MIT
ICAT
16.72 Air Traffic Control
Overview
Prof. R. John Hansman
MIT Department of Aeronautics
and Astronautics
MIT
ICAT
ICAT
US Capacity Issue
•
Prior to 9/11 the US Air Transportation system is approaching a critical
saturation threshold where nominal interruptions (e.g. weather) resulted in a
nonlinear amplification of delay
•
US and Regional Economies highly dependant on Air Transportation
 Business travel (stimulated by info technology)
 Air Freight
 Personal travel
•
System is highly complex and interdependent
•
Need better understanding of system dynamics and real constraints to guide
and justify efforts to upgrade NAS
•
Current efforts will not provide capacity to meet demand which will re-emerge
when the economy cycles up
•
Impact of upcoming capacity crisis is not well understood
 Operational Impact
 Economic Impact
•
Similar Issues in Europe
•
Different Issues in Emerging Regions (eg China, India)
MIT
ICAT
MIT
ICAT
ICAT
Passenger Traffic by Region
Scheduled Revenue Passenger-Kilometers by Region
1400
1200
North America
RPK (billion)
1000
Europe
800
Asia and
Pacific
Latin America
& Caribbean
600
Middle East
400
Africa
200
0
1970
1975
1980
1985
1990
Source: ICAO, scheduled services of commercial air carriers
1995
2000
2005
MIT
ICAT
ICAT
Trends in Aircraft Size
150
Average seats per departure
140
130
120
110
100
90
80
1990
1991
1992
1993
1994
1995
1996
1997
1998
Source: DOT Form 41 data (including Regional Jets and Turboprops)
1999
2000
2001
2002
2003
MIT
ICAT
ICAT
Data source: FAA Operational Network (OPSNET)
US Flight Delays
from 1995 to 2006
2006
Cancellations
Thousands
MIT
ICAT
ICAT
Flight Cancellations
250
200
150
100
50
0
1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004
Source: BTS, Airline On Time Performance data
MIT
ICAT
ICAT
Source: FAA OPSNET data
US Flight Delays
from 2000 to 2006
2006
MIT
Delays at Chicago O’Hare
ICAT
ICAT Pressure for Demand Managment
ORD: Total Delays
16000
14000
Total Delays
12000
2004
10000
2003
8000
2002
6000
2001
4000
2000
2000
0
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Month
Source: FAA OPSNET data
25
20
15
10
Consumer complaints
Southwest
USAirways
United
TWA
Northwest
America West
Delta
Continental
Alaska
American
5
Note: Year 2005 consumer complaints is an extrapolation using data from Jan-Mar 2005
Data source: DOT Aviation Consumer Protection Division, available at: http://airconsumer.ost.dot.gov/
20
04
20
02
2200
000
98
19
96
19
94
19
92
19
90
19
88
19
86
19
84
19
19
19
82
0
80
Complaints to the DOT (000)
MIT
ICAT
ICAT
MIT
ICAT
ICAT
Air Traffic Control Functions
• Aircraft Separation Assurance
• Traffic Congestion Management
• Flight Information
• Search and Rescue
• Example of a Command, Control and Information System
MIT
COMPONENTS OF AIR
ICAT
ICAT TRANSPORTATION INFRASTRUCTURE
• Airports





Runways
Terminals
Ground transport interface
Servicing
Maintenance
• Air traffic management




Communications
Navigation
Surveillance
Control
• Weather
 Observation
 Forecasting
 Dissemination
• Skilled personnel
• Cost recovery mechanism
MIT
ICAT
ICAT
Current Control Structure
 Surface Control
 “Ground”
 Local Control
 “Tower”
 Terminal Area Control (TRACON)
 “Approach and “Departure”
 Enroute Control (ARTCC)
 “Center”
 Oceanic Control (FIR)
 “Oceanic”
 Flow Control (ATCSCC)
 “Central Flow”
US Air Route Traffic Control Center
MIT
ICAT
ICAT
(ATRCC) Airspace - 20 Centers
ZSE
ZMP
ZLC
ZBW
ZAU
ZOB
ZDV
ZOA
ZNY
ZID
ZKC
ZDC
ZME
ZLA
ZAB
ZTL
ZFW
ZJX
ZHU
ZMA
MIT
ICAT
ICAT
High Level Sectors
257
MIT
ICAT
ICAT
Low Level Sectors
378
MIT
ICAT
ICAT
TRACONS
MIT
ICAT
ICAT
Overall Structure
ATM System Current
Functional Structure
MIT
ICAT
ICAT
Planning - Strategic Level
Desired
Sector
Loads
Schedule of
Capacities
Weather
Execution - Tactical Level
Filed
Approved
Planned
Flight Flight
Flow
National
Facility
Sector
Flight
Plans
Plans
Rates
Flow
Flow
Traffic
Planning
Planning
Planning
Planning
Flight
Schedule
Airline
hrs - day
CFMU
hrs
5-20 min
TMU
D-side
Approved
Handoffs
Negotiate
Handoffs
Real State
Plan/Intent
Measurement
Requests
Efficiency
Clearance
Requests
Clearance
Requests
Sector
Traffic
Control
5 min
AOC
Aircraft
Aircraft State
Vectors
Guidance and
Clearances
Navigation
< 5 min
Traffic
Sensor
R-side
Pilot
AC State
Sensor
Throughput
Other Aircraft
States
Safety
Increasing Criticality Level
Source: A. Haraldsdottir Boeing
MIT
ICAT
ICAT
F lig ht
S trip s
ATC Control Loo
Radar Surveillance Limits
Decision
Aids
ATC
Displays
Flight Plan
Amendments
Surveillance:
Enroute: 12.0 s
Terminal: 4.2 s
ADS :
1s
Vectors
Voice
AOC:
Airline
Operations
Center
Pilot
ACARS
(Datalink)
CDU
MCP
Tra je cto r
C omm ands
Initial
Clearances
Controls
S tate
C omm ands
F lig ht
M anagem ent
C om p ute r
Manual Control
Autop ilo t
Auto thrust
Displays
Aircraft
S tate
N av ig atio n
Emerging Approaches: ADS-B and Multi-Lateration
MIT
ICAT
ICAT
CO 123
350C
B757 310
Radar Display Example
MIT
ICAT
ICAT
• Contract process




• ATM is a Human Centered
Process
Negotiation
Execution
Monitoring
Re-negotiation/amendment
Agents
 Controllers
• Strategic
• Tactical
 Pilots
 Airlines
• Dispatchers
• ATC coordinators
 Airports
• Resources
 Airspace
 Runway
 Airport surfaces
MIT
ICAT
ICAT
Flight Progress Strip
MIT
ICAT
ICAT
Human Factors and
Adaptation
• ATM is a human centered contract process for the allocation of
airspace and airport surface resources.
• Current NAS has evolved over 60 years
• The system has significant local adaptations resulting in
nonhomogeneity





Airspace design
Local procedures
Letters of agreement
Noise restrictions
Site specific training (FPL = 3-5 years)
• Major operational changes were event driven, enabled b
technical capability
 Positive radar control - Grand Canyon 1956
 TCAS - Los Cerritos 1982
MIT
ICAT
ICAT
New York Arrival and
Departure Tracks
MIT
ICAT
ICAT
ATC Workload as a
System Constraint
MIT
ICAT
ICAT
Blocking
Block the flow to
maintain number
of aircraft within
the acceptable
level (ex. Set by
AAR or OALT)
Air traffic
controller
Blocking
action
Saturated
resource
Controlled
resource
Finite capacity
buffer
AAR is Airport Acceptance Rate
OALT is sector Operationally
Acceptable Level of Traffic
Number of aircraft
threshold corresponding
to, for example, AAR or
OALT, (varies depending
on current conditions)
For example,
rate set by
AAR or OALT
MIT
ICAT
MIT
Downstream
Flow
Constraints
ICAT
ICAT
Flow Management
Ramp
Gate
Restrictions Restrictions
Gates
Ramp
Taxi
Runway
Restrictions Restrictions
Taxiways
Runways
Capacity
Constraint
ex. Miles In
Trail through
exit fix
Terminal
area and
exit fixes
Capacity
Constraint
ex. Airport
Capacity Acceptance
Constraint Rate (AAR)
ex. Sector
Operationally
Acceptable
Level of Traffic
(OALT)
En route
sector
airspace
Departure
flow
Destination
airports
MIT
ICAT
ICAT
Dallas Fort-Worth
• Aircraft are condensed into distinct flows feeding 4 arrival fixes.

June 20, 2001
12:19 p.m.
153 Aircraft In-bound
MIT
ICAT
ICAT
• Atlanta
Condensation and merges have reduced 116 trajectories at airport to 4
ARTCC
Boundaries
Route
Flown
• June 14, 2001
11:15 a.m.
116 Aircraft In-bound
MIT
ICAT
ICAT
• San Francisco
Special use airspace provides additional constraints
Special Use
Airspace
Route
Flown
Flight Plan
• June 11, 2001 4:13 p.m. 78 Aircraft In-bound
MIT
ICAT
ICAT
Projected % Developmental
Controllers
From: ATCS Workforce Plan Briefing
MIT
ICAT
ICAT
Capacity Limit Factors
• Demand
 Peak Demand
 Hub & Spoke Networks
• Airspace Capacity
 Airspace Design
 Controller Workload
 Balkanization
• Airport Capacity




Runways
Gates
Landside Limits (including Security)
Weather
• Environmental Limits
 Noise (relates to Airport)
 Emissions (local, Ozone, NOX, CO2)
MIT
ICAT
ICAT
• Runways
• Weather
Airport Syste
Capacity Limit Factors
 Capacity Variability
 Convective Weather
• Landside Limits
 Gates
 Terminals & Security
 Road Access
• Downstream Constraints
• Controller Workload
• Environmental
 Community Noise
 Emissions
• Safety
Airport Capacity Envelope
MIT
ICAT
Boston (BOS)
100
VFR
90
ASPM - Jul/Aug 2000 - Visual
Approaches
Calculated VMC Capacity
68,50
80
Arrivals per Hour
ASPM - Apr 2000 - Visual Approaches
70
Optimum Rate (BOS)
60
50
40
30
20
10
0
0
10
20
30
40
50
60
70
80
90
100
Departures per Hour
100
80
Arrivals per Hour
ASPM - Apr 2000 - Instrument Approaches
IFR
90
ASPM - Jul/Aug 2000 - Instrument Approaches
Calculated IMC Capacity
Reduced Rate (BOS)
70
60
44,44
50
40
30
20
10
0
0
10
20
30
40
50
60
70
Departures per Hour
80
90
100
Source: FAA Benchmark Data
MIT
ICAT
Separation Requirements for
Arrival (Same Runway)
• Wake Turbulence Requirement
Radar Separation requirements
Trailing Aircraft
Leading Aircraft
Heavy
B757
Large
Small
Heavy
4
4
3(2.5)
3(2.5)
Large
5
4
3(2.5)
3(2.5)
Small
5
5
4
3(2.5)
Visual Separation requirements
• Pilots Discretion
• Preceding arrival must be clear of runway at touchdown
Runway Occupancy time
Variable Capacity Effects
MIT
ICAT
1995 Delays vs Operations
Data from FAA Capacity Office, CY95
60
SFO
Delayed Flights (per 1000)
50
40
LGA
STL
EWR
30
ORD
LAX
BOS
20
DFW
ATL
JFK
IAH
10
SJU
CLT
MEM
0
0
200000
PIT
HNL
400000
PHX
DEN
LAS
600000
800000
1000000
Total Operations (CY95)
From John Andrews, MIT Lincoln Lab
MIT
ICAT
MIT
ICAT
Safety vs Capacity
•
The current airborne system is extremely safe but conservative
•
Increased capacity with current infrastructure implies Reduced Operational
Separation





•
Airborne Separation Standards
Runway Occupancy Times
Wake Vortex
Controller Personal Buffers
...
How do you dependably predict the safety impact of changes in a complex
interdependent system?
 Statistics of small numbers
 Differential analysis limited to small or isolated changes
 Models??
•
Safety Veto Effect
•
Runway Incursions are an area of concern
MIT
ICAT
EN ROUTE MINIMA HAVE NOT
CHANGED DESPITE 5 x IMRPOVEMENT
IN RADAR PERFORMANCE
Azimuth resolution at maximum
range as % of en route minima
120
100
80
5 nm en route separation minima
1950
2000
1950
60
40
1960
2000
20
2000
1960
2000
0
ARSR-1
ARSR-4
Long range
primary radars
ASR-6
ASR-9
Medium range
primary radars
Mode A
Mode S
Medium range
secondary radars
MIT
ICAT
PERSONAL
SAFETY BUFFER
SEPARATION ASSURANCE
CONSIDERATIONS
SURVEILLANCE
UNCERTAINTY
MINIMUM
SEPARATION
STANDARD
PROCEDURAL
SAFETY
BUFFER
HAZARD
ZONE
MIT
ICAT
IMPROVED SURVEILLANCE HAS NOT
LED TO REDUCED EN ROUTE MINIMA
WHEN STANDARDS
WERE DEVELOPED
(e.g. 1950s for en route radar)
IMPROVED SURVEILLANCE
ENVIRONMENT
(e.g. today for en route radar)
Minimum
Separation
Standard
•
Surveillance has improved, but separation minima have not changed:
procedural safety buffer has implicitly increased
MIT
ICAT
Schedule Factors
•
Peak Demand/Capacity issue driven (in part) by airline Hub and Spoke
scheduling behavior
 Peak demand often exceeds airport IFR capacity (VFR/IFR Limits)
 Depend on bank spreading and lulls to recover
 Hub and Spoke amplifies delay
•
Hub and spoke is an efficient network
 Supports weak demand markets
•
Schedules driven by competitive/market factors
 Operations respond to marketing
 Trend to more frequent services, smaller aircraft
 Ratchet behavior
 Impact of regional jets
•
Ultimately, airlines will schedule rationally
 To delay tolerance of the market (delay homeostasis)
•
Limited federal or local mechanisms to regulate schedule
MIT
ICAT
Hub and Spoke Network
Completely Connected Network = 2(N-1) Flights
(eg., 50 Airports, 98 Flights)
MIT
ICAT
Fully Connected Network
Completely Connected Network = N(N-1)
(eg., 50 Airports, 2450 Flights)
MIT
ICAT
Capacity Exampl
(50 Flights/hr)
70
60
50
40
30
20
10
0
Time
250
200
150
100
50
0
Time
MIT
ICAT
Capacity Exampl
(40 Flights/hr)
70
60
50
40
30
20
10
0
Time
250
200
150
100
50
0
Time
MIT
ICAT
Capacity Exampl
(30 Flights/hr)
70
60
50
40
30
20
10
0
Time
250
200
150
100
50
0
Time
MIT
ICAT
Classic Delay vs Demand
Curve
DELAY
Non-Linear
Region
Linear
Region
Capacity
Limit
DEMAND
MIT
ICAT
LGA Air 21 Impact
LaGuardia Airport
200
180
160
140
120
100
80
60
40
20
0
Maximum Hourly
Operations Based on
Current Airspace &
ATC Design
06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23
Time of Day
Historic Movements
AIR-21 Induced Svc.
Source: William DeCota, Port Authority of New York
LGA
Capacity
Limit
DELAY
MIT
ICAT
DEMAND
Source: William DeCota, Port Authority of New York
MIT
ICAT
Source: FAA OPSNET data
Flight Delays at LGA
LGA
from 2000 to 2006
2006
Suggested Political
MIT
ICAT Solutions to Capacity Shortfall
• Full or Partial Privatization (eg AIR-21)
 May improve modernization,costs and strategic management
 Limited impact on capacity
•
Re-regulation
 Increased Costs to Consumer
• Demand Management (eg Peak Demand Pricing)
 Reduced service to weak markets
 Need to insure that revenues go to improved capacity
• Run System Tighter
 Requires improved CNS
 Safety vs Capacity Trade
•
Build more capacity
 Local community resistance
• Multi-modal transportation networks
MIT
ICAT
•
Conclusion
Technology in Pipeline will have limited impact on peak Capacity at
Currently Stressed Airports
 20% to 40% Optimistically
•
System will become (is) Capacity Restricted
•
Airlines will Schedule in Response to Market Demand




Delay Homeostasis (at Hubs)
Increased Traffic at Secondary Airports
High Frequency Service
Changing Value for Reliability
•
Overall system response is not clear
•
Need
 Protection of Airport and Spectrum Resources
 More runways in critical locations
 New ATM paradigms
•
Need for leadership
MIT
ICAT
Multi-Stakeholder Considerations &
Role of System Transition Plans
Demand
Performance
Aggregate
Inefficiency
Metrics
System
Capability
Implementation
Change
Process
Corrective
Actions
Selected
Actions
Stakeholder
Values Vi
Stakeholder
Objectives Oi
Stakeholder
Stakeholder
Stakeholder
Decisions
Stakeholder
i
Decisions
Decisions
Optimization
Monitoring &
Forecasting
Stakeholder
Awareness or
Perception Ai
Transformative
Actions
Capabilities
Research and
Development
Road Map
Research Needs
System Concepts
FAA>OEP
JPDO > NGATS
EU > CESARE
MIT
ICAT
Focus is the OEP 35 airports
More? Oakland Burbank Long Beach John Wayne-Orange County Tucson Albuquerque San Antonio Houston Hobby Palm Beach
MIT
ICAT