Internet: www.gap-projekt.de
Contact: [email protected]
GERMAN AIRPORT
PERFORMANCE
GERMAN AVIATION
BENCHMARKING
partner/sponsor:
Simulation and SIMMOD Validation of Independent
Parallel Runway Operations with Regard to Delay - An
Example of the new Airport BER
Branko Bubalo
GAP Research Project
European SIMMOD User Group (ESUG) Meeting
Berlin, October 28th 2011
ESUG Meeting Berlin 28.10.2011 Branko Bubalo Page 1
GERMAN AIRPORT
PERFORMANCE
GERMAN AVIATION
BENCHMARKING
Personal Background
Graduate in Business Administration and Engineering from Berlin School of Economics and Law and University of Applied Sciences Berlin
Diploma Thesis (2009): Benchmarking Airport Productivity and the Role of Capacity Utilization
5th year in German Airport Performance (GAP) Research Project
PhD student a the University of Hamburg under supervision of Stefan Voss
Page 2ESUG Meeting Berlin 28.10.2011 Branko Bubalo
GERMAN AIRPORT
PERFORMANCE
GERMAN AVIATION
BENCHMARKING
Introduction
Externalities of transportation, such as noise or delays, are increasingly important to the passenger and for the society, therefore
Measuring the productivity of airports should include level of service, i.e. delay.
FAA AC 150/5060-5 - Airport Capacity And Delay points to SIMMOD
Earlier work from my thesis includes a web-crawler, which currently collects scheduled and actual flight times and aircraft types for about
100 European airports on a daily basis.
-> Peak day flight schedules from a number of airports are at hand for
conducting studies
Obtained a free copy of AirportTools VisualSIMMOD and the SIMMOD engine from the FAA
Page 3ESUG Meeting Berlin 28.10.2011 Branko Bubalo
GERMAN AIRPORT
PERFORMANCE
GERMAN AVIATION
BENCHMARKING
Motivation for this Study
Challenge for BER: Existing airports (SXF and TXL) will be replaced fully
by Single Airport Berlin-Brandenburg International (BER) in June 2012
Aim 1: Looking for a tool for political decision-making and evaluation of official planning figures, in this case the planning approval documents
(Planfeststellungsbeschluss) of BER airport.
Aim 2: Estimating the maximum capacity of a particular runway system, e.g. an independent parallel runway as build for BER, with regard to
delay and associated costs of delay.
Aim 3: Conducting a traffic forecast for BER, to answer when the maximum capacity is likely to be reached, under different assumptions
regarding traffic growth rates and future aircraft mix. The official
planning documents reveal that the maximum capacity will be reached
in 2023.
Page 4ESUG Meeting Berlin 28.10.2011 Branko Bubalo
GERMAN AIRPORT
PERFORMANCE
GERMAN AVIATION
BENCHMARKING
Model
Keep it simple approach:
1. Basic airport (runway and taxiway system) and nearby airspace (final
approach) design of the simulation model.
2. Only 1 representative gate, having the total parking stand capacity
3. Coordinates and background satellite pictures
come from Google Earth
4. Usage of AIP information
Page 5ESUG Meeting Berlin 28.10.2011 Branko Bubalo
GERMAN AIRPORT
PERFORMANCE
GERMAN AVIATION
BENCHMARKING
Model cont.
5. First-Come-First-Served (FCFS)
6. Independent parallel runway operated in segregated mode
7. One operational direction, East-to-West, Runway 25R for landings,
Runway 25L for take-offs.
8. Only final approach and departure paths, reaching approx. 15NM
to/from the runway has been modelled as airspace
Main observed output: Level-of-Service Indicator Average delay per flight (in minutes), from cumulative waiting times at the different queues, such as departure queue, gate waiting and holding airspace.
Page 6ESUG Meeting Berlin 28.10.2011 Branko Bubalo
GERMAN AIRPORT
PERFORMANCE
GERMAN AVIATION
BENCHMARKING
Simulated Scenarios and Traffic Mixes
Scenario 0 Scenario I Scenario II Scenario III
Airport SXF TXL BER BER BER BERTurbulence Class Share Flights Share Flights Share Flights Share Flights Share Flights Share FlightsHEAVY 0% 3% 12 2% 12 5% 32 15% 95 5% 32LARGE 100% 155 95% 454 96% 609 95% 603 80% 508 84% 533SMALL 0% 3% 14 2% 14 0% 0 5% 32 11% 70Sum 100% 155 100% 480 100% 635 100% 635 100% 635 100% 635
Mix Index (MI) 100% 102% 102% 110% 125% 99%MI = 3 x (% HEAVY) + (% LARGE)
Page 7ESUG Meeting Berlin 28.10.2011 Branko Bubalo
GERMAN AIRPORT
PERFORMANCE
GERMAN AVIATION
BENCHMARKING
Methodology
Inputs for the baseline model are the combined busy day flight schedules of Tegel and Schoenefeld airports, which consist of
scheduled and actual gate-on/gate-off times.
Baseline aircraft mix is taken from the schedules
Cumulative distribution functions for turnaround times by aircraft category, HEAVY, LARGE, SMALL, are taken from the schedules, and
have been assigned as gate blocking time to arriving aircraft at the gate.
Random lateness of between 0 and 4 minutes is applied to flight injection times to disperse the flights, especially needed around the
peaks - no early arrivals have been modelled, i.e. no delay < 0 min.
Since arrivals are injected in the airspace, the gate-on times needs to be adjusted by uncongested landing time plus taxiing times to the gate.
SIMMODs cloning function has been used for the simulation of the growth scenarios: 11 growth scenarios (-20%,0%,20%,40%,,200%) at 10 iterations each
Page 8ESUG Meeting Berlin 28.10.2011 Branko Bubalo
GERMAN AIRPORT
PERFORMANCE
GERMAN AVIATION
BENCHMARKING
Presenting Interim Results for Scenario 0
Simulated maximum throughput Capacity of 82-84 Flights per hour at 120-140% growth above baseline, but
Recommended Practical long-term capacity of 76 Flights per hour! (1 operation every 47 seconds) The animation proves this.
Demand higher than Capacity already at 60% growth, based on a Level-of-Service of below 6 minutes average delay per flight
Best fit curve has the function:
Avg. Delay per Flight (in min) = 0.0906e0.0572 x peak hourly flights
Queuing delays increase exponentially beyond 60% growth.
Upper and lower boundaries of historic growth rates at TXL and SXF underline the assumption that demand=capacity reached between
2018 and 2026 (at 3 to 6% annual growth).
Each 20% growth increment results in approximately doubling of daily delays.
Change in Traffic Mix in other Scenarios has only a minor impact on maximum throughput and delay, sheer growth is more critical
Page 9ESUG Meeting Berlin 28.10.2011 Branko Bubalo
GERMAN AIRPORT
PERFORMANCE
GERMAN AVIATION
BENCHMARKING
Results: Simulated Delays at BER:
Scenario 0 (*Note: 42 average cost per minute of delay (Eurocontrol 2009: Standard Inputs for CBA Analyses)
Growth from Baseline
Daily Flights
Peak Hour Demand
Hourly Capacity
Mean Delay per Flight
Daily Delay Minutes
Daily Delay Costs at 42 * (without cancellation costs) Cancellations
-20% 511 40 40 1.1 543 EUR 22,806 0
0% 635 48 48 1.4 887 EUR 37,254 020% 758 55 54 2.3 1760 EUR 73,920 040% 886 69 71 3.7 3287 EUR 138,054 060% 1012 78 76 5.9 5955 EUR 250,110 080% 1145 94 80 11.5 13223 EUR 555,366 0
100% 1270 90 82 21.2 26968 EUR 1,132,656 1120% 1400 96 82 26.8 37501 EUR 1,575,042 134140% 1517 98 84 27.4 41538 EUR 1,744,596 440160% 1639 110 83 58.2 95364 EUR 4,005,288 807
Page 10ESUG Meeting Berlin 28.10.2011 Branko Bubalo
GERMAN AIRPORT
PERFORMANCE
GERMAN AVIATION
BENCHMARKING
Results Scenario 0: Simulated Peak Throughput at BER
Not taking into account the delays per flight!
48
54
71
7680 82 82
84 83
48
55
69
78
9490
96 98
110
0
10
20
30
40
50
60
70
80
90
100
110
120
Scenario 0 0% Scenario 0
20%
Scenario 0
40%
Scenario 0
60%
Scenario 0
80%
Scenario 0
100%
Scenario 0
120%
Scenario 0
140%
Scenario 0
160%
De
man
d a
nd
Cap
acit
y in
Flig
hts
pe
r h
ou
r
Projected Growth of Traffic at Berlin-Brandenburg Airport in Comparison to Runway Capacity (Scenario 0:"Nullszenario")
Hourly Capacity
Hourly Demand
Maximum Capacity SIMMOD
Maximum Capacity Planning Permission 2023
Page 11ESUG Meeting Berlin 28.10.2011 Branko Bubalo
Demand > Capacity
GERMAN AIRPORT
PERFORMANCE
GERMAN AVIATION
BENCHMARKING
Results Scenario 0: Demand and Capacity over time of day
Page 12
0
20
40
60
80
100
120
00:0
0 -
01:0
0
01:0
0 -
02:0
0
02:0
0 -
03:0
0
03:0
0 -
04:0
0
04:0
0 -
05:0
0
05:0
0 -
06:0
0
06:0
0 -
07:0
0
07:0
0 -
08:0
0
08:0
0 -
09:0
0
09:0
0 -
10:0
0
10:0
0 -
11:0
0
11:0
0 -
12:0
0
12:0
0 -
13:0
0
13:0
0 -
14:0
0
14:0
0 -
15:0
0
15:0
0 -
16:0
0
16:0
0 -
17:0
0
17:0
0 -
18:0
0
18:0
0 -
19:0
0
19:0
0 -
20:0
0
20:0
0 -
21:0
0
21:0
0 -
22:0
0
22:0
0 -
23:0
0
23:0
0 -
24:0
0
24:0
0 -
25:0
0
25:0
0 -
26:0
0
26:0
0 -
27:0
0
Flig
hts
pe
r ho
ur
Time of day
Scenario 0 Growth 0% - Demand Scenario 0 Growth 20% - Demand Scenario 0 Growth 40% - DemandScenario 0 Growth 60% - Demand Scenario 0 Growth 80% - Demand Scenario 0 Growth 100% - DemandScenario 0 Growth 120% - Demand Scenario 0 Growth 140% - Demand Scenario 0 Growth 160% - DemandScenario 0 Growth 0% - Capacity Scenario 0 Growth 20% - Capacity Scenario 0 Growth 40% - CapacityScenario 0 Growth 60% - Capacity Scenario 0 Growth 80% - Capacity Scenario 0 Growth 100% - CapacityScenario 0 Growth 120% - Capacity Scenario 0 Growth 140% - Capacity Scenario 0 Growth 160% - Capacity
ESUG Meeting Berlin 28.10.2011 Branko Bubalo
GERMAN AIRPORT
PERFORMANCE
GERMAN AVIATION
BENCHMARKING
Results: Simulated Daily and Peak hourly Delays at BER
1.1 1.42.3
3.7
5.9
11.5
21.2
26.8 27.4
Avg Delay = 0.1624e0.0036x
R = 0.9883
Avg. Delay = 0.0906e0.0572x
R = 0.96340 20 40 60 80 100 120 140
0
5
10
15
20
25
30
35
0 200 400 600 800 1000 1200 1400 1600 1800 2000
Peak Hourly Flights
Ave
rage
De
lay
pe
r Fl
igh
t
Daily Flights
Average Delay per Flight from Simulation of BBI Baseline Scenario 0
Average Delay per Daily Flights
Average Delay per Peak Hour Flights
Expon. (Average Delay per Daily Flights)
Expon. (Average Delay per Peak Hour Flights)
sustainable
Not sustainable
Page 13
GERMAN AIRPORT
PERFORMANCE
GERMAN AVIATION
BENCHMARKING
Capacity limits reached at BER
The practical capacity of 76 Flights per Hour and 60% growth in daily traffic can be reached after:
16 years at 3% average growth of traffic (2010-2026)
12 years at 4% average growth of traffic (2010-2022)
10 years at 5% average growth of traffic (2010-2020)
8 years at 6% average growth of traffic (2010-2018)
The discussion on a 3rd runway at BER should be started in 2016 the latest, as a result from Scenario 0 (3% growth), if 10 years planning and
construction time until realisation are considered.
Legal, political and environmental approval time will increase further in the future, so do the opportunity costs from building a runway too late.
%591)06.01(1)1( 8nr
%631)05.01(1)1( 10nr
%601)04.01(1)1( 12nr
%601)03.01(1)1( 16nr
Page 14
GERMAN AIRPORT
PERFORMANCE
GERMAN AVIATION
BENCHMARKING
Capacity limits reached at BER
Page 15ESUG Meeting Berlin 28.10.2011 Branko Bubalo
GERMAN AIRPORT
PERFORMANCE
GERMAN AVIATION
BENCHMARKING
European Comparisons
Peer group of independent, far parallel-runway airports (Separation of at least 1500 meters)
Best Practices with comparable runway layout and highest productivity (2008) in comparison to BER (SXF + TXL): LHR and MUC
BER has the potential to rank third-biggest European airport in peer group (before Palma-de-Mallorca (PMI) and Oslo (OSL)) by opening in 2012
Airport Annual PAX (million)
Annual Flights (thousand)
PAX per Flight
Peak Hour Throughput
London-Heathrow 67.2 473 142 103
Munich 34.5 409 84 93
BER (TXL+SXF 2008) 21.2 212 100 48
BER Planned Final 2023 (Planning Permission document from 2004)
30 301-359 100-91 90
BER Practical Final at 2024 (60% growth from 2008 at 3%; This study)
28.1 303 93 76
Page 16ESUG Meeting Berlin 28.10.2011 Branko Bubalo
GERMAN AIRPORT
PERFORMANCE
GERMAN AVIATION
BENCHMARKING
Validation of the simulation model (hourly flights and delay)
Page 17
0.0
5.0
10.0
15.0
20.0
25.0
30.0
35.0
0 10 20 30 40 50 60 70 80 90 100
Ave
rage
De
lay
pe
r S
che
du
led
Flig
ht
in M
inu
tes
Scheduled Flights per Hour
0.0
5.0
10.0
15.0
20.0
25.0
30.0
35.0
0 10 20 30 40 50 60 70 80 90 100
Ave
rage
De
lay
pe
r Sc
he
du
led
Flig
ht
in M
inu
tes
Scheduled Flights per Hour
MUC
LHR
OSL
PMI
ESUG Meeting Berlin 28.10.2011 Branko Bubalo
Observations from LHR, MUC, OSL, PMI airports from May to June 2011
Lower bound envelopes most likely onlyfrom Congestion Delays
Output from SIMMOD simulationof independent parallel runway in
Segregated mode with BER schedule
GERMAN AIRPORT
PERFORMANCE
GERMAN AVIATION
BENCHMARKING
How far should we think for the future of BER?
Airport planners must be creative and realistic in their forecasts and long term views, even if politically unpopular.
Runways take 10 years of planning, approval and construction time (economical life over 40 years)
The future is uncertain, but market outlooks predict further growth of air transportation, in line with growth of GDP, in many developing parts in
the world. Europe is a main connector between Asia and the Americas.
We can learn from other airports already:
London Heathrow in 2010: saturated most of the day
Munich in 2010: congested during daily peak hours
Possible near Future for BER? ->
Page 18ESUG Meeting Berlin 28.10.2011 Branko Bubalo
GERMAN AIRPORT
PERFORMANCE
GERMAN AVIATION
BENCHMARKING
Extension Plan for London Heathrow
Strong public opposition against 3rd runway, plans have been scrapped
Environmental and Legal battle would have be endless with uncertain outcome
Page 19ESUG Meeting Berlin 28.10.2011 Branko Bubalo
GERMAN AIRPORT
PERFORMANCE
GERMAN AVIATION
BENCHMARKING
Current Plans for Munich (2020?)
Munich 2 had been restricted from the beginning.
3rd runway is strongly opposed by the surrounding communitites, but will most likely be built
GARS Workshop at ILA Berlin 10.06.2010 Branko Bubalo Page 20
GERMAN AIRPORT
PERFORMANCE
GERMAN AVIATION
BENCHMARKING
Vision
Commitment and Continuing Discussion towards aviation needs.
Carefully conduct cost-benefit analyses (CBA), which also look at the secondary and welfare effects of air transportation and connectivity
Neighbours of airports should participate in future developments, even if 20 years or more in advance.
Master plans must present long-term forecasts, adjustable to fluctuating traffic growth.
Runways are long-term investments, which are typically depreciated over 40 to 50 years, so long-sighted visions needed.
Apart from looking at runway capacity, environmental capacity, e.g. caps on amount of noise, could become more critical in the future.
Simulation is the tool of choice to transparently communicate airport expansion plans.
GARS Workshop at ILA Berlin 10.06.2010 Branko Bubalo Page 21
GERMAN AIRPORT
PERFORMANCE
GERMAN AVIATION
BENCHMARKING
Thank you for your attention! Questions?
Suggestions and Comments are welcome.
www.gap-projekt.de
Page 22ESUG Meeting Berlin 28.10.2011 Branko Bubalo