A Model and Algorithm for Passenger-oriented Train Timetabling on Urban Rail Transit Network

Abstract

This paper focuses on optimizing passenger train timetable on an urban transit network with given predetermined long and short routing operation mode. A double objective integer programming model is established in order to reduce the difficulty of solving train timetabling problem of large transit network. A Lagrangian relaxation algorithm was designed to decrease the complexity of solving the problem by transforming the complicated coupling problem into two independent sub-problems of trains and agents respectively. In response to two sub-problems of Lagrangian relaxation, a time-space network was constructed to solve the shortest path problem of passengers and trains by dynamic programming. Finally, the model and algorithm are verified and analyzed by a simple example.