Modeling Airlines Decisions On Route Selection Using Discrete Choice Models - Data and Supplementary Materials

Listed in Datasets

This publication contains the data and supplementary results used for the analysis and estimation of airlines' decision on route selection.

Version 1.0 - published on 29 Oct 2015 doi:10.4231/R74747TG - cite this Archived on 25 Oct 2016

Licensed under Attribution-NonCommercial-ShareAlike 4.0 Internat. (CC BY-NC-SA)


The supplementary results are documented in Appendix.pdf

The following are the data files included in this set;

1. finalData****    - Has the data processed by applying filters to raw BTS data
2. airportsList        - Identifies the airports included in our network topology
3. unionSet        - Matrix that indicates the set of routes that existed in given years

Details of the data files.

1. finalData****
This data file includes the filtered data from BTS that we used for our analysis. The '****' indicator in the name indicates the year for which this data file is created. The included matrix has the following columns:
possRoutes:   This is the set of 8646 possible routes, which, if all of them exist, would make for a fully connected network. They are identified by a ten digit code, the first five digits of which are  the Airport ID of the origin airport as assigned in the BTS dataset, and the last five digits are the same for the destination airport.
DISTANCE:    This is the distance between the airport pairs.
ROUTES:       A column that identifies the routes that exist in the given year. '1' indicates the route exists while '0' indicates that it does not.
ADD:              These are the routes that were added in the current year; they are identified by '1'.
DEL:              These are the routes that were deleted in the current year; they are identified by '1'.
HUBS:           These identify the 'hub-level' of the route. '0' indicates both the terminal airports on this route are non-hubs, '1' indicates that one of the two airports is a hub, and '2' indicates that both airports are hubs.
DIRECT_COST:    This is the direct operating cost on the route. It is calculated as a weighted by number of operations sum using data in the BTS Schedule P-5.2.
SEG_PAX:    This is the segment demand the the route as reported in the BTS T-100 dataset.
MKT_PAX:    This is the market demand the the route as reported in the BTS T-100 dataset.

2. airportsList
This file has three columns. The first column is the airport code, the second column is the five digit airport ID given in the BTS dataset, and the third column identifies (using indicator '1') if the airport is a hub.

3. unionSet
This file identifies the set of routes that existed in the years under consideration. Each of the columns are for a year and correspond to the 'ROUTES' column in the finalData file as described above.

Cite this work

Researchers should cite this work as follows:



This is the first publication of this project. 

The Purdue University Research Repository (PURR) is a university core research facility provided by the Purdue University Libraries, the Office of the Executive Vice President for Research and Partnerships, and Information Technology at Purdue (ITaP).