measuring airline networks
play

Measuring Airline Networks Chantal Roucolle (ENAC-DEVI) Joint work - PowerPoint PPT Presentation

Measuring Airline Networks Chantal Roucolle (ENAC-DEVI) Joint work with Miguel Urdanoz (TBS) and Tatiana Seregina (ENAC-TBS) This research was possible thanks to the financial support of the Regional Council of Midi Pyrenees. Airline networks


  1. Measuring Airline Networks Chantal Roucolle (ENAC-DEVI) Joint work with Miguel Urdanoz (TBS) and Tatiana Seregina (ENAC-TBS) This research was possible thanks to the financial support of the Regional Council of Midi Pyrenees.

  2. Airline networks  Airline networks are complex and dynamic  Number of Airports served  Number of markets served  Direct or connecting flights, frequencies, schedules… 2

  3. US airports 3

  4. US domestic network in July 2005 4

  5. US domestic network in July 2010 5

  6. US domestic network in July 2013 6

  7. Airline networks  Airline networks are complex and dynamic  Number of Airports served  Number of markets served  Direct or connecting flights, frequencies, schedules…  Airlines make different choices 7

  8. Frontier Airlines, July 2005 8

  9. Southwest Airlines, July 2005 9

  10. Airline networks  Airline networks are complex and dynamic  Number of Airports served  Number of markets served  Direct or connecting flights, frequencies, schedules…  Airlines make different choices and their decisions evolve over time 10

  11. Evolution: Delta Airlines July 2005 11

  12. Evolution: Delta Airlines July 2010 12

  13. Evolution: Delta Airlines July 2015 13

  14. Airline network  Airline networks are complex and dynamic  Number of Airports served  Number of markets served  Frequencies, schedules…  Airlines make different choices and their decisions evolve over time  Two questions to address:  Network characterization  Network evolution: what are the drivers of the choice?  Usefulness of network analysis:  Does the network structure affect costs, prices, profitability, delays…? 14

  15. Literature  In most of the cases perfect hub-and-spoke networks (left) are compared with fully connected networks (right),  For instance Brueckner (2004), Alderighi et al. (2005), Barla and Constantatos (2005), Flores Fillol (2009) or Silva et al. (2014).  However reality is more complex  Wojahn (2001) studies whether a mixed model can be preferred to minimize costs  We want to get closer to this reality 15

  16. Our objective  Step 1: Network characterization  Our approach: combine Graph theory and Principal Component Analysis (PCA) Graph theory: set of mathematical measures and tools to study networks  Already used for airlines:  Wandelt and Sun (2015), Dunn and Wilkinson (2016) or Du et al. (2016) study different network • properties focusing on the country level. Burghouwt and Redondi (2013) present a compilation of connectivity indicators for airports built from • graph theory. Lordan et al.(2016) study resilience with a sample of airlines from Europe, North America and China. • PCA aims to explain most of the information of the dataset through a reduced number of  new variables, called principal components, calculated as linear combinations of the original variables Main findings: Airline network could be characterized by three indicators: Hubness , Resilience , Size Traditional distinction between LCC and Legacies could be reconsidered 16

  17. Our objective  Step 2: Network evolution: what are the drivers of the choice?  Our approach: explain the evolution of the three indicators over time Use of macroeconomic indicators, air market characteristics, airline type as  explanatory variables Estimation of a system of equations on panel data  Main findings: Network Hubness , Resilience and Size have distinct drivers Strategies in terms of network evolution depend on the type of airline 17

  18. Data  Official Airline Guide, OAG Worldwide data on frequencies, schedules and aircrafts for the last 10 years   Monthly data for the third quarter 2005-2015  Focus on the United States Domestic Market Advantage: most studied market with available data on fares  Disadvantage: lack of international flights   Data cleaning: Eliminating routes shorter than 200 miles or routes with less than 10 seats per flight  Recoding of regional/feeder airlines   Final data set: 125358 observations  28 operating carriers ranging from 19 to 25 per year  City number from 413 to 517 and airports ranging from 435 to 537 per year.  18

  19. Step 1: Network characterization  Graph Theory measures: a lot of correlated indicators 19 19

  20. Step 1: Network characterization  Reduction of the information: use of PCA Three principal components explain 94.69% of the sample variability: “ Hubness ”,  “ Resilience ” and “ Network Size ” Theoretical representation of network Hubness / Resilience map 20

  21. Step 1: Network characterization PC2 : RESILIENCE point-to-point hub-and-spoke hub-and-spoke with several hubs with a unique hub PC1 : HUBNESS path or circle  The distinction between LCCs and Legacies is nowadays unclear as highlighted in Jarach et al. (2009) or Bitzan and Peoples (2016).  No distinction in term of Hubness (PC1) between Legacies and LCCs.  Higher Resilience (PC2) values on average for LCCs 21

  22. Step 1: Network characterization Network representation Size / Hubness and Size / Resilience maps WN WN  When the network size increases, Hubness decreases to some level between 0 and -3, both for LCCs and Majors  When the network size increases, Resilience seems to approach a level around -1, except for LCCs (Southwest case) 22

  23. Step 2: Network evolution: what are the drivers of the choice?  Simultaneous Equations Model 𝐻 𝐽 𝑗𝑘𝑢 = 𝛽 𝑗𝑘 + 𝛾 𝑗1 𝑢 𝑀𝐷𝐷 + 𝛾 𝑗2 𝑢 𝑀 + 𝛾 𝑗3 𝑧 𝑢−1 + 𝛾 𝑗4 𝑔 𝑢−1 + 𝛾 𝑗5 𝑒 𝐸𝑀𝑂𝑋 + 𝛾 𝑗6 𝑒 𝑉𝐵𝐷𝑃 + 𝛾 𝑗7 𝑒 𝑋𝑂𝐺𝑀 + 𝛾 𝑗7 𝑒 𝐵𝐵𝑉𝑇 + 𝜁 𝑗𝑘𝑢 where 𝑗 ∈ {1,2,3} indexes one of three network indicators: Hubness, Resilience, Size  j indexes the airlines  t indexes the year  Explanatory variables:  Time trends: 𝑢 𝑀𝐷𝐷 and 𝑢 𝑀 depending on airline type  Macroeconomic indicators: 𝑧 𝐻 represents the output gap, and f the jet fuel prices,  US domestic market characteristics: dummies to control for the 4 mergers occurred during the considered time frame 23

  24. Estimation results Hubness Resilience Size (i=1) (i=2) (i=3) 𝛾 𝑗1 𝑢 𝑀𝐷𝐷 -0.067 0.071 7.681 (1.015) (1.613) (10.352)*** 𝛾 𝑗2 𝑢 𝑀 0.141 0.107 -0.934 (2.399)** (2.895)*** (0.459) 𝐻 𝛾 𝑗3 𝑧 𝑢−1 0.131 0.078 -0.364 (4.689)*** (2.342)** (0.700) 𝛾 𝑗4 𝑔 -0.345 -0.339 -1.081 𝑢−1 (2.686)*** (2.641)*** (0.521) 𝛾 𝑗5 𝑒 𝐸𝑀𝑋𝑂 -1.533 0.160 148.757 (3.283)*** (0.774) (4.101)*** 𝛾 𝑗6 𝑒 𝑉𝐵𝐷𝑃 -1.358 0.104 259.326 (4.523)*** (0.602) (22.377)*** 𝛾 𝑗7 𝑒 𝑋𝑂𝐺𝑀 -0.096 -0.236 74.008 (0.409) (0.985) (8.299)*** 𝛾 𝑗8 𝑒 𝐵𝐵𝑉𝑇 -0.886 -0.555 88.720 (1.249) (3.828)*** (2.749)*** Constant WN -0.922 2.928 409.236 Legacy AA 1.680 -3.665 -35.346 (3.101)*** (9.959)*** (75.839)*** (4.752)*** (18.206)*** (2.089)** LCC B6 3.594 -2.087 -359.558 Legacy AS -0.271 -4.757 -181.121 (24.247)*** (12.668)*** (48.688)*** (0.677) (15.322)*** (7.079)*** LCC F9 5.827 -3.722 -380.930 Legacy CO 1.250 -3.868 -85.541 (44.737)*** (26.114)*** (75.420)*** (3.615)*** (22.771)*** (3.553)*** LCC FL 4.372 -2.436 -339.937 Legacy HA 3.695 -1.620 -374.144 (25.315)*** (33.692)*** (27.508)*** (12.375)*** (6.563)*** (28.786)*** LCC G4 2.290 -3.671 -322.645 Legacy DL 0.930 -3.830 100.584 (7.760)*** (22.403)*** (49.086)*** (2.281)** (21.224)*** (3.261)*** LCC NK 1.376 0.560 -395.394 Legacy NW 0.348 -4.060 -36.161 (2.052)** (1.393) (68.473)*** (1.314) (24.469)*** (1.663)* LCC SY 4.500 -2.965 -421.380 Legacy UA -0.289 -3.718 15.540 (6.298)*** (19.657)*** (37.913)*** (1.293) (21.894)*** (1.475) LCC VX 3.725 1.335 -443.266 Legacy US -0.998 -3.564 16.677 (11.992)*** (1.373) (54.925)*** (2.852)*** (19.464)*** (0.595) ρ 0.3629 Observations 211 211 211 * p<0.1; ** p<0.05; *** p<0.01

Recommend


More recommend