Modeling Passengers' Preference on High-Speed Trains: Mixed Logit Model Development

This paper investigates passengers' choice behavior on different high-speed trains. Using the method of RP and SP survey, questionnaires are designed to collect data regarding different level-of-service on trains, including departure time, ticket price, and travel time. The survey is conducted on the high-speed railway of east China. Then the data collected is analyzed by adopting multinomial logit model and mixed logit model. The results show that mixed logit model is better for explaining passengers' taste "heterogeneity." Different passengers have different choice preference on train departure time and ticket price. Gender, career, age, income, travel purpose, and day return are all important factors impacting passengers' choice of departure time and price sensitivity. The conclusion obtained in this article can provide a theoretical support and decision-making basis for high-speed railway operation departments to design train schedules and flexible ticket price system.