
6050 40
Landing Direction
30.
pii
20
10
0
I
-u---I
I
I
I
--- T
J
1
2
3
4
5
7
6
Exit
Figure 3.4.2.7
.
r1
8
9
10
11
12
-"~~--p---
Exit Use Runway 27 Block 8 19 Landings All Classes
100
90
80
E
70
All Classes
60
J
50
Landing Direction
40 ,
30
20
10
1
4
3
2
11
10
9
8
7
6
5
I
I
II
I
I
I
I
I
I
0
12
Exit
Figure 3.4.2.8
Runway 27 Block 10 12 Landings
Exit Use per Aircraft Class
60 50 -
E Small
40
O Large
-7
SHeavy
30
20
Landing Direction
10
a00
I
1
•
I
•
i'
2
I
I
I
3
4
5
I
•
•
7
6
.L
I
Exit
Figure 3.4.2.9
iI
m
•
•
I
I
l
li
l
8
9
10
11
12
-- ---I
--
411*-.---C- l-LU-I
.
I
. -~i~W-i
iLIIY IL
CL
~ 14 11 I1 -^I - C
~
_ __ _i~-*--_~
Exit Use per Aircraft Class Runway 33L Block 2 27 Landings
60
M Small
50
O Large
40
SHeavy
30
20
Landing Direction
l
l
I-
10T-
II
,-
C
C
I
II
I
IC
t
O
I L--.
L --- j-
!-
00
I
O'
,:.M.1,
I
O
Exit
Figure 3.4.2.10
Exit Use per Aircraft Class Runway 33L Block 5 37 Landings
60
50
E]
Small
O
Large
Landing Direction
40
30
S Heavy
20
10
MI
I
O
00
Exit
Figure 3.4.2.11
0
0
I-I
L
rn
1-4
Ci
C
11
1-
; _.__
____li~L~I
LIC~_~_~___;_ _i~lq__~_
Figure 3.4.2.12
Exit Use per Aircraft Class Runway 15R Block 7 28 Landings
45
40
35
30
a
25
E Small
O
Large
* Heavy
20
15
Landing Direction
pp
10
5
0
l
0
ON
00
r-
Exit
Figure 3.4.2.13
O
l
lE
3.4.3 Exit Velocities
The angle of every exit plays a significant role in the exit velocity of the aircraft.
As figures 3.4.3.1 through 3.4.3.13 show, whenever the landing aircraft are using the
obtuse angled exits the exit velocities are significantly higher than the other ninety
degrees or acute angled exits. However, this is only true for high speed exits which are
accompanied with long exit segments and give the pilot room to brake (exit 6 runway 27:
38 knots and exit 5 runway 33L : 40 knots). Exit 8 of runway 4R, although it is obtuse
angled, the short exit segment that follows does not allow high exit velocities. Each
figure presents, using columns the average exit velocity, and with a line, one high and
one low value which corresponds to the average exit velocity plus or minus one standard
deviation (see figure legend). The letter H denotes a high speed exit.
Average Exit Velocity All Classes Rnway 4R Block 1
60 --
Avg. Exit Vel.
50
A.E.V. + SD
40
A.E.V. - SD
" Avg. Exit Vel.
30
S20
Landing Direction
H
W
-
-I
-
H
-
Exit
Figure 3.4.3.1
C
C
M
-00
-
-
Average Exit Velocity AM Classes Runway 4R Block 6
O Avg.
Ex. Vel.
A.E.V. + SD
50
50
A.E.V. - SD
40 " Avg. Ex. Vel.
30
Landing Direction
20
101
H
-_
0
f
I
S
O
3.43.
SExit
Exit
-0 10
sr
Q
Figure
Figure 3.4.3.2
Aveg e xit Velocity Al Classes Runway 4R Block 9
60 --
O AvgExVel
50 -
A.E.V. + SD
40
A.E.V. - SD
• AvgExVel
30
20
Landing Direction
10
N
en
It
%n
,rh
.f.l.
00
. O
Exit
Figure 3.4.3.3
. 0
.. -
.rn.
m
.
.O
Average Exit Velocity All Classes Ranway 22L Block 8
60 T
0
AvgExVel
50 -
A.E.V. + SD
40
A.E.V. - SD
~ 30
AvgExVel
Landing Direction
w 20
S10
0*
.m
- - -a
~---~-~---~
DO
in
t
en
Clq
-
0
-
00
Exit
Figure 3.4.3.5
t-
O
v
11
- IC14
-4
a
Average Exit Velocity All Classes Ranway 27 Block 2
60
O AvgExVel
50
A.E.V. + SD
40
>
- SD
S4A.E.V.
30
AvgExVel
20
0
Landing Direction
S10
H
1
2
3
4
5
7
6
8
9
10
11
12
Exit
Figure 3.4.3.6
Average Exit Velocity All Classes Runway 27 Block 3
OAvgExVel
60 50
A.E.V. + SD
40
A.E.V. - SD
SAvgExVel
> 30
0
Landing Direction
-20
10
10
H
1
2
3
4
i
7
6
5
Exit
Figure 3.4.3.7
61
8
9
10
11
12
Average Exit Velocity All Classes Runway 27 Block 8
60
O Avj gExVel
S50
50
A.E .V. + SD
S40
A.E .V. - SD
o
>
30
Landing Direction
"Av gExVel
- 20
10 -F1
1
2
3
4
5
8
7
6
9
10
11
12
Exit
Figure 3.4.3.8
Average Exit Velocity All Classes Runway 27 Block 10
O AvgExVel
50 -
A.E.V. + SD
40 -
A.E.V. - SD
u
AvgExVel
Landing Direction
>w 30
20
10
0
-
--
-1t
-
-
1
2
3
4
5
.th
H
-1
6
7
6
Exit
Figure 3.4.3.9
62
8
F . .l.
9
10
11
12
way 33L Block 2
Average Exit Velocity All Classes Rm
60
O
AvgExVel
50
A.E.V. + SD
40
A.E.V. - SD
" AvgExVel
30
20
Landing Direction
10
po
-
0
_
1
,
I
I
N
uu I
I
-- -
-
I
I1
I
I
I
l
I
1
Exit
3.4.3.10
Fiure
Figure 3.4.3.10
Average Exit Velocity All Classes R~mway 33L Block 5
O
60
AvgExVel
A.E.V. + SD
50
A.E.V. - SD
40
-
AvgExVel
20
10
I
I
1-4
CI4
SI
I
-- I
I
I
tn
-
I
I
I
-I
i
!-
z
00
-
.
-
-
O
Exit
..
Figure 3.4.3.11
63
rrmfl
~
-
- .- -
0
-
-
•
.-
el
.
.11.
.
em
.
I
i
Average Exit Velocity All Classes Runway S1R Block 4
O AvgExVel
60
A.E.V. + SD
50
A.E.V. - SD
40
" AvgExVel
30
20 -
[]
Landing Direction
10- +
|
I
0
-
C
00 Ex it
0
\
.n.
|
r
I.
c
I
I
i
Exit
Figure 3.4.3.12
Average Exit Velocity All Classes Runway ISR Block 7
O
60
AvgExVel
A.E.V. + SD
50
A.E.V. - SD
40
" AvgExVel
30
Landing Direction
20
10
I
I
I
>.
.i
I
I
..
Exit
Figure 3.4.3.13
64
~ril
1O
I
kI
1*
I.
e
I.I
C'
-
---
-----
------ -
Figures 3.4.3.14 through 3.4.3.22 show the average exit velocities per aircraft class.
These velocities vary from aircraft class to aircraft class but the variation is not consistent
and no conclusions can be drawn in favor of one class or another.
Average Bxit Velocity per Aircraft Class Rmaway 4R Block I
50 45 -
I
40 -
Heavy
O Large
35 30 -
E Small
25 20 15 Landing Direction
10
5
o0
..
I
-
("4
I
n¢
I ,q
I u'n
,0
I t
.
I0
..
.rfil
.
I
Figure 3.4.3.14
-4
C14
IO Cn
N
.
.
W
.
-H~Cb
ll~eB4~LISMn-'i~"'.--cnrra~-B~*M I -..-.
-~""
~I~-
~PLIII~ -~lca;--.---.;
Average Exit Velocity per Aircraft Class Runway 4R Block 6
50 -r
45 40 -
4 Heavy
Landing Direction
35 -
O Large
30 .
25 -
*
Small
20 15 10 50
I
1-4
a
C14
II
I
en
,,
S ..
I--
II
I
i
\O
I
I
I
----
00 E
r.-
.n..4 I"..0~.
-
en
-
kn
Edit
Figure 3.4.3.15
Average Exit Velocity per Aircraft Class Runway 22L BlocklO
50
45
S40
Landing Direction
E Heavy
35
0
>
S30
25
0 Large
0
H
Small
20
15
10
5
Exit
Figure 3.4.3.16
66
;--tl.C~C~i~PI~ii:
Avenge Exit Velocity per Aircraft Class Runway 27 Block 2
50
?
45
40
H
M Heavy
35
;
S30
25
.
20
Landing Direction
O
Large
E
Small
15
to
50
1
I
I
I
0
2
3
I
I
4
5
6
7
8
I
I
I
I
I
9
10
I
I
11
12
Exit
Figure 3.4.3.17
Avenrge Exit Velocity per Aircfaft Class R way 27 Block 10
Landing Direction
..
50
1 45
U Heavy
S40
H
35
30
0 Large
>w 25
U Small
"
.
20
S15
10
5
1
2
3
4
5
7
6
Exit
Figure 3.4.3.18
8
9
10
11
12
I
Average xit Velocity per Aircraft Class Runway 33L Block 2
50 -r
45 Landing Direction
40-
* Heavy
35 -
O Large
30 25 -
0 Small
20 15 10
-
50
I
I
,,.2
"4 q
I
,
en
,
0
o
\O
n
,
I
,
I
O\
I
l
,
l
W-4
i,
Cq
I
I
M
Exit
Figure 3.4.3.19
Average xit Velocity per Aircraft Class Runway 33L BlockS
50
U Heavy
- 45
o40
H
40
Large
H
M Small
35
30
S25
.
20
Landing Direction
15
10
5
MC
'
-
O
00
Exit
Figure 3.4.3.20
68
O 0
CI
eExit
I
:i'------------- -I
^~-
-
"1-1"-'--"~'"~~-"1-p~~~'-~"~~'"""
;:
-,"~,-~,-F~cx~--~.,c~--~li----~----------- I-----
~I;
~'-
Figure 3.4.3.21
Average Exit Velocity per Aircraft Class Ruaway 15R Block 7
50 -
* Heavy
45 -
O Large
40-
B Small
35 30 -
-
25 20-
Landing Direction
15 -
,
1050
I
I
I
IO
1
.!-
\O
SExit
Exit3.4.3.22
Figure
Figure 3.4.3.22
69
69
i kn
I
en
I
e
I
-
I
----F--~a.----.-
3.4.4 Velocity Profiles
3.4.4.1
Landing Velocity Profiles
Figures 3.4.4.1.1 through 3.4.4.1.13 show the landing profiles of all aircraft that
landed in each runway. The aircraft that used a particular exit are grouped together. We
can observe the different exit velocities and the higher or lower deceleration that occur,
depending on the angle and the location of each exit. We must note the fact that some
aircraft actually speed up after the runway threshold.
Landing Velocity Profile Runway 4R Block 1
----------------------
1.2 ----------------------------------
I
----- ---- --- -----------------0.8 ----------------------- MV
T
h
----
----------
0.6 -----------------06e
---
0.4 -"-------------
een
.
Mean Exit 7
K-
Mean Exit 8
Mean Exit I
----
---------h
Sd
-
-
"
--
i
Radar Hits from Runway Threshold
Radar Hits from Runway Threshold
Figure 3.4.4.1.1
Mean Exit 12
Mean Exit
Exit 113
Mean
Landing Velocity Profile Runway 4R Block 6
1.2
1
0
In
,O
0
0.8
.0
I-
-
-
S0.6
5!0
o
~0.4
T
h
r
e
--------------- I
h
6--
Mean Exit 12
------ Mean Exit 13
-X-
0
1
0.2
Radar Hits from Runway Threshold
Figure 3.4.4.1.2
Mean Exit 8
Mean Exit 14
Landing Velocity Profile Runway 4R Block 9
1.2
1
0.8
I-
S0.6
0.4
-X-
Mean Exit 7
0---
-- Mean Exit 8
A
Mean Exit 11
4-
Mean Exit 12
0.2
Radar Hits from Runway Threshold
Figure 3.4.4.1.3
Landing Velocity ProfIle Runway 22L Block 8
.
>0.8 ----------------------
0.6 ----------------
0.8e
0
0.4 ---------
-----
X
---
-----------
x
X ---------
-X-
Mean Exit 6
-&
Mean Exit 8
0
Mean Exit 10
4
Mean Exit 11
T
h
0
Fu
va-
'
%C
0
ay Threshold
Figure 3.4.4.1.4
73
IR
00
C4
-Threshold
%0
0
Landing Velocity Profile Rumway 22L Block 10
1.4
1.2
O
1
S0.8
S0.6
T
h
0.4
----
Mean Exit 6
-X-
Mean Exit 7
O
Mean Exit 10
A
Mean Exit 11
0.2
Radar Hits from Runway Threshold
Figure 3.4.4.1.5
74
Landing Velocity Profile Runway 27 Block 2
1.4
1.2
S1
S0.8
=
T
------------ h -
08o.6
==
1
S0.4
- Mean Exit 6
0
0.2
0
N
0
c
%0
en
-...,..4
0
- a4
in
00
"-1
q
4
Radar Hits from Runway Threshold
Figure 3.4.4.1.6
75
r-
V
-6
0
4
m
(4
.0
4
Landing Velocity Profle
Runway 27
Block 3
1.2
1
0.8
A
0.6
pse
----------------
0.4
0.2
o
t-
~
.4
C~c
4
00
V- -
t
.
T-
ON
04
0
c%
0
n
'0C
Radar Hits from Runway Threshold
Figure 3.4.4.1.7
C-4 %nh 00
V-
Landing Velocity Profyle Runway 27 Block 8
----------------- m------
-------------------- -------
0.8 -
0.6 -
------------- -- ------------- T
0.4
h
r
2
------
-
-
-"
Mean Exit 8
e
-------------r--1
1
d
0.2
--
------------
0-----
N±I I H-I 1+H +H4IIH
AI
M
------
N
SV)
-
4C4N
Radar Hits from Runway Threshold
Figure 3.4.4.1.8
00
-4
( C4
Landin Velocity Profile Runway 27 Block 10
1.4
1
0.8
0.6
---
----------T--h
r
e
0.4
h
0.2
Radar Hit from Runway Threshold
Figure 3.4.4.1.9
Landing Velocity Proffle Ruway 33L Block 2
1.6
1.4
1.2
1
0.8
-----
0.6
-
0.4
4-"
Mean Exit 8
X-
Mean Exit 10
A
Mean Exit 12
0.2
0
Radar Hits from Runway Threshold
Figure 3.4.4.1.10
Mean Exit 5
Landing Velocity Profile Runway 33L Block 5
1
0.8
0.6
r
e
-----
Mean Exit 5
s
0.4
3-
h
------------------
0
----
0.2
Radar Hits from Runway Threshold
Figure 3.4.4.1.11
80
~U' I--UIICII Ill
~-tl~i~C -~D~CI
UI
~~
rnE~I~I~CI~
IC*~MI*l~
~
e~
ll*r)ll~
B
Mean Exit 8
--
Landing Veocity Profile RnPway 15R Block 4
1
0.8
0.6
-----------T
h
r
e
0.4
--
&-X-
0.2
Radar Hits from Runway Threshold
Figure 3.4.4.1.12
-
Mean Exit 7
Mean Exit 10
Mean Exit 11
La8ding Velocity Profile Runway 1SR Block 7
1.4
1.4
-------------------------------- ----------- --------
IV
1.2
xhxx xx
X
X-
Mean Exit 4
--- X-
Mean Exit 5
--- -T
13
0.-
e
SrI
d
oo
dT
X
0. --
- -
--
Mean
-------Mean Exit
Exit 81
-----------------
Radar Hits from Runway Threshold
Figure 3.4.4.1.13
82
82
Mean Exit 11
.
-
Final Approach Velocity Profiles
3.4.4.2
In the final approach velocity profile figures ( 3.4.4.2.1 through 3.4.4.2.13) we
observe that in most runways, there is a small deceleration during the final approach. In
some runways, 27 in particular, this deceleration is quite significant (Figure 3.4.4.2.6). In
the same runway, during night operations (Figure 3.4.4.2.8) we see much more smooth
final approach velocity. In runway 15 R, we note that even under different weather
conditions, aircraft seem to accelerate before the runway threshold.
Overall, the standard deviations of the landing velocities fall between a value of
0.2 for a period of approximately 10 radar hits before the threshold and from then on it
increases significantly. This could be due to the changing size of the available number of
data points (radar hits), as further away from the threshold some aircraft target are not
picked up by the surface radar.
Final Approach Velocity Profile
1
-
-
-.
,m
Runway 4R Block 1
-"---
- 0.9
-
S0.7
,-- Mean
,-------------
--
0.6
' 0.5--------------M0.4
E*
S.50.3
0
1
2
3
4
5
6
7
8
9
10
11
12
Radar Hits from Runway Threshold
Figure 3.4.4.2.1
13
14
15
16
17
Final Aproach Velocity Profile
Runway 4R Block 6
1.2
1
0.8
0.6
0.4
0.2
0
1
2
3
5
4
6
7
8
9
10
Radar Hits from Runway Threshold
Figure 3.4.4.2.2
Final Approach Velocity Profile
Runway 4R Block 9
1.2
.
m--------------------------------
m..--------.--.m
---------------
1
0.8
---------------------
--------------------
Mean
--- --
0.6
---------------------
------------13
Stdev.
------- ------ m--------
---------------------
0.4
0.2
0
1
2
3
4
5
6
7
8
9
Radar Hits from Runway Threshold
Figure 3.4.4.2.3
84
10
11
12
Final Approach Velocity Profile
----------------------------------
--------------------
1.2
10.8
-----------------
m---------------------
-
E
---
-- - -- -- ----
Z0.6
Mean
-- --- -- - --- - --- -- -- -- -
-- - -- -- -- --S
S0.4
Runway 22L Block 8
tdev.
---- ------------
---------------------
S00.4
.2
. .
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
.
17
.
18
.
19
Radar Hits from Runway Threshold
Figure 3.4.4.2.4
Final Approach Velocity Profile
1.21
1
-----------
Runway 22L Block 10
--------------------------------
i
m------------------------------------m-
Mean
S--
--- Stdev.
-----
--------m
0.6
0.4
0.2
-----------------------------------------------------------
0.2
1
2
3
4
5
6
7
8
9
10
11
12
Radar Hits from Runway Threshold
Figure 3.4.4.2.5
85
13
14
15
16
Final Approach Velocity Profile
Runway 27 Block 2
-----------------------------------------------------------
1.4
---
1.2
0.8
------------------Mean----------------Stdev.
S0.6
:
------------------
0.4
---
0.2
-
.-- mm ------------
------------------
-------- m ------
-------------------- m --------- m -------------------- m....
0
1
2
3
4
5
6
7
8
10
9
11 12
13 14
15 16 17
18
19 20
Radar Hits from Runway Threshold
Figure 3.4.4.2.6
Final Approach Velocity Profile
----------------------------------------------------------
1.2
..----
!
Runway 27 Block 3
- -
--------------
----------
0.8
0.8 -----------------------------------------------------------n
Mean
0.6
S-------
Stdev.
-----------------------
S0.4 --------------------
1
0.2 -----------------------------------------------------------0.4
1
2
3
4
5
6
7
8
9
10
11
12
13
14
Radar Hits from Runway Threshold
Figure 3.4.4.2.7
86
15
16
17
18
19
Final Approach Velocity Profile
0.9 -----
RInway 27 Block 8
-----------....---....-------------------------------------
0.9
0.8
----------------------------------------------------------
e0.7
----------------------------------------------------------
~0.6
---------------------
-
Mean
-
--
-
0.5
0.4
*--*-Stdev.
--
. 0.3
0.2
--m--------------------------------------------------------
0.
1
2
3
4
5
6
7
8
10
9
11
12
13
14
15
16
17
Radar Hits from Runway Threshold
Figure 3.4.4.2.8
Final Approach Velocity Profile
1.2
1.2
Runway 27 Block 10
----------------------------------------------------------
0.8
- - -
-
--------------------
-----"-
0.6
------------------
Mean
LFStdev.---
_-
------ ---------------
0.4
---------------------
0.2
---------------------------------- -----------------------
0.2
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
Radar Hits from Runway Threshold
Figure 3.4.4.2.9
87
16
17
18
19
20
Final Approach Velocity Profle
1.2
----------------------------------------------------------
1
r
-n.--
"-
-
-
-
-
-
-
----------------------------------------------------------
0.8
I0.8
-
--
-
0.6
Mean
430.4
Runway 33L Block 2
-
Stdev.
-------- -------
-------------------
0.2
0 -------------------------- m------------------0 W
1
2
I
3
iI I i
5
6
4
7
-
II
10 11
9
8
12
------
----------.....
13
14
I I
15
I
16
Radar Hits from Runway Threshold
Figure 3.4.4.2.10
Final Approach Velocity Profile
Runway 33L Block 5
--------------------
----------------------------
1.2
"
0.8
&--
0-
Mean
4
Stdev.
S0.6
-------------------
* 0.4 ----------0
0.2
1
2
3
4
5
6
7
8
9
10
11
12
Radar Hits from Runway Threshold
Figure 3.4.4.2.11
88
13
14
15
16
17
Ranway 15R Block 4
Final Approach Velocity Profile
1.4 ------------------------------------------------------------
S1.2
----
-----------------------------------------..
.
m
.,.
0- --- ------------------------------------------------------S0.8
0.6
'
>
Mean
0.4
-
--
---------------
-----------------
-
-
. ...........
2
00.4
1
2
3
4
5
6
8
7
9
10
11
12
13
Radar Hits from Runway Threshold
Figure 3.4.4.2.12
Rnway 15R Block 7
Final Approach Velocity Profile
1.4 ------------------------------------------------------------
S----
----------------------------
-------------------------------
0.8
S0.6 ---
-----
-
-
--------
----------------------------
Mean
S
tdev. -----
0.4 --- ------------------------------ ----------- --------------
---------------------------
' 0.2
0
1
2
3
4
5
7
6
8
9
Radar Hits from Runway Threshold
Figure 3.4.4.2.13
89
10
11
12
13
3.5 Intersection Analysis
3.5.1 Intersection Crossing Times
When an aircraft approaches an intersection it is either cleared to cross it by the
ground controller, or it is instructed to stop or slow down to allow another aircraft to
cross in front of it. Occasionally, in periods of heavy traffic a pilot will have to wait in a
queue to cross a particular intersection or possibly wait for a queue of aircraft to pass
through an intersection.
Figure 3.5.1.1: Typical Intersection
Due to the complexity of the runway and taxiway system at Logan airport, a
departing or arriving aircraft has to cross a significant number of intersections in its way
to the gate or departing runway. In order to measure the intersection crossing time as
Rots)
Zoom In)
Zoning to full airt
ZomOut
FulView)
Redrw)
Dbpl
r)
Edit)
viw
D klopd by tho Flight Tmnsportaton Laboratory, MIT. Cambride Mnssachusetts.
Qut)
accurately as possible a distance of fifty meters before and after the intersecting runway
or taxiway centerline was chosen. Since the radar hits are skin returns from the center of
the aircraft' s fuselage, such a distance would ensure that the calculated crossing time
would include the initial crossing of the front part of the aircraft and will end after its tail
has cleared the crossing runway or taxiway (Figure 3.5.1.1).
The calculated crossing time was split in two segments. The first (time 1) being
the time from the start until the aircraft crosses the intersecting centerline and the second
segment (time 2) from the centerline until the aircraft clears the runway or taxiway. The
following table lists the series of airport links corresponding to each intersection number.
Intersection
Series of Links
1
AZ6 A17 A18
2
A02 A01 A19
3
A22 A23 A48
4
A18 A24 A46
5
A08 A09 A44
6
A74 A75 A83
7
A52 A51 A86
8
A48 A47 A70
9
A24 A46 A60
10
A44 A43 A00
11
A44 A43 A10
12
A35 A43 AOO
13
A35 A43 A10
14
A35 A42 A20
Table 3.5.1
Figures 3.5.1.3 through 3.5.1.12 show the average crossing times along with the
standard deviations for every intersection in the ten blocks of collected data. As figures
3.5.1.13 and 14 show there is a significant difference in average crossing time depending
in the runway configuration and thus in the direction of use of some intersections. For
example, when intersections 10, 11, 12, 13 and 14 are used in the inbound direction
(towards the terminal area), usually after arrivals in runway 27, the crossing times seem
to be much smaller compared to those of departing aircraft which use the same
intersections but in the opposite direction (outbound), and often have to form a queue
while waiting to depart from runway 9 and thus cross these intersections very slowly.
Similarly in intersection 6 the inbound direction of crossing is much quicker (aircraft
landing on runway 4R) that the outbound one (aircraft waiting to depart from 22L and
22R).
Comparing the crossing times of the two different segments of each intersection
(time 1 and time 2), we observe that usually when an intersection is used in the inbound
direction time 2 is larger than time 1 and when it used in the outbound time 1 is larger.
This could be possibly due to the fact that when a pilot is on his way to the gate, after
crossing the intersection it has to slow down since the connecting taxiway segments
A75-A83, A51-A86, A47-A70 etc. are short as they intersect with the circuferential outer
taxi lane which is often congested. On the other hand, in the outbound direction usually
time 2 is smaller than time 1 probably because the connecting taxiways that lead away
from the terminal area are longer and therefore the pilot accelerates faster.
Figures --- through --- show the average crossing times per aircraft class (size) for
intersections where ten or more aircraft crossed them. Larger aircraft are heavier and
logically should have longer crossing times but the graphs show this is the case only in
few blocks of data (block 1).
Average Inersection Crossing Times
Block I
50
45
40 ,
E Mean Time 1
S35 -"
O Mean Time 2
30 Stdev Time 1
P 25 20
Stdev Time 2
15
10
Intersection Number
Figure 3.5.1.3
Average Intersection Crossing Times
Block 2
50
45 -
l Mean Time 1
40 -
O Mean Time 2
35 30 -
SStdev Time 1
25 -
MStdev Time 2
20 15 10.
5.
0-
LEEi
lrc--
I----
r- --
r- -~
u
m
I ----
IEILEdUIn
I----- ~--
--- - ----
--- --
I-----
Intersection Number
Figure 3.5.1.4
-- --
-
I
I___I ~__I_ _________1____
rr___1
_I/_~_n__
1_
Average Intrsection Crossing Times
Block 3
50
45
-
40 -
'35
-
0
Mean Time 1
O
Mean Time 2
30 Stdev Time 1
.
25
-
.
U
20
Stdev Time 2
15
10
Cl
'cn
'r
\O
r-
00
ON
0
C4
Intersection Number
Figure 3.5.1.5
Average Intersection Crossing Times Block 4
50
El Mean Time 1
45
40
Mean Time 2
I-1O
35
.3o -
Stdev Time 1
l
20
Stdev Time 2
15
•
-10
0o
Intersection Number
Figure 3.5.1.6
Average I[tersection Crossing Times
Block 5
50
45
40
S35
* 30
..
25
25
= 20
15
10
5
o
S
e
Intersection
00 Number
Intersection Number
Figure 3.5.1.7
Average Intersection Crossing Times
50
45
40
35 30 25
20
15
O Mean Time
O
Block 6
1
Mean Time 2
* Stdev Time 1
Stdev Time 2
I-I
I
10
5
:;r u
I I I
r::
0
Intersection
00 Numb er
Intersection Number
Figure 3.5.1.8
0
-
W-4
C4
-
en
1-
Averan
e atersection Crossing Times
Block 7
El Mean Time 1
0
Mean Time 2
__
Stdev Time 1
Stdev Time 2
I
"I I
SIntersection
Number
0
-
Ce
Intersection Number
Figure 3.5.1.9
Average Ontersection Crossing Times
Block 8
50
45
40
El Mean Time 1
35
30
E Mean Time 2
25
20
El Stdev Time 1
15
10
5
0
M Stdev Time 2
--
If-
C
oc)
i
N
L
(
I -0
CD
Intersection Number
Figure 3.5.1.10
o
C
CO
_____ ~i__~~__
lls--*-C..--~---I-3----.I~ ~-.._-.1.~_X-l
-~D1---lr-m--~h -C-iII
Average ntersection Crossing Times
I----i---l~^-~ ~-X
I__ _-1411Ull
Block 9
68
0.Mean Time 1
0 Mean Time 2
SStdev Time 1
Stdev Time 2
1
SIntersectinO
C~
00 Numb er
Intersection Number
Figure 3.5.1.11
~erage Intersection Crossing Times
Block 10
50
45 40 Mean Time 1
40
S35
30
-l
O MeanTime2
Stdev Time 1
- 25 S20
U
Stdev Time 2
15 U
10
5
SIntersectO
ion
00
O
Intersection Number
Figure 3.5.1.12
0Number
rCI
Inbond Direction
Average Intersection Crossing Time
50
El Block3 Tim.
45
1
40
35
0
30
* Block8 Tim. 1
25
20.
SBlock8 Tim. 2
15
1 Blockl0 Tim. 1
10
50
0
Block3 Tim. 2
SBlocklO Tim.
imt2
5-ElBlckO
I
IL
Intersection
Figure 3.5.1.13
Obom Direction
Average latersection Crossing Time
50
El Blockl Tim. 1
45
IO Blockl Tim. 2
40
S35
~
Block4 Tim. I
i
Block4 Tim. 2
30
25
. 20
8 15
o 10
M
5
E Block6
Tim. 1
0
Block6 Tim. 2
l
Block9 Tim. 1
0
10
11
12
13
Intersection
Figure 3.5.1.14
14
I Block9 Tim. 2
I~_1_C__:
r~_
--~-CII----~MI- ~-_):._ _i_i-. --)~
Average Intersection Crossing Time
...
Inbound Direction
50 -
SBlock1 Tim. 1
45 -
O Blockl Tim. 2
40 35 -
I
30 25 -
Block6 Tim. 1
SBlock6 Tim. 2
20 15 -
* Block9 Tim. 1
10-
E
50. *
iiiiiiiiiitiiTAi
-------
*-
-
I -VIOM
- - r
Block9 Tim. 2
I
6
Intersection
Figure 3.5.1.15
Average [ntersection Crosing Time
Outboid Directic
E Block2 Tim. 1
0
Block2 Tim. 2
1 Block3 Tim. 1
Block3 Tim. 2
* Block5 Tim. 1
E Block5 Tim. 2
* Block8 Tim. I
[1 Block8 Tim. 2
Eff::
.::
A
BlocklO Tim. 1
6
Intersection
Figure 3.5.1.16
'100
* BlocklO Tim. 2
_i.
i-)-1C13--r-
r -nr(-------
Average Intersection Crossing Times
per Aircraft ClasI
UYI~
O
Avg. T1 Heavy
o Avg. T2 Heavy
Block 1
1 Stdev T1 Heavy
1 Stdev T2 Heavy
O
Avg. T1 Large
0
Avg. T2 Large
B Stdev T1 Large
Stdev T2 Large
IE-I
6
8
Intersection
per Aircraft Class
Block 2
15
12
S9
..
5
7
Intersection
Figure 3.5.1.18
101
* Avg. T2 Small
I
Figure 3.5.1.17
Average Intersection Crossing Times
* Avg. T1 Small
Stdev TI Small
;.. _ -;
___ _. ____._
Average Intersection Crossing Times
per Aircraft
Block 4
15
12
9
I
6ijiii
i:
iiiiii
3
I"~'
L,*,I.
.
.
~m
m
4Mn
I
Intersection
Figure 3.5.1.19
Average Intersection Crossing Times
Block 5
per Aircraft Class
1 Avg. T1 Heavy
O Avg. T2 Heavy
* Stdev T1 Heavy
* Stdev T2 Heavy
O Avg. Tl Large
E Avg. T2 Large
E Stdev T1 Large
D
Stdev T2 Large
* Avg. T1 Small
0
Intersection
Figure 3.5.1.20
102
Avg. T2 Small
* Stdev T1 Small
-~LP~
-
-- .--. nU--
~-~
.LiiiP u.
rrAMIAIMM"
~
Y. ~
~~--r --. -41~. -M.~~~~,~;~~-
.4 mv.li-i-
U
~-:-
3.5.2 Intersection Use
The following table provides information about the number of aircraft that used
each intersection. Figures 3.5.2.1 through 3.5.2.4 show the intersection use under the four
most popular runway configurations at Logan airport. As we can see, the use of a specific
exit is closely related to the operating runway configuration.
Number of Aircraft Using Each Intersection
Intersection Block Block Block Block Block Block Block Block Block Block
1
2
3
4
5
6
7
8
9
10
1
0
2
0
0
6
0
0
0
1
0
2
0
15
0
0
23
0
0
0
0
0
3
0
5
0
4
21
0
7
0
0
0
4
0
2
0
12
1
0
9
0
0
0
5
0
13
3
0
1
0
0
1
0
4
6
13
4
25
2
7
26
0
36
6
45
7
0
18
0
7
16
1
6
1
1
1
8
11
6
3
5
20
3
11
3
11
4
9
8
30
0
11
50
0
11
2
6
3
10
0
9
2
0
1
0
0
3
0
1
11
0
3
2
0
0
0
1
4
0
5
12
11
0
6
25
0
9
7
8
9
3
13
4
0
1
4
1
3
2
7
1
6
14
11
1
1
11
0
10
3
5
8
0
Table 3.5.2
103
-I
.l
_:I~_ ___I__.;__;
s__ll ri~i __i
~______~_~
_ ~rrlC~
__
~lil_~n_
Number of Aircrafts Using Each Intersection
Configuration:
Arrivals 4R, 4L and Departures 9, 4L
50
45
-
40
SBlockl
35
30
O
-
25
20
Block6
--
Il Block9
15
10
5
0
-a
AMmIN)I
n _.)~i, ,
t
m~~
I
-
C4
Intersection
00 Number
M
cr,
C
-
~
Intersection Number
Figure 3.5.2.1
Configuration:
Number of Aircrafts Using Bach Intersection
Arrivals 33L and Departures 33L, 22R
50
45
a
- 40 -
M Block2
S35
< 30
30
O Block5
25
20
-
15 -S10
z
I
I
5 - II
-C
I
I
O Intersect ion
00 Nu
Intersection Number
Figure 3.5.2.2
104
I
I
I
I
Cmber
. I.
Number of Aircrafts Using Each Intersection
Configuration:
Arrivals 27, 22L and Departures 22R, 22L
Block3
O
Block8
SBlock0
Cl4
I
I_.
e
MOM,..
!
Intersection Number
Figure 3.5.2.4
105
3.6 Taxiway Analysis
3.6.1 Taxiway Average Velocity
The average velocity of an aircraft moving on an airport taxiway system depends
on many factors. Before the analysis of the collected data, we expected that the distance
of the taxiing segment would be a significant determinant of this velocity. Usually when
pilots are moving on a short segment, prefer to taxi slowly since they expect soon to
arrive at an intersection and might be instructed by the controller to stop for crossing
traffic. On the other hand when a pilot sees that he has a long stretch in front of him with
no imminent intersection, he taxies at higher speeds.
The location of the taxiway segment should also play an important role. Taxiways
far away from the terminal area are more likely to exhibit higher average velocities since
they tend to be less congested and their surrounding areas are usually free of obstacles.
Other variables that affect the taxiway velocities, are the complexity of the taxiway
system at the particular airport and the level of familiarity that each pilot, who operates
there, has with the system. The first variable usually remains constant while the second
one can vary, and cannot be very easily quantified.
Trying to test if the taxiway length and location relates to the average taxiway
velocity, we categorized each taxiway link that did not belong to a runway or was not an
exit link, into two groups. Those links that were shorter than 500 meters were assigned a
letter S and those that were longer a letter L. Each taxiway was also assigned either a
letter C (close) if it was located inside the outer taxilane, or a letter F (far) otherwise.
106
Figure 3.6.1.1 shows the average velocity of all taxiways for each data block
independent of segment length or location. The velocities range between 7 and 16 knots.
However, standard deviations vary from 6 to 22 knots. Data block 3 exhibits the lowest
average velocity for its taxiways (7 knots), possibly due to the heavy traffic of that time
period. To traffic congestion can be also attributed the low velocities that we observe in
data block 4. Bad weather conditions (thunderstorms and fog) are the probable reasons
for low velocities in blocks 1 and 9.
Average Velocity in Taxiways
35
" 30
.Mean
25
ml Stdev.
5 20
>
15
U 10
S5
0
1
2
3
4
5
6
7
8
9
10
Data Block
Figure 3.6.1.1
The results of the previous taxiway categorization can be seen in figures 3.6.1.2
and 3.6.1.3. Longer segments almost always display higher average velocities than the
shorter ones. We must note though that in periods of heavy traffic the differences in speed
between longer and shorter taxiways become smaller. As figure 3.6.1.3 shows, taxiways
that are located far from the terminal area, in data blocks 1, 6, 9, 4 and 7, have higher
average velocities while in the other blocks lower. It is interesting to note that blocks that
have higher velocities, the runway configuration is the same (block 1, 6, 9 arrivals in 4R
107
departures from 9). The same pattern (same configuration for all blocks that exhibit
similar velocity characteristics) occurs in the other blocks of data, where the combination
(higher velocity - taxiway location) is opposite.
Average Velocity in Taxiways
35
S30
25
20
Length : Short I Long
I
*E Mean
Short
Mean Short
0E SSd Shor
Short
20
>. 15
SMean Long
5
Sd Long
S
Cl
I W
O
00
Oi
I
Data Block
Figure 3.6.1.2
Figure 3.6.1.4 shows two groups of taxiways. In the first group, belong taxiways
that are located close to the terminal area and are short in length, while in the second one
the taxiways that are located further away and are longer. Here, the differences in
velocities between the far and long and the close and short becomes even larger in favor
of the longer and further away ones compared to the previous two figures. However,
these differences, in the blocks where close and short taxiways have higher velocities,
become smaller.
Overall, we can conclude that there exist a close relationship between the length
of the taxiway and the taxiing velocity. The location of the taxiway segment seems to be
also critical. We must also note that traffic congestion has a more deleterious effect on the
average taxiing velocity than bad weather conditions.
108
Figure 3.6.1.3
Average Velocity in Taxiways Location : Close I Far Length :
Short / Long
1)
0
1
5-
5
Mean C & S
OSd C&S
B
o-
O
Mean F & L
SSdF&L
501
Data Block
Figure 3.6.1.4
109
3.6.2 Taxiway Use
Figure 3.6.2.1 through 3.6.2.10 present the use of the taxiway segments that
constitute the inner and outer taxilanes. All data blocks show an almost identibal taxiway
use no matter what is the runway configuration. The most often used taxiway links are
B03-B04 and B04-B05 of the inner taxilane. Only block 3 demonstrates a higher
percentage wise use of the outer taxilane, probably due to the heavy surface traffic of that
day.
In figure 3.6.2.11 through 3.6.2.20 the use of the supporting taxiways is presented.
As supporting taxiways are classified all the taxiway links that do not belong to either a
runway or the inner and outer taxilanes, but feed traffic to the terminal and runway areas.
A very strong relationship between the use of the these taxiway links and the particular
runway configuration seem to exist, as figures of data blocks with the same configuration
seem identical.
110
r
c~
~
e:
?
CD
W
0\
B08A92
B09A93
A50A20
A35A42
A97A98
A98A80
A80A99
A99B00
BOOB01
BO1B02
B02B03
B03B04
B04B05
B05B06
B06B07
B07B08
B08B09
B09A95
A95A96
A97A79
A79A81
A81BI0
B10A82
A82A83
A83A84
A84A85
A85A86
A86A70
A70A60
A60A00
AOOAI0
A10A20
A50A91
A91A92
A92A93
A93A94
A94A95
00
~
0
00
0
Number of Aircraft
B08A92
B09A93
A50A20
A35A42
A97A98
A98A80
A80A99
A99B00
BOOBO0
B01B02
B02B03
B03B04
B04B05
B05B06
B06B07
B07B08
B08B09
B09A95
A95A96
A97A79
A79A81
A81BIO
B 10A82
A82A83
A83A84
A84A85
A85A86
A86A70
A70A60
A60A00
AOOAIO
A10A20
A50A91
A91 A92
A92A93
A93A94
A94A95
t-O
Number of Aircraft
-
'.
4
iD
%a
e
B08A92
B09A93
A50A20
A35A42
A97A98
A98A80
A80A99
A99B00
BOOB01
B01B02
B02B03
B03B04
B04B05
B05B06
B06B07
B07B08
B08B09
B09A95
A95A96
A97A79
A79A81
A81B10
B10A82
A82A83
A83A84
A84A85
A85A86
A86A70
A70A60
A60A00
AOOAI0
AIOA20
A50A91
A91A92
A92A93
A93A94
A94A95
Number of Aircraft
B08A92
B09A93
A50A20
A35A42
A97A98
A98A80
A80A99
A99B00
BOOB01
B01B02
B02B03
B03B04
B04B05
B05B06
B06B07
B07B08
B08B09
B09A95
A95A96
A97A79
A79A81
A81B10
B10A82
A82A83
A83A84
A84A85
A85A86
A86A70
A70A60
A60A00
AOOAI0
A10A20
A50A91
A91 A92
A92A93
A93A94
A94A95
(.A
0
PA
PCD
N)
Number of Aircraft
(,A
t)
w
CD
A82A83
A83A84
A84A85
A85A86
A86A70
A70A60
A60A00
A00AI0
A10A20
A50A91
A91A92
A92A93
A93A94
A94A95
B10A82
.
B08A92
B09A93
A50A20 ...
A35A42 ....
A97A98 L
A98A80
A80A99
........
'.::.":
A99B00
BOOB01 ..........
B01B02
----..--B02B03
B03B04 .............
.
B04B05 ..
B05B06 ........
....
B06B07 ...
B07B08
B08BO9 "
SBOA95
A95A96
A97A79
A79A81
p- A81B0
Number of Aircraft
,.
-I!
A97A79
A83A84
A84A85
A85A86
A86A70
A70A60
A60A00
A00AI0
A10A20
A50A91
A91A92
A92A93
A93A94
A94A95
BI0A82
A82A83
A79A81
- A81BIO
C
- ---
.
B08A92
B09A93
A50A20
A35A42
A97A98
A98A80
A80A99
.........
A99B00
BOOB01
B01B02
B02
B02B03 .........
B03B04
B04B05
B05B06 ""
B06B07 .
B07B08
- B08B09
B09A95
A95A96
-
-
Number of Aircraft
...
C
.r.
e
B08A92
B09A93
A50A20
A35A42
A97A98
A98A80
A80A99
A99B00
BOOB01
BO1B02
B02B03
B03B04
B04B05
B05B06
B06B07
B07B08
B08B09
B09A95
A95A96
A97A79
A79A81
A81B10
B10A82
A82A83
A83A84
A84A85
A85A86
A86A70
A70A60
A60A00
AOOA10
A1OA20
A50A91
A91A92
A92A93
A93A94
A94A95
Number of Aircraft
-s
OQ
-
BB09
B08BO9
B09A95
A95A96
A97A79
A79A81
A81BI0
B10A82 ~
(.
A82A83
A83A84 .................
A84A85
A85A86
A86A70 ................
A70A60 ................
A60A00 ...........
AOOAI0
A10A20
A50A91
A91A92
A92A93
A93A94
A94A95
B08A92
B09A93
A50A20
A35A42
A97A98
A98A80
.
A80A99
A99B00
............
BOOB01
BOIBO2
B02BO3
B034B04
................
Number of Aircraft
CD
ito
ED
0t
me
111
B08A92
B09A93
A50A20
A35A42
A97A98
A98A80
A80A99
A99B00
BOOB01
BO1B02
B02B03
B03B04
B04B05
B05B06
B06B07
B07B08
B08B09
B09A95
A95A96
A97A79
A79A81
A81BI0
B 10A82
A82A83
A83A84
A84A85
A85A86
A86A70
A70A60
A60A00
AOOAI0
A10A20
A50A91
A91A92
A92A93
A93A94
A94A95
Number of Aircraft
a\
r
t3
h)
i~h
~f
W
rS!
A80BIO
A20A40
A39A37
A37A31
B04A60
A18A24
A16A17
A33A32
A32A31
A18A9
AO2AOI
A02A04
A04A07
A07A90
.
B06AI0
A35A43
A44A43
B03A70
A48A47
Pr B14B3
A98A81
BOOA83
A74A73
A73A71
A71 A69
A67A66
A66A63
B02A86
A99A82
B05A00
A35B13
A34A33
B06A50
A34A33
B06A50
B05AOO
A35B3
B02A86
A99A82
A67A66
A66A63
A73A71
A71A69
BOOA83
A74A73
B14B13
A98A81
A35A43
A44A43
A04AO7
A07A90
B06A10
A02A01
A02A04
A32A31
Al8A19
A16A17
A33A32
B04A60
A18A24
A39A37
A37A31
A80B10
A20A40
ggg
C.
B03A70
A48A47
=
0
gggg
|iggg
ttmg
Iss
ggg
------
Number of Aircraft
0
LA
0
I--
I.-
t.A
0
A
Number of Aircraft
w
0D
LA
-
m
cQQ
A8OBIO
A20A40
A39A37
A37A31
B04A60
A18A24
A16A17
A33A32
A32A31
A18A19
A02A01
A02A04
A04A07
A07A90
B06AI0
A35A43
A44A43
B03A70
A48A47
B14B13
A98A81
BOOA83
A74A73
A73A71
A71A69
A67A66
A66A63
B02A86
A99A82
B05A00
A35B3
A34A33
B06A50
0
ur
0
ur
-
Number
wr
0
Aircraft
w
0
A8OBI0
A20A40
A39A37
A37A31
B04A60
A18A24
A16AI7
A33A32
A32A31
A18A9
AO2AO1
A02A04
A04A07
A07A90
B06A10
A35A43
A44A43
B03A70
A48A47
B14B13
A98A81
B00A83
A74A73
A73A71
A71A69
A67A66
A66A63
B02A86
A99A82
B05A00
A35B3
A34A33
B06A50
1 --
Crn
U
EMB
J6
0
urn
0
3
Number of Aircraft
Is)
CA
,
C
*
S
w
*
B04A60
Al8A24
A16A17
A33A32
A32A31
A18A9
AO2AO1
AO2A04
A04A07
A07A90
B06AI0
A35A43
A44A43
B03A70
A48A47
B14B13
A98A81
BOOA83
A74A73
A73A71
A71A69
A67A66
A66A63
B02A86
A99A82
B05A00
A35B13
A34A33
B06A50
A80BI0
A20A40
A39A37
A37A31
B04A60
I
A18A24
A16AI7
A33A32
A32A31
A18AI9
AOAOOI1
A02A04
A04A07
A07A90
SB06A10
. A35A43
A44A43
"
B03A70
Ch A48A47
P
Bl4B13
A98A81
BOOA83
A74A73
A73A71
A71A69
A67A66
A66A63
B02A86
A99A82
B05A00
A35B13
A34A33
B06A50
A8OBIO
A20A40
A39A37
A37A31
Number of Aircraft
- ---
Number of Aircraft
.
P
*
S
A67A66
A66A63
B02A86
A99A82
B05A00
A35B3
A34A33
B06A50
A71A69
A35A43
A44A43
B03A70
A48A47
B14B13
A98A81
B00A83
A74A73
A73A71
BO6AIo0
A07A90
A04A07
A80B10
A20A40
A39A37
A37A31
B04A60
A18A24
A16A17
A33A32
A32A31
A18A19
AO2AO1
A02A04
Number of Aircraft
A8OBIO
A20A40
A39A37
A37A31
B04A60
A18A24
A16A17
A33A32
A32A31
A18A9
A02A01
A02A04
A04A07
A07A90
B06A10
A35A43
A44A43
B03A70
A48A47
B14B13
A98A81
BOOA83
A74A73
A73A71
A71A69
A67A66
A66A63
B02A86
A99A82
B05A00
A35B13
A34A33
B06A50
I
0
P-
T
L.A
P-
I
0C>
t
I
___
J
Number of Aircraft
9slIlllBlllllllsllIlI~
I[]
i
U
t
SIJ
0>
I
-
0
wJ
Chapter 4
Conclusions
4.1
Introduction
In this chapter the major observations of the previous analyses are discussed and
final conclusions are drawn. These conclusions are divided into three sections; runways,
intersections and taxiways. Finally, the last part provides some directions for future
research.
121
4.2
Runways
The runway analysis showed some interesting results. The occupancy time on
landing does vary, as expected, with aircraft size but this variation is mainly due to the
fact that heavier aircraft tend to exit further down the runway, resulting in higher
occupancy times. However, aircraft using a given exit have similar occupancy time,
independently of aircraft size. The particular angle of the exit used also plays a very
significant role. Aircraft that use obtuse angled exits, usually exit at higher speeds, and
therefore maintain a higher average landing velocity resulting in similar occupancy times
compared to exits that are located much closer to the runway threshold. The standard
deviation of the occupancy time is usually smaller for aircraft using the first exits, than
those exiting further down the runway. Visibility also affected the occupancy time,
mainly due to the increased use of exits located further down the runway. In similar
weather conditions, aircraft using exit 6 (high speed) of runway 27, in day light had
average occupancy times of 35 seconds (lowest overall), while at night, in similar
weather conditions, most aircraft used exit 8 (low speed) and their occupancy times were
significantly increased (53 seconds).
In the analysis of exit usage, as mentioned above, heavier aircraft tend to use exits
located further away from the threshold, while smaller sized ones require shorter landing
distances and exit earlier. The specific turning angle of the exit was also a determinant
factor of its use. As a result, independently of runway, most aircraft tended to prefer the
use of obtuse angled (high speed) exits.
Exit velocities, as expected, were closely related to the angle of the exit, and
whenever aircraft where using obtuse angled exits they exited at significantly higher
speeds. However, this was only true, for high speed exits that were accompanied with
122
long exit segments and allowed the pilot enough distance to brake. The exit velocity did
not vary uniformly between aircraft classes and it seemed that exit velocity does not
depend on aircraft size.
The landing velocity profiles showed a close relationship between the amount of
deceleration and the type of exit (high or low speed) used after landing. In the final
approach, we observed a slight deceleration as the aircraft were approaching the
threshold. Data from runway 27, illustrated the importance of visibility, as in day light,
this deceleration was quite significant whereas during night operations we saw a much
smoother final approach profile. In runway 15R, under different weather conditions,
aircraft seem to even accelerate before the threshold.
4.3
Intersections
There is a significant difference in average crossing time dependingjn the runway
configuration and thus in the direction of use of some intersections. For example, when
intersections 10, 11, 12, 13 and 14 are used in the inbound direction (towards the terminal
area), usually after arrivalsjn runway 27, the crossing times seem to be much smaller
compared to those of departing aircraft which use the same intersections but in the
opposite direction (outbound), and often have to form a queue while waiting to depart
from runway 9 and thus cross these intersections very slowly.
Comparing the crossing times of the two different segments of each intersection
(time 1 and time 2), we observe that usually when an intersection is used in the inbound
direction time 2 is larger than time 1 and when it used in the outbound time 1 is larger.
This could be due to the fact that often pilots slow down and approach the terminal area
123
cautiously, while on their way to the departing runway, they accelerate faster out of an
intersection since the taxiways that lead away from the terminal are usually longer.
Aircraft size did not seem to be a determinant of crossing times as larger size aircraft had
longer crossing times only in few blocks. Exit usage was closely tied to the operating
runway configuration.
4.4
Taxiways
In the taxiway analysis we saw that there exists a close relationship between the
length of the taxiway and its velocity. Taxiways far away from the terminal almost
always exhibit higher average velocities since they tend to be less congested and free of
surrounding obstacles. We must note though, that in periods of heavy traffic the
differences in speed between longer and shorter taxiways become smaller. The use of the
inner and outer taxilanes under different configurations is almost identical. However, a
very strong relationship between the use of the supporting taxiways around the terminal
area, and the particular runway configuration seems to exist.
4.5
Directions for Future Research
Although after the analysis of the data many questions regarding the motion
characteristics of aircraft moving on the surface of Logan airport have been answered,
many more have been raised. More research should be done on the factors that affect
airport surface traffic in order to successfully develop and implement future surface
124
traffic automation systems. This study was limited to only Logan airport but a future one
should include and compare data from a wide variety of airports.
Such a study could include analysis of other variables that affect surface traffic
such as congestion and examine in further detail their potential effects. Most importantly,
a much larger size of data must be gathered. This requires some form of radar
surveillance at the airport. In our case, even though almost 12 hours of airport operation
data were made available for this study, after the breakdown of all aircraft by aircraft
type, runway and exit used we were left with very small sample for each variable that we
wanted to measure, limiting the accuracy of our results. There is much more data
available for Logan if further confidence in the measurements is required.
Hopefully, the content of this thesis will act a catalyst in attracting interest and
consequently, more studies will be under taken in the future, for a more complete
understanding of the surface traffic variables and a more efficient use of the airport
surface.
125
References
1.
Bussolari, Steven (1991), Real-Time Control Tower Simulationfor Evaluation of
Airport Surface Traffic Automation, International Symposium on Aviation
Psychology Proceedings Vol. 1 (A92-4401)
2.
Harrison, Michael (1993), Airport Surface Traffic PlannerDesign Concept,
Systems Engineering and Development FAA, 19 April 1993.
3.
Lyon, Ervin F. (1991), Airport Surface Traffic Automation, The Lincoln
Laboratory Journal, Volume 4, Number 2, pp. 151-187.
4.
Marquis, Douglas (1991), A PreliminaryDescriptionof the Airport Surface
Traffic Automation (ASTA) System, ATC Project Memorandum No 41PMASTA-0006, Lincoln Laboratory, 9 January 1991.
5.
Mathaisel, Dennis F. (1994), A Simulation of Aircraft Motion on the Airport
Surface, MIT Flight Transportation Laboratory, 16 March 1994.
6.
Swedish, William (1972), Some Measures of Aircraft Performanceon the
Airport Surface, MIT FTL Report R72-4, May 1972.
7.
Weiss, W. E., Analysis of Runway Occupancy Time and SeparationData
Collectedat Laguardia,Boston, and Newark Airports, MTR-84W228
MITRE.
8.
Flight Transportation Associates (1982), Airfield CapacityAnalysis for BostonLogan InternationalAirport, for Massport, November 1982.
9.
Airborne Instruments Laboratory (1960), Runway Characteristicsof Jet
Transportsin Routine Operation,Report No 5791-15, Federal Aviation
Agency, March 1960.
10.
Airborne Instruments Laboratory (1960), Use of Exit Taxiways at New York
International Airport, Supplement to Report No 7601-1, July 1960.
11.
Boston Logan InternationalAirport Capacity Enhancement Plan, Airports
Division FAA, October 1992
12.
A Report on Noise Abatement, Massachusetts Port Authority Noise Abatement
Office, January 1994.
126