Low-level modeling for trains routing and scheduling in railway stations

Overview & Synopsis

This thesis focuses on the trains routing and scheduling problem faced by railway station managers. The objective is to develop a decision model for railway station managers to generate a conflict-free timetable. The timetable has to fit to the existing infrastructure while respecting the safety and the commercial schedules. Unlike traditional methods that allocate all resources to setup a route for a train until the route is released, our work focuses on the progressive reservation of resources used by trains along their route on the railway station. This technique allows a larger number of trains to be routed simultaneously in a railway node and thus increases railway capacity. In this research work, we focus on mathematical model using mixed-integer linear programming. This mathematical model transforms railway systems into linear programming models. We wanted to build the most general model possible in order to be able to study the railway traffic management problem from several points of view. Thus, the model builds an abstract model based on elements of the railway system: section and connector. This abstract model is able to adapt to all railway networks. A subgroup partitioning strategy based on rolling horizon methodology is proposed. This Math-Heuristic approach divides the problem into many sub-problems in order to reduce the combinatorial explosion. The mathematical models were implemented using the CPLEX tool. We run numerical experiments on the data of the railway station of Bordeaux for full day timetable. The results show the good performance and effectiveness of our models. The resulting timetables are displayed and verified by three Gantt Chart separately corresponding to sections, connectors and trains.