Energy-Efficient Train Control: A Practical Application

Abstract

This research presents a practical application of the Energy Efficient Train Control (EETC) problem, which involves a collaboration between the Operations Research group of the University of Bologna and ALSTOM Ferroviaria SpA. The work is carried out within the framework of project Swift, funded by the Emilia-Romagna regional authority. Given a train running on a certain line, the problem requires to determine a speed profile that minimizes the traction energy consumption. In particular, we consider the setting of a real-time application, in which the speed profile has to be recomputed due to changes in the schedule caused by unpredictable events. We introduce three solution methods: a constructive heuristic, a multi-start randomized constructive heuristic, and a Genetic Algorithm. Numerical experiments are performed on real-life instances. The results show that high quality solutions are produced and that the computing time is suitable for real-time applications.