capacity and delays at newark airport, theoretical and experimental approaches
DESCRIPTION
Capacity and delays at Newark Airport, theoretical and experimental approaches. Thomas Morisset. Objectives. Review and apply essential notions studied in class: Theoretical capacity Actual capacity (based on maximum throughput) Delay estimates based on queuing theory Actual delays - PowerPoint PPT PresentationTRANSCRIPT
Capacity and delays at Newark Airport,
theoretical and experimental approaches
Thomas Morisset
Objectives
• Review and apply essential notions studied in class:– Theoretical capacity– Actual capacity (based on maximum throughput)– Delay estimates based on queuing theory– Actual delays
• Compare and understand the differences between theory and reality.
Newark Airport
• 443,000 movements• 36 million passengers• 13th busiest airport in the United States• 22 minutes mean arrival delay, congested• EWR-JFK-LGA metroplex: – 1,276,000 movements– 110,000,000 passengers– busiest multi-airport system in the world
figures for 2007
Experimental approach to capacity [1]
26.7 mov per 15 min
Experimental approach to capacity [2]
24 mov per 15 min
Experimental approach to capacity [3]
22 mov per 15 min
27262524232221201918171615141312111098765432100.0%
1.0%
2.0%
3.0%
4.0%
5.0%
6.0%
7.0%
8.0%
9.0%
VFRIFR
movements realized per 15-minute intervals
freq
uenc
y in
200
7 98%
per
centi
le
Experimental approach to capacity [4]
22 mov per 15 min
• Capacity varies depending on runway configuration, weather conditions:
Experimental approach to capacity [5]
Utilization 3rd runway % time % movts mov/15min (98 %ile)
With 11/29 36.6% 39.0% 23
Without 11/29 62.8% 60.4% 21
Weather condition % time % movts mov/15min (98 %ile)
VFR 83.4% 84.3% 22
IFR 16.6% 15.7% 21
• Run the MACAD* capacity simulation program• Treat the simple case of 2 close parallel
runways (4R|4L or 22L|22R)• Data needed:– aircraft mix: ASPM– IFR separation standards: FAA AC’s– VFR separation estimates**– approaching speeds, runway occupancy times
Theoretical approach to capacity
* Odoni, Stamatopoulos ** Trani
• Run the MACAD* capacity simulation program• Treat the simple case of 2 close parallel
runways (4R|4L or 22L|22R)• Data needed:– aircraft mix: computed from ASPM– IFR separation standards: FAA AC’s– VFR separation estimates**– approaching speeds, runway occupancy times
Theoretical approach to capacity
* Odoni, Stamatopoulos ** Trani
RESULTS:
- IFR theoretical capacity: 18 movements per 15 minutes
- VFR theoretical capacity: 20 movements per 15 minutes
Theoretical capacities are slightly lower than capacities yielded by theoretical approaches
Method / Sourcearrivals+departures in 15
minutes capacity
98th percentile 22
Pareto envelope (>100 obs) 22
Scheduled demand 23
FAA benchmark report 21-23
Macad Model (VFR) 20
All capacities
Experimental approach to delays [1]
7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 220
5
10
15
20
25
30
35
40
Arrival delay at EWR in 2007
hourly delayAverage
Hours of the day
min
utes
Experimental approach to delays [2]
-60 -40 -20 0 20 40 60 80 100 120 140 160 1800.00%
0.50%
1.00%
1.50%
2.00%
2.50%
3.00%
3.50%
4.00%
4.50%
5.00%
EWR arrival VFR 2007
arrival delays (1-minute intervals)
freq
uenc
y
Experimental approach to delays [3]
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
-10.0
0.0
10.0
20.0
30.0
40.0
50.0
60.0
70.0
80.0
90.0
-10.0
0.0
10.0
20.0
30.0
40.0
50.0
60.0
70.0
80.0
90.0
arrival delay departure delay +taxi out delay schedule Actual
hours
num
ber o
f mov
emen
ts
min
utes
Experimental approach to delays [4]
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 230.0%
10.0%
20.0%
30.0%
40.0%
50.0%
60.0%
70.0%
80.0%
90.0%
100.0%
Runway configurations
Rest
29, 4R / 4L
11, 4R / 4L
4R / 4L
11, 22L / 22R
22L / 22R, 29
22L / 22R
Experimental approach to delays [5]
Performance of different runway configurations during peak hour (4pm – 8pm)
Runway configuration All 22L | 22R 11, 22L | 22R 4R | 4L 4R, 11 |
4L4R, 29 |
4L
Frequency 100% 30.5% 27.8% 14.4% 13.9% 8.0%
Arrival delay 31.2 43.9 14.5 48.0 18.4 25.4
Departure delay 27.1 35.1 16.8 37.5 20.0 23.4
+ Taxi out delay 20.8 17.7 20.6 19.9 28.2 23.2
• Run a queuing model program: DELAYS*• For different levels of capacity• For a particular day of demand (Feb 4 2007), NOT average demand
Theoretical approach to delays
* Odoni, Kivestu
Conclusions
• Capacity:– theoretical and experimental approach within 15% of each
other, a good match– can assist with the choice of runway configuration,
comparing airports, etc.• Delays:– theoretical estimates much lower than reality– delays cannot be explained by the limited capacity of 1
single airport– must consider the entire network of airport to include
delay propagation