Post on 02-Apr-2020
Nicole Adler, Hebrew University
Eran Hanany, Tel Aviv University
Stef Proost, KU Leuven
Managing Change in Air Traffic Control:
a Regional Forerunner Approach
Motivation present ATC system in EU is composed of 37 national providers
compared to FAA, EU system is 34% more costly (2011)
barriers to cost efficiency:
ownership form: governmental organizations
fragmentation: missing economies of scale
protectionism: power of Labor Unions & member states
weak regulation: failure to implement FABs or strict price-caps
barriers to increasing capacity:
SESAR estimated cost: 30 billion euro
relatively low congestion currently
opposition to change / fear of technology
ACCHANGE: Adler, Hanany & Proost
how could cost efficiency and technology
adoption be encouraged simultaneously?
ACCHANGE: Adler, Hanany & Proost
changes in ownership form
horizontal integration
vertical integration
privatization
changes in pricing regulation
strict individual price-caps
peak / off peak charges
hybrid price-caps
changes in capacity
Pilot Common Project
SESAR Step 1
Outline of talk Literature search
Methodology to analyse ATM market: 2-stage game 1st stage: ANSPs 2nd stage: airline response
Case study Western Europe covering 6 ANSP areas Data collected
Scenarios analysed Base case Pricing Technology adoption: PCP and SESAR Step 1 Horizontal mergers: FABs Vertical mergers: Regional forerunner
Conclusions & Future Research
ACCHANGE: Adler, Hanany & Proost
Literature Search
Air Traffic Control On the US system:
Morrison & Winston (2008)
Nextor (2010)
On the European system: Grushka-Cockayne, De Reyck and Degraeve (2008)
Castelli, Labbe and Violin (2013)
Jovanovic, Tosic, Cangalovic, Stanojevic (2014)
Button and Neiva (2014)
Network congestion games Rosenthal (1973) Monderer & Shapley (1996) Application:
Road networks: de Palma & Nesterov (1998), Lindsey & Verhoef (2000)
Computer networks: Roughgarden (2009)
ACCHANGE: Adler, Hanany & Proost
2 Stage Game
ACCHANGE: Adler, Hanany & Proost
Stage 1:
ANSPs set peak/off-peak charges according to regulatory rules
ATC terminal set charges and limit flights in peak (slot allocation) Cost recovery
Profit maximization
Peak/Off peak price-caps
Stage 2 - Airlines choose flight paths given schedules
3 cost components: operational, congestion & ATC en-route charges
+ Revenue loss: flying off-peak lowers airfares
Option to ‘not fly’ necessary for demand elasticity
Note: Congestion is non-linear
Closer to capacity: the higher the delays
2 Stage Game
ACCHANGE: Adler, Hanany & Proost
Stage 1: ANSPs set peak/off-peak charges according to regulatory rules ATC terminal set charges and limit flights in peak (slot allocation)
Cost recovery
Profit maximization
Peak/Off peak price-caps
Stage 2 - Airlines choose flight paths given schedules 3 cost components: operational, congestion & ATC en-route charges + Revenue loss: flying off-peak lowers airfares Option to ‘not fly’ necessary for demand elasticity
User optimal: Airlines set flight paths to minimize own costs
Minimize only self-inflicted delays
System optimal: central planner set flight paths to minimize sum of airline costs Bird’s eye view: Minimize total sum of all delay
Case Study of Western Europe
ACCHANGE: Adler, Hanany & Proost
5
1
3
6
2
4
7
8
9
10
13
16 15
11
14
18
23
24
25
26
12
19
17
21 20 22
34
29
28
33
35
32
27
30
31
Case Study of Western Europe
ACCHANGE: Adler, Hanany & Proost
5
1
3
6
2
4
7
8
9
10
13
16 15
11
14
18
23
24
25
26
12
19
17
21 20 22
34
29
28
33
35
32
27
30
31
Players
6 Air navigation service providers: (~50% of EU traffic)
NATS (U.K.) LVNL (Netherlands) DFS (Germany) AENA (Spain) Belgocontrol (Belgium) DSNA (France)
5 Airlines: 3 alliances:
Star (Lufthansa)
Oneworld (BA)
SkyTeam (AF-KLM)
Low cost carrier (EasyJet) Unaligned carrier (Emirates)
• 9 ATC terminal providers:
• Large airports (slot constrained):
• AMS
• BRU
• CDG
• FRA
• LHR
• MAD
• Small airports (no slots):
• BER
• MAN
• PMI
ACCHANGE: Adler, Hanany & Proost
• ANSP
• ATC Terminal by country
Key Data Source: Eurocontrol. (2013). ATM cost-effectiveness Benchmarking Reports
ANSP Revenues
(000 €)
Variable Costs
(000 €)
Fixed Costs
(000 €)
Total Distance in Charging Area
(Km)
Avg. Charge per km
DSNA 1,167,138 804,653 113,876 1,463,618,011 0.80
DFS 739,112 631,983 129,285 1,007,485,777 0.73
Aena 794,710 498,756 135,599 859,175,623 0.93
NATS 651,366 368,015 153,001 707,474,135 0.92
LVNL 169,365 102,058 11,378 191,563,198 0.88
Belgocontrol 155,805 82,605 13,884 166,751,138 0.93
Country IFR airport movements
controlled by ANSP Fixed Cost (€) Variable Cost per Movement (€)
Income From Charges per
Movement (€)
Germany 2,059,372 41,208,000 86 101
France 1,892,868 31,704,000 127 122
Spain 1,854,896 49,253,000 117 116
UK 1,746,362 9,863,000 87 115
Netherlands 485,525 5,313,000 99 113
Belgium 380,572 9,208,000 130 69
ACCHANGE: Adler, Hanany & Proost
Cases
Base case: reproduce 2011
Pricing: peak/off-peak price, price regulation vs. no regulation
Technology: pilot common project & SESAR step 1
Horizontal Integration: FABs
Belgocontrol & DSNA
LVNL & DFS
DSNA & DFS
Belgocontrol, LVNL & DFS (MUAC lower airspace)
Vertical Integration: Regional Forerunner
Germany: DFS, FRA, BER and Lufthansa
France: DSNA, CDG & Air France
Spain: AENA, MAD, PMI & British Airways/Iberia
ACCHANGE: Adler, Hanany & Proost
Base case: reproducing current equilibria 2011
User optimal (cost recovery)
0.053
0.054
0.073
0.088
0.073
-150,000,000
-100,000,000
-50,000,000
-
50,000,000
100,000,000
150,000,000
200,000,000
250,000,000
300,000,000
ANSP Profits in €
-40,000,000
-30,000,000
-20,000,000
-10,000,000
-
10,000,000
20,000,000
30,000,000
LHR AMS FRA BRU CDG MAD MAN BER PMI
ATC Terminal Profits in €
BA/IB 6,891,443,879
LH 7,615,174,955
AF/KLM 4,288,508,980
Low Cost Carrier
11,371,448,750
Emirates 7,799,266,681
AIRLINE COSTS IN €
0.053
0.054 0.073
0.088
0.073
System Optimal
Conclusions:
Possible to reorganize flight paths to reduce congestion
Some airlines worse off thus unlikely to agree to central planner approach
0.053
0.054
0.073
0.088
0.073
-0.70%
-0.60%
-0.50%
-0.40%
-0.30%
-0.20%
-0.10%
0.00%
0.10%
0.20%
0.30%
0.40%
BA LH AF LC Rest
-0.05%
-0.48%
-0.61%
0.30%
0.01%
% Change in Airline Costs
-5.00%
0.00%
5.00%
10.00%
15.00%
20.00%
25.00%
30.00%
35.00%
40.00%
NATS UK LVNLNetherlands
DFSGermany
Belgocontrol DSNA France AENA Spain
0.00% 0.00% 0.19%
36.00%
-1.00% -1.01%
% Change in ANSP Profits
ACCHANGE: Adler, Hanany & Proost
Pricing scenarios ANSP profit maximization
Consequences compared to base-run: ANSP prices by average magnitude of 10 Airline CASKs doubled LH stops flying off-peak and cuts down operations during peak by half Airlines cut down operation during peak, only REST business as usual Different peak, off-peak prices Insufficient competition to remove price regulation in current market
0
1
2
3
4
5
6
7
8
9
10
NATS UK LVNL Netherlands DFS Germany Belgocontrol DSNA France AENA Spain
ANSP Charge per Km
Base Case Profit Max
ACCHANGE: Adler, Hanany & Proost
Peak versus off-peak pricing
ANSP Base price Peak-price
(120%) Off-peak price
(80%)
NATS UK 0.921 1.1052 0.7368
LVNL
Netherlands 0.884 1.0608 0.7072
DFS Germany 0.734 0.8808 0.5872
Belgocontrol 0.934 1.1208 0.7472
DSNA France 0.797 0.9564 0.6376
AENA Spain 0.925 1.11 0.74 0.00%
0.50%
1.00%
1.50%
2.00%
BA LH AF LC Rest
1.27%
1.05%
1.38%
1.95% 1.99%
% Change in Airline Costs
• Single price-cap: ANSPs charge maximum
• 2 price-caps: ANSP revenues equivalent or increase
• Off-peak price: low (negative) to induce airline changes
• Revenue loss:
• greater than ANSP charges hence no impact on airlines
• half revenue loss parameter not impact results
ACCHANGE: Adler, Hanany & Proost
Technology: Pilot Common Project
PCP cost = € 2.5 billion
Airlines (~16% of cost): congestion: reduction of 8.7% en-route, 12.51% at airport
operational costs: reduction 0.63% (technology less fuel)
ANSP (~65% of cost) variable cost: reduction of 8.4%
fixed cost: increase of 22%
constant price
ATC Terminal (~5% of cost): airport capacity: increase of 4%
variable cost: reduction of 8.4%
fixed cost: increase of 104%
constant price
ACCHANGE: Adler, Hanany & Proost
Technology: Pilot Common Project
Results: airlines ANSPs ATC terminal
-40.0%
-30.0%
-20.0%
-10.0%
0.0%
10.0%
20.0%
30.0%
40.0%
1.913%
31.379%
1.558%
20.281% 17.290%
-38.963%
% Change in ANSP Profits
-1.20%
-0.80%
-0.40%
0.00%
BA LH AF LC Rest
-0.72% -0.78% -0.79%
-0.90% -0.87%
% Change in Airline Costs
ACCHANGE: Adler, Hanany & Proost
Technology: SESAR Step 1
Step 1 cost = € 30 billion
Airlines (~50% of cost): congestion: reduction of 27% en-route,14% at airport
variable costs: increase of 0.1%
ANSP (~16% of cost): variable cost: reduction of 8.4%
fixed cost: increase of 64%
price: reduction of 6.1%
ATC Terminal (~13% of cost): airport capacity: increase of 14%
variable cost: reduction of 8.4%
fixed cost: increase of 600%
price: reduction of 6.1%
ACCHANGE: Adler, Hanany & Proost
Technology: SESAR Step 1, 2030
Results: Airlines ANSPs ATC
Conclusions: based on expected demand growth of 38.7%
ANSP & ATC prices insufficient
-9.00%
-8.00%
-7.00%
-6.00%
-5.00%
-4.00%
-3.00%
-2.00%
-1.00%
0.00%
BA LH AF LC Rest
-1.77%
-8.91%
-2.46% -2.58% -1.92%
% Change in Airline Costs
-200,000,000
-100,000,000
-
100,000,000
200,000,000
300,000,000
400,000,000
500,000,000
NATS UK LVNLNetherlands
DFS Germany Belgocontrol DSNA France AENA Spain
ANSP Profits: Base Case vs. SESAR Step 1
Base Case SESAR Step 1
ACCHANGE: Adler, Hanany & Proost
Technology: SESAR Step 1 2030 + Price Cap
• ATC terminal: worthwhile
• Conclusions:
• 20% increase in peak for both ANSP and ATC terminal sufficient
• Airlines: wipes out all savings
0.00%
1.00%
2.00%
3.00%
4.00%
5.00%
6.00%
7.00%
8.00%
9.00%
BA LH AF LC Rest
0.75%
8.31%
0.76% 1.44% 1.68%
% Change in Airline Costs
-200,000,000
-100,000,000
-
100,000,000
200,000,000
300,000,000
400,000,000
500,000,000
600,000,000
NATS UK LVNLNetherlands
DFSGermany
Belgocontrol DSNAFrance
AENA Spain
ANSP Profits
SESAR step 1 Price-cap
ACCHANGE: Adler, Hanany & Proost
Horizontal Integration: Functional Airspace Blocks
Horizontal Integration: FABs = 30% fixed cost savings
250,000,000
300,000,000
350,000,000
separately FAB
Belgocontrol & DSNA France: Profit
-80,000,000
-60,000,000
-40,000,000
-20,000,000
-
separately FAB
LVNL Netherlands & DFS Germany: Profit -0.060%
-0.040%
-0.020%
0.000%
0.020%
0.040%
0.060%
BA LH AF LC Rest
0.011%
0.042% 0.049%
-0.045%
0.046%
% Change in Airline Costs
-0.30%
-0.20%
-0.10%
0.00%
0.10%
BA LH AF LC Rest
-0.02%
0.03%
-0.12%
0.05%
-0.27%
% Change in Airline Costs
0.00%
0.20%
0.40%
0.60%
BA LH AF LC Rest
0.28% 0.30%
0.05%
0.38%
0.57%
% Change in Airline Costs
-
100,000,000
200,000,000
300,000,000
separately FAB
DSNA France & DFS Germany: Profit
Horizontal Integration: FABs
MUAC lower airspace FAB:
With lowest price (Germany) – MUAC lower airspace FAB
-0.25%
-0.20%
-0.15%
-0.10%
-0.05%
0.00%
0.05%
BA LH AF LC Rest
-0.05%
0.04%
-0.04% -0.03%
-0.21%
% Change in Airline Costs
-60,000,000
-40,000,000
-20,000,000
-
separately FAB
DFS Germany, Belgocontrol & LVNL Netherlands: Profit
-0.45%
-0.40%
-0.35%
-0.30%
-0.25%
-0.20%
-0.15%
-0.10%
-0.05%
0.00%
BA LH AF LC Rest
-0.10%
-0.05%
-0.24%
-0.17%
-0.41% -56,000,000
-54,000,000
-52,000,000
-50,000,000
-48,000,000
-46,000,000
-44,000,000
-42,000,000
separately FAB
Vertical Integration: Regional Forerunner
ANSP cost recovery & PCP implementation
German Coalition:
French Coalition:
Spanish Coalition:
Member Before After
Lufthansa (Cost) 7,615,174,955 7,529,739,840 DFS -111,244,998 -109,512,293 FRA 19,871,750 23,998,701 BER 492,550 -1,985,179
Member Before After
AF/KLM (Cost) 4,288,508,980 4,216,128,108 DSNA 265,366,417 311,248,918 CDG -19,397,050 -17,132,471
Member Before After
BA/Iberia (Cost) 6,891,443,879 6,841,060,627 AENA 19,932,172 12,165,925 MAD -12,088,200 -17,202,306 PMI -5,043,880 -7,465,139
ACCHANGE: Adler, Hanany & Proost
Conclusions
Modeling air traffic control via network congestion game enables cost-benefit analysis
Ownership form regional forerunner approach preferable to functional airspace blocks
but dependent on partnership
commercialization appears to be necessary to change behavior e.g. through time dependent tenders
Pricing approach insufficient competition to ensure ANSPs charge reasonable prices
peak & off-peak price caps necessary to separate charges
New technology (capacity) pilot common project is positive for airlines but unlikely to succeed without
permitting ANSPs to charge higher prices
SESAR step 1 even more unlikely to occur without re-balancing of charges &/or subsidies
hybrid price-caps necessary to balance cost & benefits of SESAR technologies
ACCHANGE: Adler, Hanany & Proost