Train schedule optimization based on schedule-based stochastic passenger assignment

Abstract

In this study, we propose a new schedule-based itinerary-choice model, the mixed itinerary-size weibit model, to address the independently and identically distributed assumptions that are typically used in random utility models and heterogeneity of passengers’ perceptions. Specifically, the Weibull distributed random error term resolves the perception variance with respect to various itinerary lengths, an itinerary-size factor term is suggested to solve the itinerary overlapping problem, and random coefficients are used to model heterogeneity of passengers. We also apply the mixed itinerary-size weibit model to a train-scheduling model to generate a passenger-oriented schedule plan. We test the efficiency and applicability of the train-scheduling model in the south China high-speed railway network, and we find that it works well and can be applied to large real-world problems.