george mason university transportation lab air transportation system limitations, constraints and...

George Mason UniversityTransportation Lab

Air Transportation System Limitations, Constraints and Trends

George L. DonohueMarch 19-20, Wye Woods Conference Center© George Donohue 2002

2 George Mason University

Demand has grown Faster than National Infrastructure

Relative Growth in Transportation Modes

0.0

2.0

4.0

6.0

8.0

10.0

12.0

14.0

1960 1965 1970 1975 1980 1985 1990 1994

Year

Mile

s /

Mile

s in

19

60

AirCarrierBus

Truck

Car

Source: DOT Statistics


Initial Observationsand an Hypothesis

FACTS: Airspace above Airport Runway Thresholds (Operational

Capacity) is a Limited, Nationally Allocateable Commodity National Airport and Airspace Management Infrastructure

growth has seriously lagged behind Growth in Air Transportation Demand

Utilization of this Capacity Commodity is Constrained by Airline Schedule Conflicts, Delay Tolerance, FAA Ground Delay Programs and Aircraft Safety (i.e. Aircraft Spacing)

HYPOTHESIS: A DoT Supervised Auction System may be Required to

Efficiently allocate Airport Capacity within Delay and Safety constraints


Incentives for Operational Improvements and Modernization

Key Decision Points DP 1 NATCA Contract Negotiations and

Controller Mass Retirement Threat (Avg. Age=50 + Service=25) ~2007

DP 2 Termination of Slot Controls - 2007 DP 3 Sector Congestion and limits of Radio

Frequency Spectrum Availability ~ 2010 Transition Barriers

- Ground Based Infrastructure L----M----H- Airborne Equipment L----M----H - Labor Issues L----M----H- Regulation L----M----H- Required Culture Change L----M----H- Communication Bandwidth L----M----H- LACK OF INCENTIVES TO CHANGE !!!!


Outline

Limitations on Air Transportation Capacity

Safety, Capacity and Delay System Network EffectsFuture Security EffectsObservationsFuture Vision


Operational Capacity is a Limited Commodity

CMAX = 2 C AR MAX S i (XG)i Ri {Airports}

–K AK(t) {Airspace Management Intervention}

S = f ( Safety, ATC , Wake Vortex, etc.) ~ 0.6

AK(t) = (A/CREQUEST – A/CACCEPT) ~ [ 0 to >1,000] AK(t) = f ( GDP:Weather, Sector Workload Constraints )

C AR MAX ~ 64 Arrivals/Hour (set by Runway Occupancy Time)

Ri = Number of Runways at ith Airport

XGi = Airport Configuration Factor at ith Airport i = 1 to N, where N is approximately 60 Airports K = 1 to M, where M is typically much less than 100 Sectors


Regional Distribution of Airport Infrastructure is Uneven

TABLE 1 Regional Air Transportation Capacity Fraction For (57) Major AirportsNUMBER Estimated % Avg 8 yr TAF

HUB # A/C TURN Number Ops/Hr Cap97/ Growth 1997 ENP OPERATIONS

REGION R/W POINTS MODEL 1997 CapMAX Rate % X10E6 2012 1997

NORTH EAST 14 420 348 294 84 9 54 1,950,000 1,645,786

PACIFIC SOUTHWEST 9 262 403 298 74 10 43 2,205,000 1,670,280

PACIFIC NORTHWEST 22 353 693 455 66 8 62 3,364,000 2,549,603

NOTHERN MIDWEST 42 773 1090 684 63 32 99 5,522,000 4,040,088

ATLANTIC COAST 13 269 438 241 55 8 31 1,701,000 1,347,458

CENTRAL MIDWEST 12 205 237 131 55 3 19 1,496,000 1,114,207

WEST 22 415 758 405 53 9 62 3,180,000 2,270,307

SOUTHEAST 21 424 776 391 50 -2 54 2,704,000 2,190,557

FLORIDA & LATIN AM 14 322 602 287 48 18 48 2,114,000 1,608,673

SOUTH SOUTHWEST 27 380 892 433 48 16 59 3,468,000 2,424,105

TOTAL 196 3823 6239 3620 58 11 532 27,704,000 20,861,064

% NATIONAL TOTAL 89 78 77

Donohue and Shaver, TRB 2000


Airport Diseconomies of Scale

Airport Runway Diminishing Returns

0.00

0.20

0.40

0.60

0.80

1.00

1.20

0 2 4 6 8

RUNWAYS

XG

for

Typ

ical

Air

port

s

XG FactorPower (XG Factor)


Non-Linear Network Characteristics

NAS is a Highly Non-Linear, Adaptive System Controller-in-the-Loop AOC-in-the-Loop Independent Network

Schedules Stochastic In Nature May exhibit Chaotic

Behavior under Some Conditions

Additive Improvements DO NOT result in Additive Increases in NAS Capacity ie. pFAST, Runways, etc.

EXAMPLE OF AIR TRANSPORTATION SYSTEM NON-LINEARITY

0

50

100

150

200

250

300

350

400

450

500

0 100 200 300 400 500

TOTAL AIRPORT CAPACITY (OPS/HR)

SYST

EM C

APAC

ITY

(OPS

/HR)

LINEAR SYSTEM

NON-LINEAR SYSTEM

3 Airport Network at 100 operations each + 50 operations increase


Airline Schedule has a Strong Effect on Network Performance – Model Prediction to 20% Airport Capacity Increase

DPAT Simulation, benchmark capacity, airports ranked by delay extent, with sector

0

0.5

1

1.5

2

2.5

3

Ori

gin

alL

GA

EW

RP

HL

BO

SM

CO

JFK

PH

XO

RD

SL

CT

PA

IAD

MIA

DE

NL

AS

DC

AL

AX

DT

WA

TL

MS

PC

LT

CV

GS

FO

SE

AS

TL

DF

WIA

HP

ITS

AN

BW

IM

EM

original1 2 3 4 5 6 7 8 9 1011 12131415161718192021222324252627282930

(min

)

Average Arrival Delay( queue on runway)

Linear System Response

MITRE DPAT MODEL


Outline




Capacity vs. Delay Penalty

5

10

15

20

25

Arrival_Rate5

10

15

20

25

Departure_Rate

0

200

400

600

Delay

5

10

15

20

25

Arrival_Rate

[3] “ACE 1999 Plan,” CD-ROM. Federal Aviation Administration – Office of system capacity.

IAD

Arrivals

Depart

Delay


NY LaGuardia: A non-Hub Maximum Capacity Airport

1 Arrival Runway 1 Departure Runway 45 Arrivals/Hr (Max) 80 Seconds Between Arrivals 11.3 minute Average Delay 77 Delays/1000 Operations 40 min./Delay

TOTAL SCHEDULED OPERATIONS AND CURRENT OPTIMUM RATE BOUNDARIES

0

4

8

12

16

20

24

28

32

7 8 9 10 11 12 13 14 15 16 17 18 19 20 21

Schedule Facility Est. Model Est.


New York LaGuardia Airport Arrival- Departure Spacing VMC

0

10

20

30

40

50

60

0 10 20 30 40 50 60

Departures per Hour

Arr

iva

ls p

er

Ho

ur

ASPM - Apr 2000 - Visual Approaches

ASPM - Oct 2000 - Visual Approaches

Calculated VMC Capacity

Optimum Rate (LGA)

40,40

Each dot represents one hour of actual traffic

during April or October 2000

45 Arr./Hr/RW@ 80 sec separation

DoT/FAA


Atlanta: A Maximum Capacity Fortress Hub Airport

2 Runways – Arrivals 2 Runways – Departures 50 Arrivals/Hr/RW – Max 72 Seconds Between Arrivals 8.5 minutes Average Delay 36 Delays/1000 Operations 38 min./delay

TOTAL SCHEDULED OPERATIONS AND CURRENT OPTIMUM RATE BOUNDARIES

0

10

20

30

40

50

60

7 8 9 10 11 12 13 14 15 16 17 18 19 20 21

Schedule Facility Est. Model Est.


Atlanta AirportArrival-Departure Spacing VMC

0

20

40

60

80

100

120

0 20 40 60 80 100 120

Departures per Hour

Arr

iva

ls p

er

Ho

ur

ASPM - April 2000 - Visual Approaches

Calculated VMC Capacity

Optimum Rate (ATL)100,100

Each dot represents one hour of actual traffic

during April 2000

50 Arr/Hr/Rw

@72 sec

Separation

DoT/FAA


Major US Airport Congestion

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9

DEN

BWI

JFK

PIT

CLT

DFW

DTW

IAD

EWR

PHL

SFO

LGA

MSP

SEA

ORD

STL

ATL

LAXA

IRP

OR

T

DEMAND / CAPACITY RATIO

J. D. Welch and R.T. Lloyd, ATM 2001

Queuing Delays Grow Rapidly


Aircraft Arrival Rate:Distance-Time Relationship

0

10

20

30

40

50

60

70

80

0 1 2 3 4 5 6 7

DISTANCE ( NMi)

AR

RIV

AL

S /

RW

/ H

R 120 Knots

130 Knots

140 Knots

Spacing

(sec)

60

90

180

120

72


LGA Aircraft Inter-Arrival Time Distribution

LGA Arrival Seperation Histogram

-5

0

5

10

15

20

25

30

35

0 50 100 150 200 250 300 350

Aircraft Inter-Arrival Time (seconds)

Nu

mb

er o

f A

ircra

ft i

n 2

0 S

ec. B

ins

LGA VMC

96 Sec. WV Separation Expected Value

μ = 134 sec., σ = 73 sec.

70 sec

SAFETYCAPACITY


Possible Relationship Between Safety and Capacity: ATM Technology Effect

Hypothesis: SAFETY-CAPACITY SUBSTITUTION CURVES

-

20.00

40.00

60.00

80.00

100.00

120.00

140.00

0.00 2.00 4.00 6.00 8.00 10.00 12.00

MILLION DEPARTURES/HULL LOSS

U.S

. MIL

LIO

N D

EPA

RT

UR

ES/

YE

AR

15 HULL LOSS/YR5 HULL LOSS/YR2 HULL LOSS/YRS-C @ CURRENT TECH.S-C @ REDUCED SEP.

US


Outline




The Semi-Regulated Market Does Not Act to Minimize Delay: LGA Air 21 Impact

Source: William DeCota, Port Authority of New York

LaGuardia Airport

0

2040

60

80

100120

140

160180

200

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.

Maximum Hourly Operations Based on Current Airspace & ATC Design


Annual and Seasonal Delay Trends(Note Possible Effect of Air 21 on LGA & System)

OPSNET Total System Delays

0

10

20

30

40

50

60

Month

Th

ou

sa

nd

s o

f D

ela

ys

2000

1999

1998

1997

1996

1995


Outline




Observations Approximately 10 of the Top US Hub Airports are

Operating close to Maximum Safe Capacity Demand / Capacity Ratio’s Greater than 0.7 lead to Very

Rapid Increase in Arrival and Departure Delays Higher Delays Lead to Loss of Schedule Integrity 25 New Runways Not a Solution

Airline Hub and Spoke Network System Produces a Highly Non-Linear, Connected System Weather, Security or Terminal Delays Propagate System

Wide Airline Schedules are part of the Problem & Solution

ATC Sector Controller Workloads and Weather also Produce Network Choke-Points that Produce Capacity Constraints


Observations (cont.) 100% EDS Baggage Screening will either Increase

Delays or Travel Block Times for Commercial Ops Current Regulations on Airlines and Airports do

not provide Incentives for either Safe or Efficient Operations Airlines are over-scheduling Major Airports ATC is spacing Aircraft at the limits of current

technology leading to growing safety concerns Airlines are moving to Smaller aircraft to increase

frequency of operations and profitability, leading to increased congestion and delays

Airlines are resisting modernizing their aircraft with the technology required to decrease spacing and increase capacityIncentives are to be last to equip


Outline




Vision: Incentives for Operational Improvements and Modernization

Brief Summary of Vision:

Major Hub Airports will Allocate Slots by DoT Auctions:-Both Strategic, Near Term and Spot Auctions-Peak runway loading will be reduced to Government Established

Safety and Capacity optimized schedules-Aircraft Size will be driven by a combination of airline profits and maximum enplanement opportunities

Business travel will migrate to Travel on Demand via air-taxi or private aircraft ownership and operation

Increased En-route Traffic density will be accommodated by Aircraft Self Separation-Technology-Equipped Flight Corridors

Auctions will provide incentives for aircraft technology insertion and a government contract to provide enhanced benefits

George Donohue


Outline




Key Airport System FlowsMIT Queuing Model

ArrivalArrivalPathsPaths

Dept.Dept.PathsPaths

RunwaysRunways TaxiwaysTaxiways RampRamp GatesGates

PaxPaxScreenScreen

Ckd BagCkd BagScreenScreen

Check-InCheck-InIDID

Drop-offDrop-offParkingParking

EntryEntryFixFix

DepartureDeparture FixFix

AirsideAirside GroundsideGroundside

ArrivalsArrivals

DeparturesDepartures

PassengersPassengers

Bags/CargoBags/Cargo

GndGndTransTrans


0

100

200

300

400

500

600

700

800

900

1000

6 7 8 9 10 11 12 13 14 15 16 17 18

Time of Day

Nu

mb

er

of

Ch

ec

k-I

n B

ag

s p

er

3-M

inu

te

Inc

rem

en

t Scheduled Demand, No Spread Demand Spread = 30 mins (flat)

Total Check-InBags = 56,516

Baggage: Actual and Spread Demand for 1998 DFW Case (RAND Study)

Dr. R. Shaver


Planned Time of Arrival According to Passenger Propensity to Accept Risk

0

15

30

45

60

75

90

105

120

0.1%1.0%10.0%100.0%

Probability that Bag Misses Plane

Pla

nn

ed

Tim

e o

f A

rriv

al

Be

fore

Sc

he

du

led

D

ep

art

ure

--In

clu

des

15 m

inu

te U

np

lan

ned

D

ela

y B

efo

re B

ag

ga

ge C

he

ck-I

n (

min

ute

s)

Machines Deployed = [17:24] Machines Deployed = [19:28] Machines Deployed = [20:29] Machines Deployed = [22:32] Machines Deployed = [24:34] Machines Deployed = [15:22]

“The Passenger’s View”

R. Shaver, RAND

george mason university transportation lab air transportation system limitations, constraints and...

Documents

airport capacity

george mason university

capacity commodity

trends george

safety constraints

h lack

h regulationl

th airport xg