Discrete choice model for Amtrak Acela Express revenue management

In this article, we propose a pricing strategy for Amtrak Acela Express focusing on business class passenger departing from Washington, DC. A two‐step process is proposed to model passenger demand. In the first step, passenger choice model of booking time is estimated using a multinomial logit model. In the second step, a linear regression determines passenger demand in response to fare price; the effects of departure day of week and destination specific are incorporated. The proposed models, estimated on ticket reservation data, are incorporated into a non‐linear programming problem to maximize expected revenue. The results indicate a potential for significant revenue improvements.