Appendix B Ultimate Airport Capacity and Delay Simulation
Appendix B Ultimate Airport Capacity and Delay Simulation Modeling Analysis
ULTIMATE AIRPORT CAPACITY & DELAY SIMULATION MODELING ANALYSIS B TABLE OF CONTENTS B.1
Introduction ................................................................................ 1
B.2
Simulation Modeling Assumption and Methodology ........ 4 B.2.1
B.3
B.4
Runway Usage .................................................................................. 4
B.2.2
Airfield and Terminal Enhancements .......................................... 5
B.2.3
Demand Levels ................................................................................ 5
B.2.4
Weather Conditions ....................................................................... 6
B.2.5
FAA Separation Initiatives ............................................................ 7
B.2.6
Statistical Evaluation Metrics ....................................................... 9
VFR Simulation Experiment Results and Analysis ............ 10 B.3.1
Present Day Separation ............................................................... 10
B.3.2
Wake Recat ..................................................................................... 10
Morning IFR then VFR Simulation Experiment Results and Analysis ..................................................................................... 12
EXHIBITS
TABLES
Exhibit B.1-1 | Existing Airfield Configuration ......................................... 3
Table B.1-1 | Existing Runway Lengths ...................................................... 2
Exhibit B.2-1 | Standard 28-1 Runway Configuration ............................. 4
Table B.2-1 | Design Day Flight Schedule Fleet Mix Summary ............. 7
Exhibit B.2-2 | Forecast Flight Schedule Rolling Hour Demand Comparison .......................................................................... 6
Table B.2-2 | Forecast B757 and B767 Aircraft Fleet Demand Levels .. 8
Exhibit B.2-3 | FAA 7110.308 Dependent Approaches Parallel Runways Spaced Less Than 2,500 feet Apart Application ........................................................................... 8 Exhibit B.6-1 | VFR with Present Day Separation Delays by Hour ...... 17 Exhibit B.6-2 | VFR with Present Day Separation Percentage of Total Operations Delayed................................................ 18 Exhibit B.6-3 | VFR with Wake Recat Delays by Hour ............................ 19 Exhibit B.6-4 | VFR with Wake Recat Percentage of Total Operations Delayed .......................................................... 20 Exhibit B.6-5 | Morning IFR then VFR with Present Day Separation Delays by Hour................................................................... 21 Exhibit B.6-6 | Morning IFR then VFR with Present Day Separation Percentage of Total Operations Delayed ..................... 22
B.4.1
Present Day Separation ............................................................... 12
B.4.2
Wake Recat ..................................................................................... 12
B.4.3
FAA JO 7110.308 ........................................................................... 13
Exhibit B.6-8 | Morning IFR then VFR with Wake Recat Percentage of Total Operations Delayed ............................................ 24
B.5
Annualized Weighted Simulation Delay Results .............. 14
B.6
Maximum Practical Capacity Analysis ................................ 16
Exhibit B.6-9 | Morning IFR then VFR Operation with FAA JO 7110.308 Delays by Hour ................................................. 25
B.7
B.6.1
VFR with Present Day Separation.............................................. 17
B.6.2
VFR with Wake Recat.................................................................... 19
B.6.3
Morning IFR then VFR with Present Day Separation ............. 21
B.6.4
Morning IFR then VFR with Wake Recat................................... 23
B.6.5
Morning IFR then VFR with FAA JO 7110.308 ......................... 25
Table B.3-1 | VFR Simulation Analysis Results ........................................ 10 Table B.3-2 | VFR Simulation with Wake Recat Analysis Results......... 11 Table B.4-1 | Morning iFR then VFR Simulation Analysis Results ....... 12 Table B.4-2 | Operation Morning IFR then VFR with Wake Recat Simulation Analysis Results ............................................... 13 Table B.4-3 | Morning IFR then VFR with FAA 7110.308 Simulation Analysis Results .................................................................... 14 Table B.5-1 | Weighted Average Delay Summary with Present Day Separation ............................................................................. 15 Table B.5-2 | Weighted Average Delay Summary with Wake Recat .. 15 Table B.5-3 | Weighted Average Delay Summary with Wake Recat And FAA JO 7110.308 .......................................................... 15
Exhibit B.6-7 | Morning IFR then VFR with Wake Recat Delays by Hour ..................................................................................... 23
Exhibit B.6-10 | Morning IFR then VFR with FAA JO 7110.308 Percentage of Total Operations Delayed ................... 26
Conclusion ................................................................................ 27
AIRPORT DEVELOPMENT PLAN – DRAFT FINAL
Ultimate Airport Capacity & Delay Simulation Modeling Analysis | B - i
SFO Airport Development Plan
Ultimate Airport Capacity and Delay Simulation Modeling Analysis
Appendix B Ultimate Airport Capacity and Delay Simulation Modeling Analysis B.1
Introduction
SFO Airport Development Plan
Ultimate Airport Capacity and Delay Simulation Modeling Analysis
Exhibit B.1-1 depicts the existing runway configuration for SFO. Table B.1-1 provides the dimensions for each runway. The ultimate capacity and delay simulation analysis focuses primarily on the maximum capacity of the standard 28-11 runway operating configuration. In this configuration, arriving flights use Runways 28L and 28R while departing flights use Runways 1L and 1R. Alternate runway configurations were not evaluated in this analysis as the 28-1 plan and the closely related 28-28 plan are used over 95 percent of the time. Table B.1-1 | Existing Runway Lengths
Evaluating the ultimate capacity of San Francisco International Airport (SFO) is necessary to understand the Airport’s growth potential. In order to evaluate the ultimate capacity of the Airport, the SFO Bureau of Planning and Environmental Affairs (BPEA) tasked Landrum & Brown Inc. (L&B) with using the FAA’s Airport and Airspace Simulation Model (SIMMOD) simulation software to evaluate the ultimate capacity and delays of the current runway system at SFO. In addition, various operational configurations were evaluated at increased demand levels to determine impacts on capacity and delays. Due to constraints which limit the ability for runway expansion at the Airport, future growth and success depends on new procedures and methods that increase capacity while using the existing airfield configuration. These were considered in the development of the model simulation.
Designation
Length
Width
1L-19R
7,500’
200’
1R-19L
8,646’
200’
10L-28R
11,381’
200’
10R-28L
11,870’
200’
The following sections of this report provide a description of the simulation modeling experiments including: x An overview of the simulation modeling methodology and assumptions x Simulated weather and operating configurations o
Visual flight rules (VFR) weather conditions – pilots visually navigate and maintain separation from clouds and other aircraft
o
Morning instrument flight rules (MIFR) weather conditions – visibility less than 3 statute miles and ceiling less than 1,000 feet above ground level
x Explanation of various simulated runway operating configurations and airspace configurations which could potentially increase runway capacity at SFO o
FAA Wake Turbulence Recategorization (Wake Recat)
o
FAA Job Order (JO) 7110.308 1.5 nautical mile Dependent Approaches to Parallel Runways Spaced Less Than 2,500 feet Apart arrival procedures on Runways 28L and 28R
x Evaluation of the maximum practical capacity of the airfield
The results of the simulation analyses are intended to provide SFO and airport stakeholders with a realistic assessment of the maximum theoretical capacity of the Airport. The maximum capacity assessment in this report can be used to guide the Airport in planning future flight demand levels, thereby ensuring flight schedule integrity and minimizing delays.
1 Runway
operation patterns at SFO are abbreviated using the code “x-y,” where x represents the primary arrival runway pair and y represents the primary departure runway pair.
Draft Final: September 2016
Appendix B | Page 1
Draft Final: September 2016
Appendix B | Page 2
SFO Airport Development Plan
B.2
Ultimate Airport Capacity and Delay Simulation Modeling Analysis
Simulation Modeling Assumption and Methodology
This section provides an overview of the methodology and assumptions that were used in the simulation modeling analyses.
B.2.1 SFO Airport Development Plan
Ultimate Airport Capacity and Delay Simulation Modeling Analysis
Exhibit B.1-1 | Existing Airfield Configuration
Runway Usage
As stated in Section B.1, the standard 28-1 operation was the primary runway configuration simulated in this analysis. In VFR conditions, the standard runway usage consists of two paired arrivals on Runways 28R and 28L and two departures on Runways 1L and 1R between each arrival pair. Departures that cannot use Runways 1L or 1R due to aircraft performance requirements use Runways 28L or 28R instead. Following a 2013 extension of Runway 10R-28L, the majority of long-haul international departures use this runway. In the standard 28-1 runway operation, the majority of general aviation and cargo departures use Runways 1L and 1R for departures. Exhibit B.2-1 depicts the standard 28-1 runway usage configuration which was evaluated in the simulation modeling analysis. Exhibit B.2-1 | Standard 28-1 Runway Configuration
Source: SFO Bureau of Planning and Environmental Affairs, SFO Airport Layout Plan, 2014; Landrum & Brown, Inc., 2016
Draft Final: September 2016
Appendix B | Page 3
Draft Final: September 2016
Appendix B | Page 4
SFO Airport Development Plan
B.2.2
Ultimate Airport Capacity and Delay Simulation Modeling Analysis
SFO Airport Development Plan
Ultimate Airport Capacity and Delay Simulation Modeling Analysis
Exhibit B.2-2 | Forecast Flight Schedule Rolling Hour Demand Comparison
Airfield and Terminal Enhancements
The future configuration of B/A B was simulated with 18 gates, which does not account for multiple aircraft ramp system (MARS) gates in the new boarding area which will allow additional aircraft to park at some gates. Existing airline gating assignments were used, and those aircraft which currently park at B/A B were assigned to the reconfigured boarding area.
B.2.3
Demand Levels
Previous SIMMOD analyses conducted for the Runway Safety Area (RSA) improvement project simulated demand levels up to 1,300 daily operations, which represented a 2014 forecast demand level. During peak summer months, the Airport currently operates demand levels that occasionally exceed 1,300 daily operations. To evaluate the impacts of increased aircraft operations on the capacity of the airspace and runway system at SFO, several forecast aircraft demand levels were simulated. As part of the update to the Airport’s ultimate forecast, L&B prepared forecast flight schedules for 2016, 2018, and 2021 demand levels. The simulated operations in the flight schedule are as follows: x
2016 demand level – 1,336 daily operations
x
2018 demand level – 1,369 daily operations
x
2021 demand level – 1,435 daily operations
In addition to these three forecast demand levels, a 5 percent flight schedule increase of the 2021 demand level was evaluated. This demand level simulated approximately 1,504 daily operations. Military and helicopter operations which were identified in the original design day flight schedules (DDFS) were not simulated in the analysis.
The rolling hour comparison of the 2016, 2018, and 2021 demand levels indicates the following peak hour demand rates anticipated at SFO:
Exhibit B.2-2 depicts a rolling hour comparison based on 10 minute time intervals (bins) of the current day 2013 demand levels as well as the three forecast demand levels which were simulated for this analysis.
x
2016 (1,336 daily operations): 55 arrivals, 58 departures, 105 total operations
x
2018 (1,369 daily operations): 55 arrivals, 58 departures, 104 total operations
x
2021 (1,435 daily operations): 59 arrivals, 62 departures, 110 total operations
The arrival, departure, and total operations peaks occur at different times of day, so the peak hour totals do not equal the sum of arrivals and departures.
B.2.4
Weather Conditions
Each demand level was evaluated under VFR conditions and morning IFR then VFR (MIFR) conditions. The MIFR condition is the result of a typical weather pattern where a layer of cool marine air draws fog through the Golden Gate and down the coast of the Bay. Usually this lasts until 10:30 am when the sun begins to evaporate the fog and visibility improves. The Airport is reduced to a single arrival runway (Runway 28R) when operating under IFR conditions. With only one primary arrival runway available, Airport capacity decreases, resulting in a greater number of delays.
Draft Final: September 2016
Appendix B | Page 5
Draft Final: September 2016
Appendix B | Page 6
SFO Airport Development Plan
B.2.5
Ultimate Airport Capacity and Delay Simulation Modeling Analysis
SFO Airport Development Plan
Ultimate Airport Capacity and Delay Simulation Modeling Analysis
Table B.2-2 | Forecast B757 and B767 Aircraft Fleet Demand Levels
FAA Separation Initiatives
FAA Wake Recat As part of its NextGen air traffic modernization initiatives, the Federal Aviation Administration (FAA) is reevaluating aircraft wake turbulence categorizations in a program known as “Wake Recat.” Analysis of certain aircraft fleets has determined that existing wake turbulence separation requirements for lead-to-trail aircraft combinations may be excessive during IFR operations. Wake Recat increases runway capacity by reducing the lead-to-trail aircraft separations for various aircraft types.
B757 B767 Total Flights
B757/B767 Operations in Forecast Flight Schedules 2016 2018 20 10 21 19 41 29
2021 0 13 13
FAA JO 7110.308 In an effort to increase arrival capacity during IFR conditions at SFO, the FAA JO 7110.308 1.5 Nautical Mile Dependent Approaches to Parallel Runways Spaced Less Than 2,500 feet Apart arrival procedure was simulated. During IFR operations, the Airport is ordinarily reduced to a single arrival runway which reduces capacity to 30-32 operations per hour. The FAA 7110.308 procedure allows ILS approaches to be operated simultaneously on Runways 28L and 28R using a staggered arrival procedure.
Table B.2-1 provides a summary of the DDFS fleet mix by Wake Recat category. Table B.2-1 | Design Day Flight Schedule Fleet Mix Summary DDFS Fleet Mix (% Of Total Aircraft Operations) Wake Category
Common A/C
2016
2018
2021
CAT A
A380
0.4%
0.6%
0.7%
CAT B
B747, B777, A340
7.7%
7.9%
8.6%
CAT C
B767, A310, MD11
1.6%
1.4%
0.9%
CAT D
B757, B737, A320
61.5%
62.0%
62.5%
CAT E
CRJ2, CRJ7, E170
19.2%
20.8%
21.4%
CAT F
E120, E190, H25B
9.7%
7.4%
6.0%
Total
100.0%
100.0%
100.0%
Daily ops
1,336
1,369
1,435
Exhibit B.2-3 depicts the basic guidelines for the application of the FAA JO 7110.308 procedure when conducted on Runways 28L and 28R. Exhibit B.2-3 | FAA 7110.308 Dependent Approaches Parallel Runways Spaced Less Than 2,500 feet Apart Application
Wake Recat primarily benefits the B757 and B767 aircraft fleets by allowing reduced in-trail separation requirements in IFR conditions. As proposed in Wake Recat, B757 aircraft are grouped with other narrowbody aircraft including the B737 and A320. The B767 is removed from the “heavy” aircraft category which contains aircraft including the B747 and B777 and placed in its own category with new in-trail separation requirements. Table B.2-2 provides a summary of the forecast B757 and B767 aircraft demand levels in the 2016, 2018, and 2021 flight schedules. The B757 is estimated to be phased out of SFO operations by the year 2021 and replaced with more efficient narrowbody aircraft such as the B737-900ER. Usage of the B767 is also expected to decline in future years, and that aircraft type will be replaced with new models such as the A350 and B787.
Draft Final: September 2016
Appendix B | Page 7
Draft Final: September 2016
Appendix B | Page 8
SFO Airport Development Plan
Ultimate Airport Capacity and Delay Simulation Modeling Analysis
The FAA JO 7110.308 arrival procedure requires the lead aircraft in a staggered pair to use the lower approach to the near end of a runway or the lower glide slope angle. Since the runway thresholds for Runways 28L and 28R are abeam each other, Runway 28L is considered the lower approach as a result of the 2.85 degree glide slope angle for instrument approaches on the runway. Narrowbody and small aircraft can always be the lead aircraft in a staggered pair. However, heavy and super heavy category aircraft are required to be the trailing aircraft to mitigate the impacts of wake turbulence on the paired aircraft. For simulation purposes, heavy and super heavy aircraft were assigned to Runway 28R. Rules dictate that arrivals can be diagonally staggered using a minimum of 1.5 nm of separation. However, given the complexity of implementing the staggered operation at SFO, this operation will most likely not achieve the 1.5 nm separation on a consistent basis during the initial implementation. It is estimated that staggered separations would range between 1.5 nm and 2.0 nm; therefore, a separation of 1.8 nm was used in the simulation.
B.2.6
SFO Airport Development Plan
B.3
VFR Simulation Experiment Results and Analysis
This section presents the results of the two VFR simulation experiments, one operating under the present day VFR separation criteria and other under the Wake Recat separation criteria. Although Wake Recat is intended primarily to reduce in-trail separations during IFR operations, this analysis simulated it in a VFR environment to determine if any delay improvements or capacity gains could be achieved. FAA JO 7110.308 was not simulated, as this procedure affects only IFR operations.
B.3.1
Present Day Separation
Table B.3-1 presents a summary of some of the standard delay evaluation metrics from the simulation model. The delays in each of these categories increase when operating at higher demand levels. Table B.3-1 | VFR Simulation Analysis Results
Statistical Evaluation Metrics
The simulation analysis used the following statistical metrics in evaluating the final results of each simulation experiment: x Average arrival delays x Average departure delays x Average overall delays
Total Operations Average Arrival Delay Average Departure Delay Average Delay
x Peak hour arrival delays
Peak Hour Arrival Delay Peak Hour Departure Delay
x Peak hour departure delays x Weighted average delays x Demand hour delay profiles for VFR and morning IFR simulations x Percentage of total operations delayed by more than 15, 30, 45, and 60 minutes The following sections of this report will discuss the statistical analysis results of the various simulation experiments. The results of the simulation analysis are used to estimate the maximum practical capacity and acceptable levels of delay for the Airport which is the primary focus of this analysis.
Ultimate Airport Capacity and Delay Simulation Modeling Analysis
B.3.2
2016 VFR
2018 VFR
2021 VFR
2021 + 5% VFR
1,336
1,369
1,435
1,504
5.7
6.0
8.4
9.8
5.1
4.8
6.8
8.1
5.4
5.4
7.6
8.9
11.9
12.6
17.9
23
10.7
7.9
11.6
20.3
Wake Recat
Table B.3-2 represents a summary of the simulated delay results of the Wake Recat simulation experiments operating in VFR conditions. For comparison purposes, the delay results from the simulation experiments under present day VFR separations are included to show the differences in delays. In general, the application of Wake Recat provides minimal delay savings under VFR conditions. The primary reason is that lead-to-trail aircraft in-trail VFR separations at SFO are typically larger than minimum IFR separations because departures from Runways 1L and 1R are sequenced in between arrivals on Runways 28L and 28R. A previous analysis of SFO radar data indicates that in VFR conditions, the minimum in-trail separation between narrowbody arrival pairs is approximately 3.8 nm, which is considered the minimum required to sequence departures from Runways 1L and 1R in between arrivals. Due to the dependent nature of arrivals and departures at SFO, Wake Recat may only yield minimal benefits in delay reductions and capacity gains. In general, Wake Recat would not provide significant benefit to SFO under VFR operating conditions.
Draft Final: September 2016
Appendix B | Page 9
Draft Final: September 2016
Appendix B | Page 10
SFO Airport Development Plan
Ultimate Airport Capacity and Delay Simulation Modeling Analysis
Table B.3-2 | VFR Simulation with Wake Recat Analysis Results 2016 VFR
Total Operations Average Arrival Delay Average Departure Delay Average Delay Peak Hour Arrival Delay Peak Hour Departure Delay
2016 VFR Recat
2018 VFR
1,336
2018 VFR Recat
SFO Airport Development Plan
B.4
2021 VFR
1,369
2021 VFR Recat
2021 + 5% 2021 + 5% VFR VFR Recat
1,435
1,504
5.7
5.5
6.0
5.9
8.4
8.2
11.6
9.8
5.1
5.9
4.8
5.1
6.8
6.4
8.4
8.1
5.4
5.7
5.4
5.5
7.6
7.3
10.0
8.9
11.9
9.4
12.6
13.7
17.9
16.1
22.4
23.0
10.7
12.4
7.9
9.6
11.6
14.6
19.2
20.3
Ultimate Airport Capacity and Delay Simulation Modeling Analysis
Morning IFR then VFR Simulation Experiment Results and Analysis
This section presents the results of the morning IFR then VFR (MIFR) simulation experiments. This weather pattern was simulated using each of the four demand levels and both of the FAA air traffic procedure improvements.
B.4.1
Present Day Separation
The morning IFR period typically occurs during the morning peak arrival push and ends around 10:30 am. With present day separation requirements, the Airport is reduced to a single arrival runway during the morning IFR period, which reduces arrival capacity to approximately 30-32 operations per hour. For simulation purposes, dual departures on Runways 1L and 1R were sequenced between the single arrival stream on Runway 28R. When the morning IFR period ends, simultaneous paired arrivals commence on Runways 28L and 28R. Table B.4-1 presents a comparison of the simulated delay results for the four demand levels of the MIFR modeling experiments. Increased demand levels when operating in IFR conditions result in significantly higher delays, particularly for arrivals. Late arrivals typically shift the peak departure banks to the mid to late afternoon periods, overlapping with the peak international departure push and thereby resulting in higher than normal departure delays. Table B.4-1 | Morning IFR then VFR Simulation Analysis Results
Total Operations Average Arrival Delay Average Departure Delay Average Delay Peak Hour Arrival Delay Peak Hour Departure Delay
B.4.2
2016 MIFR
2018 MIFR
2021 MIFR
2021 + 5% MIFR
1,336
1,369
1,435
1,504
23.8
31.9
36.4
49.8
10.6
11.8
16.1
18.2
17.2
21.8
26.3
34.1
66.4
78.3
86.1
103.5
24.4
37.5
55.6
47.6
Wake Recat
Table B.4-2 presents a summary of the simulated delay results of the Wake Recat simulation experiments operating in morning IFR then VFR conditions. For comparison purposes, the delay results from the MIFR simulation experiments under present day separations is included to show the differences in delays between the various experiments.
Draft Final: September 2016
Appendix B | Page 11
Draft Final: September 2016
Appendix B | Page 12
SFO Airport Development Plan
Ultimate Airport Capacity and Delay Simulation Modeling Analysis
Under Wake Recat separation criteria, significant gains are achieved with B757 and B767 aircraft, resulting in noticeable arrival delay improvements. However, reduced arrival in-trail separation results in fewer gaps for departures on intersecting Runways 1L and 1R, thus resulting in higher departure delays.
SFO Airport Development Plan
Table B.4-3 | Morning IFR then VFR with FAA 7110.308 Simulation Analysis Results 2016 MIFR
Table B.4-2 | Operation Morning IFR then VFR with Wake Recat Simulation Analysis Results 2016 MIFR
2016 MIFR Recat
2018 MIFR
2018 MIFR Recat
2021 MIFR
2021 MIFR Recat
2021 + 5% 2021+ 5% MIFR MIFR Recat
Total Operations Average Arrival Delay Average Departure Delay
23.8
22.3
31.9
29.7
36.4
34.6
49.8
44.0
10.6
7.8
11.8
9.4
16.1
16.3
18.2
19.3
Average Delay
17.2
15.0
21.8
19.5
26.3
25.4
34.1
31.7
66.4
62.2
78.3
76.6
86.1
76.2
103.5
96.0
24.4
21.9
37.5
29.0
55.6
40.0
47.6
52.5
Peak Hour Arrival Delay Peak Hour Departure Delay
B.4.3
1,336
1,369
1,435
1,504
FAA JO 7110.308
Table B.4-3 presents a summary of the simulated delay results of the morning IFR then VFR conditions using the FAA JO 7110.308 procedure. For comparison purposes, the delay results from the MIFR simulation experiments under existing separations is included to show the differences in delays between the various experiments. The simulation results show a significant delay savings at each of the demand levels when operating the FAA JO 7110.308 stagger arrival procedures. The ability to operate two arrival runways during morning IFR conditions increases arrival capacity and reduces delays. In addition, implementation of FAA 7110.308 improves the recovery and transition back to a VFR operation because the procedure reduces extended delays that typically occur under existing morning IFR conditions.
Draft Final: September 2016
Appendix B | Page 13
Ultimate Airport Capacity and Delay Simulation Modeling Analysis
Total Operations Average Arrival Delay Average Departure Delay Average Delay Peak Hour Arrival Delay Peak Hour Departure Delay
B.5
2016 MIFR 7110.308
2018 MIFR
1,336
2018 MIFR 7110.308
2021 MIFR
1,369
2021 2021 + 5% 2021 + 5% MIFR MIFR MIFR 7110.308 7110.308 1,435
1,504
23.8
13.4
31.9
18.8
36.4
24.3
49.8
30.5
10.6
7.3
11.8
7.0
16.1
10.9
18.2
14.5
17.2
10.3
21.8
12.9
26.3
17.6
34.1
22.5
66.4
42.3
78.3
46.4
86.1
66.9
103.5
77.5
24.4
21.7
37.5
18.4
55.6
25.2
47.6
29.8
Annualized Weighted Simulation Delay Results
Weighted average delays were computed using the results of the VFR and morning IFR then VFR (MIFR) simulation results for each of the four simulated demand levels. The delay results in each statistical category were multiplied by the annual estimated percentage in which the Airport operates in each weather condition (approximately 60% VFR, 40% MIFR). Annualizing delays provides a more realistic estimate of the expected delays on an annual basis versus using the statistical results from individual operating conditions (VFR or morning IFR then VFR). Table B.5-1, Table B.5-2, and Table B.5-3 summarize the weighted arrival, departure, and average overall delays based on the VFR and morning IFR simulation results for the present day, Wake Recat, and JO 7110.308 separation standards, respectively. When compared with the present day and Wake Recat separation simulations, operating the FAA JO 7110.308 provides noticeable delay reductions as well as increased capacity in the morning IFR period.
Draft Final: September 2016
Appendix B | Page 14
SFO Airport Development Plan
Ultimate Airport Capacity and Delay Simulation Modeling Analysis
Table B.5-1 | Weighted Average Delay Summary with Present Day Separation
Total Operations Average Arrival Delay Average Departure Delay Average Delay
B.6
2016
2018
2021
2021 + 5%
1,336
1,369
1,435
1,504
12.9
16.4
19.6
26.9
7.3
7.6
10.5
12.3
10.1
12.0
15.1
19.6
Table B.5-2 | Weighted Average Delay Summary with Wake Recat
Total Operations Average Arrival Delay Average Departure Delay Average Delay
2016
2018
2021
2021 + 5%
1,336
1,369
1,435
1,504
12.2
15.4
18.8
23.5
6.7
6.8
10.4
12.6
9.4
11.1
14.5
18.0
Table B.5-3 | Weighted Average Delay Summary with Wake Recat and FAA JO 7110.308
Total Operations Average Arrival Delay Average Departure Delay Average Delay
SFO Airport Development Plan
2016
2018
2021
2021 + 5%
1,336
1,369
1,435
1,504
8.7
11.1
14.6
18.1
6.5
5.9
8.2
10.7
7.5
8.5
11.4
14.3
Ultimate Airport Capacity and Delay Simulation Modeling Analysis
Maximum Practical Capacity Analysis
The airfield/airspace simulation analysis considered not only the maximum theoretical capacity of the existing SFO airfield, but also its maximum practical capacity. Maintaining airline schedule integrity is the primary operational goal of defining maximum practical airfield capacity. While delays during certain peak periods in certain IFR operating conditions or runway uses is expected, high peak hour delays should dissipate in the hours following to avoid excessive cancellations and missed connections. Extended delays throughout the day during predominant operating conditions are not acceptable. Therefore, (1) average daily delay levels, (2) delays in each hour of the day, and (3) percentage of flights delayed were all considered in defining the maximum practical capacity. From the maximum practical capacity analysis, a daily operations level was defined which was used to develop the constrained schedule. Based on the simulation modeling, the existing Airport runway system has a maximum practical capacity between 1,400 and 1,425 daily operations. Certain NextGen procedures could potentially allow for additional throughput with similar or lower aircraft delays. Wake Recat may provide an estimated increase of approximately 20 daily operations. Implementation of the FAA JO 7110.308 procedure could reduce overall delays and increase arrival capacity during morning IFR periods, allowing for an additional 30 operations over the already increased throughput enabled by Wake Recat. The following analysis assumes a design day capacity of 1,475 daily operations. According to the unconstrained operations forecast, the Airport is anticipated to reach this demand level between the years 2021 and 2026. The maximum practical capacity of an airport is the maximum demand an airport can sustain while maintaining an acceptable level of service. The acceptable level of service can vary between airports and is custom defined for each airport. In this analysis, the following metrics were considered at different demand levels to evaluate the level of service: x x
Maximum peak hour arrival and departure delays Time it takes for peak hourly delays to dissipate and return to lower acceptable delay levels
x
Airline operations delayed by more than 15, 30, 45, and 60 minutes
x
Maintaining airline schedule integrity
The four demand levels that were modeled were used in this analysis – 2016, 2018, 2021, and 2021 +5 percent. In this section, each demand level is referred to by its number of daily operations: 1,336, 1,369, 1,435, and 1,504, respectively. At each demand level, both all-day VFR and morning IFR then VFR conditions were modeled. No flight cancellations were taken into account in the morning IFR then VFR simulation experiments and the full flight schedule was operated. In reality, the airlines would cancel some flights to avoid both excessive delays and schedule degradation during inclement weather. For each demand level, weather condition, and separation procedure combination, an analysis was conducted based on two delay metrics: x x
Draft Final: September 2016
Appendix B | Page 15
Delay hours: The number of hours spent with average arrival or departure days exceeding either 15 minutes (VFR) or 30-60 minutes (morning IFR then VFR) Percentage of total operations delayed: The proportion of flights delayed by more than 15 minutes
Draft Final: September 2016
Appendix B | Page 16
SFO Airport Development Plan
B.6.1
Ultimate Airport Capacity and Delay Simulation Modeling Analysis
SFO Airport Development Plan
Ultimate Airport Capacity and Delay Simulation Modeling Analysis
Exhibit B.6-2 | VFR with Present Day Separation Percentage of Total Operations Delayed
VFR with Present Day Separation
Exhibit B.6-1 depicts the simulated hourly delays for the VFR experiments using present day separation criteria. The figure presents both arrival and departure delays for each of the four simulated demand levels. Peak hour delays in excess of 15 minutes do not typically span more than two hours at any given time for arrivals when operating at demand levels between 1,435 and 1,504 daily operations. Average departure delays remain below the 15 minute threshold for all demand levels except the 1,504 demand level; however, the delay dissipates within one hour. Exhibit B.6-1 | VFR with Present Day Separation Delays by Hour
Exhibit B.6-2 depicts the percentage of total operations delayed for each of the four simulated demand levels. At each demand level, a small percentage of flights experience average delays in excess of 15 minutes. When demand levels increase, a larger percentage of flights will experience these higher average delays.
Draft Final: September 2016
Appendix B | Page 17
Draft Final: September 2016
Appendix B | Page 18
SFO Airport Development Plan
B.6.2
Ultimate Airport Capacity and Delay Simulation Modeling Analysis
SFO Airport Development Plan
Ultimate Airport Capacity and Delay Simulation Modeling Analysis
Exhibit B.6-4 | VFR with Wake Recat Percentage of Total Operations Delayed
VFR with Wake Recat
Exhibit B.6-3 depicts the simulated hourly delays for the 28-1 VFR experiment with the Wake Recat separation criteria. Application of the Wake Recat separation criteria provides some overall average peak hour delay reduction in comparison to present day separation criteria (Exhibit B.6.1). Exhibit B.6-3 | VFR with Wake Recat Delays by Hour
Exhibit B.6-4 depicts the percentage of total operations delayed for each of the four simulated demand levels in VFR conditions and with Wake Recat separation criteria applied. Delays when using Wake Recat are slightly lower for arrivals and departures when compared with present day separation criteria (Exhibit B.6-2).
Draft Final: September 2016
Appendix B | Page 19
Draft Final: September 2016
Appendix B | Page 20
SFO Airport Development Plan
B.6.3
Ultimate Airport Capacity and Delay Simulation Modeling Analysis
Morning IFR then VFR with Present Day Separation
SFO Airport Development Plan
Ultimate Airport Capacity and Delay Simulation Modeling Analysis
Exhibit B.6-6 | Morning IFR then VFR with Present Day Separation Percentage of Total Operations Delayed
Exhibit B.6-5 depicts the simulated demand hour delays for the morning IFR then VFR weather condition under present day separation criteria. The morning IFR then VFR experiments were evaluated based on peak delays in excess of 30 minutes and 60 minutes for arrivals and departures, respectively. In each of the four demand levels, the capacity limitations during morning IFR conditions caused cascading arrival delays in excess of 1.5 to 7.0 hours. These cascading delays would degrade airline schedule integrity and result in excessive delays or cancelled flights. As a result of the delayed arrivals during the morning IFR period, the corresponding departure flights for those aircraft would be shifted into the early afternoon/evening period, overlapping with the peak international departure push. This higher departure demand yields higher delays and longer periods of cascading delays. Exhibit B.6-5 | Morning IFR then VFR with Present Day Separation Delays by Hour
Exhibit B.6-6 depicts the percentage of total operations delayed for each of the four simulated demand levels in morning IFR then VFR conditions under current day separation criteria. Exhibit B.6-6 depicts the impact of cascading arrival and departure delays and the resulting higher percentage of aircraft delayed more than 15 minutes.
Draft Final: September 2016
Appendix B | Page 21
Draft Final: September 2016
Appendix B | Page 22
SFO Airport Development Plan
B.6.4
Ultimate Airport Capacity and Delay Simulation Modeling Analysis
SFO Airport Development Plan
Ultimate Airport Capacity and Delay Simulation Modeling Analysis
Exhibit B.6-8 | Morning IFR then VFR with Wake Recat Percentage of Total Operations Delayed
Morning IFR then VFR with Wake Recat
Exhibit B.6-7 depicts the simulated demand hour delays for the morning IFR then VFR weather condition with the application of Wake Recat separation criteria. When compared to the simulation using present day separation criteria (presented in Exhibit B.6-7), Wake Recat slightly reduced arrival delays. Cascading delays in excess of 30 minutes for departures are reduced in all demand levels when compared with present day separation criteria. Exhibit B.6-7 | Morning IFR then VFR with Wake Recat Delays by Hour
Exhibit B.6-8 depicts the percentage of total operations delayed for each of the four demand levels in morning IFR then VFR conditions with the application of Wake Recat separation criteria. The application of Wake Recat provides some delay benefits; however, the results did not indicate significant delay reductions.
Draft Final: September 2016
Appendix B | Page 23
Draft Final: September 2016
Appendix B | Page 24
SFO Airport Development Plan
B.6.5
Ultimate Airport Capacity and Delay Simulation Modeling Analysis
SFO Airport Development Plan
Ultimate Airport Capacity and Delay Simulation Modeling Analysis
Exhibit B.6-10 | Morning IFR then VFR with FAA JO 7110.308 Percentage of Total Operations Delayed
Morning IFR then VFR with FAA JO 7110.308
Exhibit B.6-9 depicts the simulated demand hour delays for the morning IFR then VFR weather condition with the application of the FAA JO 7110.308 arrival procedure. The ability to operate two ILS approaches with staggered arrivals on Runways 28L and 28R during IFR conditions reduces arrival delays and increases throughput. As a result, the application of this procedure provides significant peak arrival and departure delay improvements when compared to the present day and Wake Recat separation criteria. Because arrival delays do not build as significantly during the morning IFR period, corresponding departing flights are not shifted into the early afternoon period and so there is no overlap with the international departure push. This scenario results in lower peak departure delays when the airport transitions back to a VFR operation. Exhibit B.6-9 | Morning IFR then VFR Operation with FAA JO 7110.308 Delays by Hour
Exhibit B.6-10 depicts the percentage of total operations delayed for each of the four demand levels in morning IFR then VFR conditions with the application of the FAA JO 7110.308 arrival procedure. The procedure provides significant delay reduction benefits and reduces the percentage of total operations delayed 15 minutes or more.
Draft Final: September 2016
Appendix B | Page 25
Draft Final: September 2016
Appendix B | Page 26
SFO Airport Development Plan
B.7
Ultimate Airport Capacity and Delay Simulation Modeling Analysis
Conclusion
The ultimate capacity and delay simulation modeling analysis provides the Airport and the airlines with guidance for determining the maximum practical capacity of the existing runway system when operating at higher demand levels. The results of the simulation analysis indicate that the maximum practical capacity for the Airport is 1,475 daily operations. It is estimated that the Airport operates in VFR conditions 60 percent of the time and IFR conditions 40 percent of the time annually. With such a high percentage of IFR operating conditions on an annual basis, increasing capacity and reducing delays is imperative to reaching the maximum practical capacity of the existing runway system at the Airport. Key NextGen airspace improvements, especially the FAA JO 7110.308 arrival procedure, would need to be implemented to reduce the average arrival delays during morning IFR periods, which will increase capacity of the Airfield to its maximum practical capacity. Key findings from the simulation modeling analysis include: x
The application of the FAA JO 7110.308 arrival procedure during morning IFR periods significantly reduces delays so that they do not cascade past when the Airport transitions back to a VFR operation.
x
Wake Recat separation criteria provides some benefit in IFR conditions for reduction of delays. However, due to the dependent nature of the arrival and departure operation and the increased arrival separation required to sequence departing flights, Wake Recat does not provide significant gains in capacity and delay reductions.
Draft Final: September 2016
Appendix B | Page 27
ULTIMATE AIRPORT CAPACITY & DELAY SIMULATION MODELING ANALYSIS B TABLE OF CONTENTS B.1
Introduction ................................................................................ 1
B.2
Simulation Modeling Assumption and Methodology ........ 4 B.2.1
B.3
B.4
Runway Usage .................................................................................. 4
B.2.2
Airfield and Terminal Enhancements .......................................... 5
B.2.3
Demand Levels ................................................................................ 5
B.2.4
Weather Conditions ....................................................................... 6
B.2.5
FAA Separation Initiatives ............................................................ 7
B.2.6
Statistical Evaluation Metrics ....................................................... 9
VFR Simulation Experiment Results and Analysis ............ 10 B.3.1
Present Day Separation ............................................................... 10
B.3.2
Wake Recat ..................................................................................... 10
Morning IFR then VFR Simulation Experiment Results and Analysis ..................................................................................... 12
EXHIBITS
TABLES
Exhibit B.1-1 | Existing Airfield Configuration ......................................... 3
Table B.1-1 | Existing Runway Lengths ...................................................... 2
Exhibit B.2-1 | Standard 28-1 Runway Configuration ............................. 4
Table B.2-1 | Design Day Flight Schedule Fleet Mix Summary ............. 7
Exhibit B.2-2 | Forecast Flight Schedule Rolling Hour Demand Comparison .......................................................................... 6
Table B.2-2 | Forecast B757 and B767 Aircraft Fleet Demand Levels .. 8
Exhibit B.2-3 | FAA 7110.308 Dependent Approaches Parallel Runways Spaced Less Than 2,500 feet Apart Application ........................................................................... 8 Exhibit B.6-1 | VFR with Present Day Separation Delays by Hour ...... 17 Exhibit B.6-2 | VFR with Present Day Separation Percentage of Total Operations Delayed................................................ 18 Exhibit B.6-3 | VFR with Wake Recat Delays by Hour ............................ 19 Exhibit B.6-4 | VFR with Wake Recat Percentage of Total Operations Delayed .......................................................... 20 Exhibit B.6-5 | Morning IFR then VFR with Present Day Separation Delays by Hour................................................................... 21 Exhibit B.6-6 | Morning IFR then VFR with Present Day Separation Percentage of Total Operations Delayed ..................... 22
B.4.1
Present Day Separation ............................................................... 12
B.4.2
Wake Recat ..................................................................................... 12
B.4.3
FAA JO 7110.308 ........................................................................... 13
Exhibit B.6-8 | Morning IFR then VFR with Wake Recat Percentage of Total Operations Delayed ............................................ 24
B.5
Annualized Weighted Simulation Delay Results .............. 14
B.6
Maximum Practical Capacity Analysis ................................ 16
Exhibit B.6-9 | Morning IFR then VFR Operation with FAA JO 7110.308 Delays by Hour ................................................. 25
B.7
B.6.1
VFR with Present Day Separation.............................................. 17
B.6.2
VFR with Wake Recat.................................................................... 19
B.6.3
Morning IFR then VFR with Present Day Separation ............. 21
B.6.4
Morning IFR then VFR with Wake Recat................................... 23
B.6.5
Morning IFR then VFR with FAA JO 7110.308 ......................... 25
Table B.3-1 | VFR Simulation Analysis Results ........................................ 10 Table B.3-2 | VFR Simulation with Wake Recat Analysis Results......... 11 Table B.4-1 | Morning iFR then VFR Simulation Analysis Results ....... 12 Table B.4-2 | Operation Morning IFR then VFR with Wake Recat Simulation Analysis Results ............................................... 13 Table B.4-3 | Morning IFR then VFR with FAA 7110.308 Simulation Analysis Results .................................................................... 14 Table B.5-1 | Weighted Average Delay Summary with Present Day Separation ............................................................................. 15 Table B.5-2 | Weighted Average Delay Summary with Wake Recat .. 15 Table B.5-3 | Weighted Average Delay Summary with Wake Recat And FAA JO 7110.308 .......................................................... 15
Exhibit B.6-7 | Morning IFR then VFR with Wake Recat Delays by Hour ..................................................................................... 23
Exhibit B.6-10 | Morning IFR then VFR with FAA JO 7110.308 Percentage of Total Operations Delayed ................... 26
Conclusion ................................................................................ 27
AIRPORT DEVELOPMENT PLAN – DRAFT FINAL
Ultimate Airport Capacity & Delay Simulation Modeling Analysis | B - i
SFO Airport Development Plan
Ultimate Airport Capacity and Delay Simulation Modeling Analysis
Appendix B Ultimate Airport Capacity and Delay Simulation Modeling Analysis B.1
Introduction
SFO Airport Development Plan
Ultimate Airport Capacity and Delay Simulation Modeling Analysis
Exhibit B.1-1 depicts the existing runway configuration for SFO. Table B.1-1 provides the dimensions for each runway. The ultimate capacity and delay simulation analysis focuses primarily on the maximum capacity of the standard 28-11 runway operating configuration. In this configuration, arriving flights use Runways 28L and 28R while departing flights use Runways 1L and 1R. Alternate runway configurations were not evaluated in this analysis as the 28-1 plan and the closely related 28-28 plan are used over 95 percent of the time. Table B.1-1 | Existing Runway Lengths
Evaluating the ultimate capacity of San Francisco International Airport (SFO) is necessary to understand the Airport’s growth potential. In order to evaluate the ultimate capacity of the Airport, the SFO Bureau of Planning and Environmental Affairs (BPEA) tasked Landrum & Brown Inc. (L&B) with using the FAA’s Airport and Airspace Simulation Model (SIMMOD) simulation software to evaluate the ultimate capacity and delays of the current runway system at SFO. In addition, various operational configurations were evaluated at increased demand levels to determine impacts on capacity and delays. Due to constraints which limit the ability for runway expansion at the Airport, future growth and success depends on new procedures and methods that increase capacity while using the existing airfield configuration. These were considered in the development of the model simulation.
Designation
Length
Width
1L-19R
7,500’
200’
1R-19L
8,646’
200’
10L-28R
11,381’
200’
10R-28L
11,870’
200’
The following sections of this report provide a description of the simulation modeling experiments including: x An overview of the simulation modeling methodology and assumptions x Simulated weather and operating configurations o
Visual flight rules (VFR) weather conditions – pilots visually navigate and maintain separation from clouds and other aircraft
o
Morning instrument flight rules (MIFR) weather conditions – visibility less than 3 statute miles and ceiling less than 1,000 feet above ground level
x Explanation of various simulated runway operating configurations and airspace configurations which could potentially increase runway capacity at SFO o
FAA Wake Turbulence Recategorization (Wake Recat)
o
FAA Job Order (JO) 7110.308 1.5 nautical mile Dependent Approaches to Parallel Runways Spaced Less Than 2,500 feet Apart arrival procedures on Runways 28L and 28R
x Evaluation of the maximum practical capacity of the airfield
The results of the simulation analyses are intended to provide SFO and airport stakeholders with a realistic assessment of the maximum theoretical capacity of the Airport. The maximum capacity assessment in this report can be used to guide the Airport in planning future flight demand levels, thereby ensuring flight schedule integrity and minimizing delays.
1 Runway
operation patterns at SFO are abbreviated using the code “x-y,” where x represents the primary arrival runway pair and y represents the primary departure runway pair.
Draft Final: September 2016
Appendix B | Page 1
Draft Final: September 2016
Appendix B | Page 2
SFO Airport Development Plan
B.2
Ultimate Airport Capacity and Delay Simulation Modeling Analysis
Simulation Modeling Assumption and Methodology
This section provides an overview of the methodology and assumptions that were used in the simulation modeling analyses.
B.2.1 SFO Airport Development Plan
Ultimate Airport Capacity and Delay Simulation Modeling Analysis
Exhibit B.1-1 | Existing Airfield Configuration
Runway Usage
As stated in Section B.1, the standard 28-1 operation was the primary runway configuration simulated in this analysis. In VFR conditions, the standard runway usage consists of two paired arrivals on Runways 28R and 28L and two departures on Runways 1L and 1R between each arrival pair. Departures that cannot use Runways 1L or 1R due to aircraft performance requirements use Runways 28L or 28R instead. Following a 2013 extension of Runway 10R-28L, the majority of long-haul international departures use this runway. In the standard 28-1 runway operation, the majority of general aviation and cargo departures use Runways 1L and 1R for departures. Exhibit B.2-1 depicts the standard 28-1 runway usage configuration which was evaluated in the simulation modeling analysis. Exhibit B.2-1 | Standard 28-1 Runway Configuration
Source: SFO Bureau of Planning and Environmental Affairs, SFO Airport Layout Plan, 2014; Landrum & Brown, Inc., 2016
Draft Final: September 2016
Appendix B | Page 3
Draft Final: September 2016
Appendix B | Page 4
SFO Airport Development Plan
B.2.2
Ultimate Airport Capacity and Delay Simulation Modeling Analysis
SFO Airport Development Plan
Ultimate Airport Capacity and Delay Simulation Modeling Analysis
Exhibit B.2-2 | Forecast Flight Schedule Rolling Hour Demand Comparison
Airfield and Terminal Enhancements
The future configuration of B/A B was simulated with 18 gates, which does not account for multiple aircraft ramp system (MARS) gates in the new boarding area which will allow additional aircraft to park at some gates. Existing airline gating assignments were used, and those aircraft which currently park at B/A B were assigned to the reconfigured boarding area.
B.2.3
Demand Levels
Previous SIMMOD analyses conducted for the Runway Safety Area (RSA) improvement project simulated demand levels up to 1,300 daily operations, which represented a 2014 forecast demand level. During peak summer months, the Airport currently operates demand levels that occasionally exceed 1,300 daily operations. To evaluate the impacts of increased aircraft operations on the capacity of the airspace and runway system at SFO, several forecast aircraft demand levels were simulated. As part of the update to the Airport’s ultimate forecast, L&B prepared forecast flight schedules for 2016, 2018, and 2021 demand levels. The simulated operations in the flight schedule are as follows: x
2016 demand level – 1,336 daily operations
x
2018 demand level – 1,369 daily operations
x
2021 demand level – 1,435 daily operations
In addition to these three forecast demand levels, a 5 percent flight schedule increase of the 2021 demand level was evaluated. This demand level simulated approximately 1,504 daily operations. Military and helicopter operations which were identified in the original design day flight schedules (DDFS) were not simulated in the analysis.
The rolling hour comparison of the 2016, 2018, and 2021 demand levels indicates the following peak hour demand rates anticipated at SFO:
Exhibit B.2-2 depicts a rolling hour comparison based on 10 minute time intervals (bins) of the current day 2013 demand levels as well as the three forecast demand levels which were simulated for this analysis.
x
2016 (1,336 daily operations): 55 arrivals, 58 departures, 105 total operations
x
2018 (1,369 daily operations): 55 arrivals, 58 departures, 104 total operations
x
2021 (1,435 daily operations): 59 arrivals, 62 departures, 110 total operations
The arrival, departure, and total operations peaks occur at different times of day, so the peak hour totals do not equal the sum of arrivals and departures.
B.2.4
Weather Conditions
Each demand level was evaluated under VFR conditions and morning IFR then VFR (MIFR) conditions. The MIFR condition is the result of a typical weather pattern where a layer of cool marine air draws fog through the Golden Gate and down the coast of the Bay. Usually this lasts until 10:30 am when the sun begins to evaporate the fog and visibility improves. The Airport is reduced to a single arrival runway (Runway 28R) when operating under IFR conditions. With only one primary arrival runway available, Airport capacity decreases, resulting in a greater number of delays.
Draft Final: September 2016
Appendix B | Page 5
Draft Final: September 2016
Appendix B | Page 6
SFO Airport Development Plan
B.2.5
Ultimate Airport Capacity and Delay Simulation Modeling Analysis
SFO Airport Development Plan
Ultimate Airport Capacity and Delay Simulation Modeling Analysis
Table B.2-2 | Forecast B757 and B767 Aircraft Fleet Demand Levels
FAA Separation Initiatives
FAA Wake Recat As part of its NextGen air traffic modernization initiatives, the Federal Aviation Administration (FAA) is reevaluating aircraft wake turbulence categorizations in a program known as “Wake Recat.” Analysis of certain aircraft fleets has determined that existing wake turbulence separation requirements for lead-to-trail aircraft combinations may be excessive during IFR operations. Wake Recat increases runway capacity by reducing the lead-to-trail aircraft separations for various aircraft types.
B757 B767 Total Flights
B757/B767 Operations in Forecast Flight Schedules 2016 2018 20 10 21 19 41 29
2021 0 13 13
FAA JO 7110.308 In an effort to increase arrival capacity during IFR conditions at SFO, the FAA JO 7110.308 1.5 Nautical Mile Dependent Approaches to Parallel Runways Spaced Less Than 2,500 feet Apart arrival procedure was simulated. During IFR operations, the Airport is ordinarily reduced to a single arrival runway which reduces capacity to 30-32 operations per hour. The FAA 7110.308 procedure allows ILS approaches to be operated simultaneously on Runways 28L and 28R using a staggered arrival procedure.
Table B.2-1 provides a summary of the DDFS fleet mix by Wake Recat category. Table B.2-1 | Design Day Flight Schedule Fleet Mix Summary DDFS Fleet Mix (% Of Total Aircraft Operations) Wake Category
Common A/C
2016
2018
2021
CAT A
A380
0.4%
0.6%
0.7%
CAT B
B747, B777, A340
7.7%
7.9%
8.6%
CAT C
B767, A310, MD11
1.6%
1.4%
0.9%
CAT D
B757, B737, A320
61.5%
62.0%
62.5%
CAT E
CRJ2, CRJ7, E170
19.2%
20.8%
21.4%
CAT F
E120, E190, H25B
9.7%
7.4%
6.0%
Total
100.0%
100.0%
100.0%
Daily ops
1,336
1,369
1,435
Exhibit B.2-3 depicts the basic guidelines for the application of the FAA JO 7110.308 procedure when conducted on Runways 28L and 28R. Exhibit B.2-3 | FAA 7110.308 Dependent Approaches Parallel Runways Spaced Less Than 2,500 feet Apart Application
Wake Recat primarily benefits the B757 and B767 aircraft fleets by allowing reduced in-trail separation requirements in IFR conditions. As proposed in Wake Recat, B757 aircraft are grouped with other narrowbody aircraft including the B737 and A320. The B767 is removed from the “heavy” aircraft category which contains aircraft including the B747 and B777 and placed in its own category with new in-trail separation requirements. Table B.2-2 provides a summary of the forecast B757 and B767 aircraft demand levels in the 2016, 2018, and 2021 flight schedules. The B757 is estimated to be phased out of SFO operations by the year 2021 and replaced with more efficient narrowbody aircraft such as the B737-900ER. Usage of the B767 is also expected to decline in future years, and that aircraft type will be replaced with new models such as the A350 and B787.
Draft Final: September 2016
Appendix B | Page 7
Draft Final: September 2016
Appendix B | Page 8
SFO Airport Development Plan
Ultimate Airport Capacity and Delay Simulation Modeling Analysis
The FAA JO 7110.308 arrival procedure requires the lead aircraft in a staggered pair to use the lower approach to the near end of a runway or the lower glide slope angle. Since the runway thresholds for Runways 28L and 28R are abeam each other, Runway 28L is considered the lower approach as a result of the 2.85 degree glide slope angle for instrument approaches on the runway. Narrowbody and small aircraft can always be the lead aircraft in a staggered pair. However, heavy and super heavy category aircraft are required to be the trailing aircraft to mitigate the impacts of wake turbulence on the paired aircraft. For simulation purposes, heavy and super heavy aircraft were assigned to Runway 28R. Rules dictate that arrivals can be diagonally staggered using a minimum of 1.5 nm of separation. However, given the complexity of implementing the staggered operation at SFO, this operation will most likely not achieve the 1.5 nm separation on a consistent basis during the initial implementation. It is estimated that staggered separations would range between 1.5 nm and 2.0 nm; therefore, a separation of 1.8 nm was used in the simulation.
B.2.6
SFO Airport Development Plan
B.3
VFR Simulation Experiment Results and Analysis
This section presents the results of the two VFR simulation experiments, one operating under the present day VFR separation criteria and other under the Wake Recat separation criteria. Although Wake Recat is intended primarily to reduce in-trail separations during IFR operations, this analysis simulated it in a VFR environment to determine if any delay improvements or capacity gains could be achieved. FAA JO 7110.308 was not simulated, as this procedure affects only IFR operations.
B.3.1
Present Day Separation
Table B.3-1 presents a summary of some of the standard delay evaluation metrics from the simulation model. The delays in each of these categories increase when operating at higher demand levels. Table B.3-1 | VFR Simulation Analysis Results
Statistical Evaluation Metrics
The simulation analysis used the following statistical metrics in evaluating the final results of each simulation experiment: x Average arrival delays x Average departure delays x Average overall delays
Total Operations Average Arrival Delay Average Departure Delay Average Delay
x Peak hour arrival delays
Peak Hour Arrival Delay Peak Hour Departure Delay
x Peak hour departure delays x Weighted average delays x Demand hour delay profiles for VFR and morning IFR simulations x Percentage of total operations delayed by more than 15, 30, 45, and 60 minutes The following sections of this report will discuss the statistical analysis results of the various simulation experiments. The results of the simulation analysis are used to estimate the maximum practical capacity and acceptable levels of delay for the Airport which is the primary focus of this analysis.
Ultimate Airport Capacity and Delay Simulation Modeling Analysis
B.3.2
2016 VFR
2018 VFR
2021 VFR
2021 + 5% VFR
1,336
1,369
1,435
1,504
5.7
6.0
8.4
9.8
5.1
4.8
6.8
8.1
5.4
5.4
7.6
8.9
11.9
12.6
17.9
23
10.7
7.9
11.6
20.3
Wake Recat
Table B.3-2 represents a summary of the simulated delay results of the Wake Recat simulation experiments operating in VFR conditions. For comparison purposes, the delay results from the simulation experiments under present day VFR separations are included to show the differences in delays. In general, the application of Wake Recat provides minimal delay savings under VFR conditions. The primary reason is that lead-to-trail aircraft in-trail VFR separations at SFO are typically larger than minimum IFR separations because departures from Runways 1L and 1R are sequenced in between arrivals on Runways 28L and 28R. A previous analysis of SFO radar data indicates that in VFR conditions, the minimum in-trail separation between narrowbody arrival pairs is approximately 3.8 nm, which is considered the minimum required to sequence departures from Runways 1L and 1R in between arrivals. Due to the dependent nature of arrivals and departures at SFO, Wake Recat may only yield minimal benefits in delay reductions and capacity gains. In general, Wake Recat would not provide significant benefit to SFO under VFR operating conditions.
Draft Final: September 2016
Appendix B | Page 9
Draft Final: September 2016
Appendix B | Page 10
SFO Airport Development Plan
Ultimate Airport Capacity and Delay Simulation Modeling Analysis
Table B.3-2 | VFR Simulation with Wake Recat Analysis Results 2016 VFR
Total Operations Average Arrival Delay Average Departure Delay Average Delay Peak Hour Arrival Delay Peak Hour Departure Delay
2016 VFR Recat
2018 VFR
1,336
2018 VFR Recat
SFO Airport Development Plan
B.4
2021 VFR
1,369
2021 VFR Recat
2021 + 5% 2021 + 5% VFR VFR Recat
1,435
1,504
5.7
5.5
6.0
5.9
8.4
8.2
11.6
9.8
5.1
5.9
4.8
5.1
6.8
6.4
8.4
8.1
5.4
5.7
5.4
5.5
7.6
7.3
10.0
8.9
11.9
9.4
12.6
13.7
17.9
16.1
22.4
23.0
10.7
12.4
7.9
9.6
11.6
14.6
19.2
20.3
Ultimate Airport Capacity and Delay Simulation Modeling Analysis
Morning IFR then VFR Simulation Experiment Results and Analysis
This section presents the results of the morning IFR then VFR (MIFR) simulation experiments. This weather pattern was simulated using each of the four demand levels and both of the FAA air traffic procedure improvements.
B.4.1
Present Day Separation
The morning IFR period typically occurs during the morning peak arrival push and ends around 10:30 am. With present day separation requirements, the Airport is reduced to a single arrival runway during the morning IFR period, which reduces arrival capacity to approximately 30-32 operations per hour. For simulation purposes, dual departures on Runways 1L and 1R were sequenced between the single arrival stream on Runway 28R. When the morning IFR period ends, simultaneous paired arrivals commence on Runways 28L and 28R. Table B.4-1 presents a comparison of the simulated delay results for the four demand levels of the MIFR modeling experiments. Increased demand levels when operating in IFR conditions result in significantly higher delays, particularly for arrivals. Late arrivals typically shift the peak departure banks to the mid to late afternoon periods, overlapping with the peak international departure push and thereby resulting in higher than normal departure delays. Table B.4-1 | Morning IFR then VFR Simulation Analysis Results
Total Operations Average Arrival Delay Average Departure Delay Average Delay Peak Hour Arrival Delay Peak Hour Departure Delay
B.4.2
2016 MIFR
2018 MIFR
2021 MIFR
2021 + 5% MIFR
1,336
1,369
1,435
1,504
23.8
31.9
36.4
49.8
10.6
11.8
16.1
18.2
17.2
21.8
26.3
34.1
66.4
78.3
86.1
103.5
24.4
37.5
55.6
47.6
Wake Recat
Table B.4-2 presents a summary of the simulated delay results of the Wake Recat simulation experiments operating in morning IFR then VFR conditions. For comparison purposes, the delay results from the MIFR simulation experiments under present day separations is included to show the differences in delays between the various experiments.
Draft Final: September 2016
Appendix B | Page 11
Draft Final: September 2016
Appendix B | Page 12
SFO Airport Development Plan
Ultimate Airport Capacity and Delay Simulation Modeling Analysis
Under Wake Recat separation criteria, significant gains are achieved with B757 and B767 aircraft, resulting in noticeable arrival delay improvements. However, reduced arrival in-trail separation results in fewer gaps for departures on intersecting Runways 1L and 1R, thus resulting in higher departure delays.
SFO Airport Development Plan
Table B.4-3 | Morning IFR then VFR with FAA 7110.308 Simulation Analysis Results 2016 MIFR
Table B.4-2 | Operation Morning IFR then VFR with Wake Recat Simulation Analysis Results 2016 MIFR
2016 MIFR Recat
2018 MIFR
2018 MIFR Recat
2021 MIFR
2021 MIFR Recat
2021 + 5% 2021+ 5% MIFR MIFR Recat
Total Operations Average Arrival Delay Average Departure Delay
23.8
22.3
31.9
29.7
36.4
34.6
49.8
44.0
10.6
7.8
11.8
9.4
16.1
16.3
18.2
19.3
Average Delay
17.2
15.0
21.8
19.5
26.3
25.4
34.1
31.7
66.4
62.2
78.3
76.6
86.1
76.2
103.5
96.0
24.4
21.9
37.5
29.0
55.6
40.0
47.6
52.5
Peak Hour Arrival Delay Peak Hour Departure Delay
B.4.3
1,336
1,369
1,435
1,504
FAA JO 7110.308
Table B.4-3 presents a summary of the simulated delay results of the morning IFR then VFR conditions using the FAA JO 7110.308 procedure. For comparison purposes, the delay results from the MIFR simulation experiments under existing separations is included to show the differences in delays between the various experiments. The simulation results show a significant delay savings at each of the demand levels when operating the FAA JO 7110.308 stagger arrival procedures. The ability to operate two arrival runways during morning IFR conditions increases arrival capacity and reduces delays. In addition, implementation of FAA 7110.308 improves the recovery and transition back to a VFR operation because the procedure reduces extended delays that typically occur under existing morning IFR conditions.
Draft Final: September 2016
Appendix B | Page 13
Ultimate Airport Capacity and Delay Simulation Modeling Analysis
Total Operations Average Arrival Delay Average Departure Delay Average Delay Peak Hour Arrival Delay Peak Hour Departure Delay
B.5
2016 MIFR 7110.308
2018 MIFR
1,336
2018 MIFR 7110.308
2021 MIFR
1,369
2021 2021 + 5% 2021 + 5% MIFR MIFR MIFR 7110.308 7110.308 1,435
1,504
23.8
13.4
31.9
18.8
36.4
24.3
49.8
30.5
10.6
7.3
11.8
7.0
16.1
10.9
18.2
14.5
17.2
10.3
21.8
12.9
26.3
17.6
34.1
22.5
66.4
42.3
78.3
46.4
86.1
66.9
103.5
77.5
24.4
21.7
37.5
18.4
55.6
25.2
47.6
29.8
Annualized Weighted Simulation Delay Results
Weighted average delays were computed using the results of the VFR and morning IFR then VFR (MIFR) simulation results for each of the four simulated demand levels. The delay results in each statistical category were multiplied by the annual estimated percentage in which the Airport operates in each weather condition (approximately 60% VFR, 40% MIFR). Annualizing delays provides a more realistic estimate of the expected delays on an annual basis versus using the statistical results from individual operating conditions (VFR or morning IFR then VFR). Table B.5-1, Table B.5-2, and Table B.5-3 summarize the weighted arrival, departure, and average overall delays based on the VFR and morning IFR simulation results for the present day, Wake Recat, and JO 7110.308 separation standards, respectively. When compared with the present day and Wake Recat separation simulations, operating the FAA JO 7110.308 provides noticeable delay reductions as well as increased capacity in the morning IFR period.
Draft Final: September 2016
Appendix B | Page 14
SFO Airport Development Plan
Ultimate Airport Capacity and Delay Simulation Modeling Analysis
Table B.5-1 | Weighted Average Delay Summary with Present Day Separation
Total Operations Average Arrival Delay Average Departure Delay Average Delay
B.6
2016
2018
2021
2021 + 5%
1,336
1,369
1,435
1,504
12.9
16.4
19.6
26.9
7.3
7.6
10.5
12.3
10.1
12.0
15.1
19.6
Table B.5-2 | Weighted Average Delay Summary with Wake Recat
Total Operations Average Arrival Delay Average Departure Delay Average Delay
2016
2018
2021
2021 + 5%
1,336
1,369
1,435
1,504
12.2
15.4
18.8
23.5
6.7
6.8
10.4
12.6
9.4
11.1
14.5
18.0
Table B.5-3 | Weighted Average Delay Summary with Wake Recat and FAA JO 7110.308
Total Operations Average Arrival Delay Average Departure Delay Average Delay
SFO Airport Development Plan
2016
2018
2021
2021 + 5%
1,336
1,369
1,435
1,504
8.7
11.1
14.6
18.1
6.5
5.9
8.2
10.7
7.5
8.5
11.4
14.3
Ultimate Airport Capacity and Delay Simulation Modeling Analysis
Maximum Practical Capacity Analysis
The airfield/airspace simulation analysis considered not only the maximum theoretical capacity of the existing SFO airfield, but also its maximum practical capacity. Maintaining airline schedule integrity is the primary operational goal of defining maximum practical airfield capacity. While delays during certain peak periods in certain IFR operating conditions or runway uses is expected, high peak hour delays should dissipate in the hours following to avoid excessive cancellations and missed connections. Extended delays throughout the day during predominant operating conditions are not acceptable. Therefore, (1) average daily delay levels, (2) delays in each hour of the day, and (3) percentage of flights delayed were all considered in defining the maximum practical capacity. From the maximum practical capacity analysis, a daily operations level was defined which was used to develop the constrained schedule. Based on the simulation modeling, the existing Airport runway system has a maximum practical capacity between 1,400 and 1,425 daily operations. Certain NextGen procedures could potentially allow for additional throughput with similar or lower aircraft delays. Wake Recat may provide an estimated increase of approximately 20 daily operations. Implementation of the FAA JO 7110.308 procedure could reduce overall delays and increase arrival capacity during morning IFR periods, allowing for an additional 30 operations over the already increased throughput enabled by Wake Recat. The following analysis assumes a design day capacity of 1,475 daily operations. According to the unconstrained operations forecast, the Airport is anticipated to reach this demand level between the years 2021 and 2026. The maximum practical capacity of an airport is the maximum demand an airport can sustain while maintaining an acceptable level of service. The acceptable level of service can vary between airports and is custom defined for each airport. In this analysis, the following metrics were considered at different demand levels to evaluate the level of service: x x
Maximum peak hour arrival and departure delays Time it takes for peak hourly delays to dissipate and return to lower acceptable delay levels
x
Airline operations delayed by more than 15, 30, 45, and 60 minutes
x
Maintaining airline schedule integrity
The four demand levels that were modeled were used in this analysis – 2016, 2018, 2021, and 2021 +5 percent. In this section, each demand level is referred to by its number of daily operations: 1,336, 1,369, 1,435, and 1,504, respectively. At each demand level, both all-day VFR and morning IFR then VFR conditions were modeled. No flight cancellations were taken into account in the morning IFR then VFR simulation experiments and the full flight schedule was operated. In reality, the airlines would cancel some flights to avoid both excessive delays and schedule degradation during inclement weather. For each demand level, weather condition, and separation procedure combination, an analysis was conducted based on two delay metrics: x x
Draft Final: September 2016
Appendix B | Page 15
Delay hours: The number of hours spent with average arrival or departure days exceeding either 15 minutes (VFR) or 30-60 minutes (morning IFR then VFR) Percentage of total operations delayed: The proportion of flights delayed by more than 15 minutes
Draft Final: September 2016
Appendix B | Page 16
SFO Airport Development Plan
B.6.1
Ultimate Airport Capacity and Delay Simulation Modeling Analysis
SFO Airport Development Plan
Ultimate Airport Capacity and Delay Simulation Modeling Analysis
Exhibit B.6-2 | VFR with Present Day Separation Percentage of Total Operations Delayed
VFR with Present Day Separation
Exhibit B.6-1 depicts the simulated hourly delays for the VFR experiments using present day separation criteria. The figure presents both arrival and departure delays for each of the four simulated demand levels. Peak hour delays in excess of 15 minutes do not typically span more than two hours at any given time for arrivals when operating at demand levels between 1,435 and 1,504 daily operations. Average departure delays remain below the 15 minute threshold for all demand levels except the 1,504 demand level; however, the delay dissipates within one hour. Exhibit B.6-1 | VFR with Present Day Separation Delays by Hour
Exhibit B.6-2 depicts the percentage of total operations delayed for each of the four simulated demand levels. At each demand level, a small percentage of flights experience average delays in excess of 15 minutes. When demand levels increase, a larger percentage of flights will experience these higher average delays.
Draft Final: September 2016
Appendix B | Page 17
Draft Final: September 2016
Appendix B | Page 18
SFO Airport Development Plan
B.6.2
Ultimate Airport Capacity and Delay Simulation Modeling Analysis
SFO Airport Development Plan
Ultimate Airport Capacity and Delay Simulation Modeling Analysis
Exhibit B.6-4 | VFR with Wake Recat Percentage of Total Operations Delayed
VFR with Wake Recat
Exhibit B.6-3 depicts the simulated hourly delays for the 28-1 VFR experiment with the Wake Recat separation criteria. Application of the Wake Recat separation criteria provides some overall average peak hour delay reduction in comparison to present day separation criteria (Exhibit B.6.1). Exhibit B.6-3 | VFR with Wake Recat Delays by Hour
Exhibit B.6-4 depicts the percentage of total operations delayed for each of the four simulated demand levels in VFR conditions and with Wake Recat separation criteria applied. Delays when using Wake Recat are slightly lower for arrivals and departures when compared with present day separation criteria (Exhibit B.6-2).
Draft Final: September 2016
Appendix B | Page 19
Draft Final: September 2016
Appendix B | Page 20
SFO Airport Development Plan
B.6.3
Ultimate Airport Capacity and Delay Simulation Modeling Analysis
Morning IFR then VFR with Present Day Separation
SFO Airport Development Plan
Ultimate Airport Capacity and Delay Simulation Modeling Analysis
Exhibit B.6-6 | Morning IFR then VFR with Present Day Separation Percentage of Total Operations Delayed
Exhibit B.6-5 depicts the simulated demand hour delays for the morning IFR then VFR weather condition under present day separation criteria. The morning IFR then VFR experiments were evaluated based on peak delays in excess of 30 minutes and 60 minutes for arrivals and departures, respectively. In each of the four demand levels, the capacity limitations during morning IFR conditions caused cascading arrival delays in excess of 1.5 to 7.0 hours. These cascading delays would degrade airline schedule integrity and result in excessive delays or cancelled flights. As a result of the delayed arrivals during the morning IFR period, the corresponding departure flights for those aircraft would be shifted into the early afternoon/evening period, overlapping with the peak international departure push. This higher departure demand yields higher delays and longer periods of cascading delays. Exhibit B.6-5 | Morning IFR then VFR with Present Day Separation Delays by Hour
Exhibit B.6-6 depicts the percentage of total operations delayed for each of the four simulated demand levels in morning IFR then VFR conditions under current day separation criteria. Exhibit B.6-6 depicts the impact of cascading arrival and departure delays and the resulting higher percentage of aircraft delayed more than 15 minutes.
Draft Final: September 2016
Appendix B | Page 21
Draft Final: September 2016
Appendix B | Page 22
SFO Airport Development Plan
B.6.4
Ultimate Airport Capacity and Delay Simulation Modeling Analysis
SFO Airport Development Plan
Ultimate Airport Capacity and Delay Simulation Modeling Analysis
Exhibit B.6-8 | Morning IFR then VFR with Wake Recat Percentage of Total Operations Delayed
Morning IFR then VFR with Wake Recat
Exhibit B.6-7 depicts the simulated demand hour delays for the morning IFR then VFR weather condition with the application of Wake Recat separation criteria. When compared to the simulation using present day separation criteria (presented in Exhibit B.6-7), Wake Recat slightly reduced arrival delays. Cascading delays in excess of 30 minutes for departures are reduced in all demand levels when compared with present day separation criteria. Exhibit B.6-7 | Morning IFR then VFR with Wake Recat Delays by Hour
Exhibit B.6-8 depicts the percentage of total operations delayed for each of the four demand levels in morning IFR then VFR conditions with the application of Wake Recat separation criteria. The application of Wake Recat provides some delay benefits; however, the results did not indicate significant delay reductions.
Draft Final: September 2016
Appendix B | Page 23
Draft Final: September 2016
Appendix B | Page 24
SFO Airport Development Plan
B.6.5
Ultimate Airport Capacity and Delay Simulation Modeling Analysis
SFO Airport Development Plan
Ultimate Airport Capacity and Delay Simulation Modeling Analysis
Exhibit B.6-10 | Morning IFR then VFR with FAA JO 7110.308 Percentage of Total Operations Delayed
Morning IFR then VFR with FAA JO 7110.308
Exhibit B.6-9 depicts the simulated demand hour delays for the morning IFR then VFR weather condition with the application of the FAA JO 7110.308 arrival procedure. The ability to operate two ILS approaches with staggered arrivals on Runways 28L and 28R during IFR conditions reduces arrival delays and increases throughput. As a result, the application of this procedure provides significant peak arrival and departure delay improvements when compared to the present day and Wake Recat separation criteria. Because arrival delays do not build as significantly during the morning IFR period, corresponding departing flights are not shifted into the early afternoon period and so there is no overlap with the international departure push. This scenario results in lower peak departure delays when the airport transitions back to a VFR operation. Exhibit B.6-9 | Morning IFR then VFR Operation with FAA JO 7110.308 Delays by Hour
Exhibit B.6-10 depicts the percentage of total operations delayed for each of the four demand levels in morning IFR then VFR conditions with the application of the FAA JO 7110.308 arrival procedure. The procedure provides significant delay reduction benefits and reduces the percentage of total operations delayed 15 minutes or more.
Draft Final: September 2016
Appendix B | Page 25
Draft Final: September 2016
Appendix B | Page 26
SFO Airport Development Plan
B.7
Ultimate Airport Capacity and Delay Simulation Modeling Analysis
Conclusion
The ultimate capacity and delay simulation modeling analysis provides the Airport and the airlines with guidance for determining the maximum practical capacity of the existing runway system when operating at higher demand levels. The results of the simulation analysis indicate that the maximum practical capacity for the Airport is 1,475 daily operations. It is estimated that the Airport operates in VFR conditions 60 percent of the time and IFR conditions 40 percent of the time annually. With such a high percentage of IFR operating conditions on an annual basis, increasing capacity and reducing delays is imperative to reaching the maximum practical capacity of the existing runway system at the Airport. Key NextGen airspace improvements, especially the FAA JO 7110.308 arrival procedure, would need to be implemented to reduce the average arrival delays during morning IFR periods, which will increase capacity of the Airfield to its maximum practical capacity. Key findings from the simulation modeling analysis include: x
The application of the FAA JO 7110.308 arrival procedure during morning IFR periods significantly reduces delays so that they do not cascade past when the Airport transitions back to a VFR operation.
x
Wake Recat separation criteria provides some benefit in IFR conditions for reduction of delays. However, due to the dependent nature of the arrival and departure operation and the increased arrival separation required to sequence departing flights, Wake Recat does not provide significant gains in capacity and delay reductions.
Draft Final: September 2016
Appendix B | Page 27
