Abstract
With the increasing competition and cost pressures, the U.S. airline industry has explored methods to reduce operating costs and diversify revenue sources for improving financial performance. Understanding the influence of operating revenues and expenses on airline profitability is imperative for the long term growth of the airlines and continued generation of profits.
This study examined the cost and revenue data of the U.S. major airlines from the Department of Transportation’s Bureau of Transportation Statistics Form 41 reports between 2009 and 2018. Using SAS Enterprise Miner software, researchers used variables representing revenue and expenses from these data to develop and test predictive models for airline profit generation. Decision trees and linear regression methods were used for two identical datasets one with monetary values and the other with percentage values to identify the best predictor of airline profitability.
From this study, decision tree models appeared to be better predictors of profitability for major airlines. Using the decision model, transport-related revenue and expenses which are incidentals to the air transportation services performed by airlines were found to be the two most influential factors in predicting the U. S. airlines’ profitability.
Scholarly Commons Citation
Choi, W.,
O’Connor, M. B.,
&
Truong, D.
(2019).
Predicting the U.S. Airline Operating Profitability using Machine Learning Algorithms.
International Journal of Aviation, Aeronautics, and Aerospace,
6(5).
DOI: https://doi.org/10.15394/ijaaa.2019.1373
