A multi-objective programming model for timetables on corridors integrating macroscopic and microscopic approaches

Abstract

This paper proposes an integrated framework for timetable design that meets the needs of service planners, who usually adopt a purely macroscopic point of view toward the demand requirements and conflicts that these timetables may cause at stations due to their particular configurations and structures. Service planners may not take these aspects into account, which require a microscopic analysis of the structure of the stations, considering possible conflicts at platforms and/or internal junctions. To this end, the proposed integrated model guarantees feasible timetables while balancing performance indexes for the interests of passengers and operators. Specified bounds are maintained for the differences between service planners optimal timetables and those that are operational at stations. The resulting problem is modeled as a multi-objective Mixed Integer Linear Programming problem. The model can be solved with sufficient accuracy and in a reasonable amount of computing time, as shown by tests on the Madrid-Zaragoza-Barcelona high speed railway network. In this case, we perform a Pareto frontier analysis and find that the problem is well-posed. In addition, the previous model forms the basis of a heuristic procedure for obtaining timetables that avoid, as much as possible, conflicts and concentrations of arrivals/departures at/from stations, thus allowing for improved timetable recoverability and robustness in case of unexpected disruptions.