A review of real-time railway and metro rescheduling models using learning algorithms

Abstract

Planning railway and metro systems includes the critical step of finding a schedule for the trains. Although buffer times and running supplements are added to the schedule to make operations resilient to minor disturbances, they do not protect against all possible events that may lead to conflicts during everyday operations. Thus, real-time train rescheduling models are needed to restore feasibility using actions such as retiming, reordering, rerouting, overtaking or cancelling of trains. Unfortunately, despite many rescheduling models that have been developed in the literature, only a few can learn actions from past, simulated, or ongoing events and cope with disturbances and disruptions’ stochastic nature. However, the last decade’s expansion of learning algorithms is gaining momentum in the train rescheduling literature by bringing promising novel ideas. This paper aims to review the state-of-the-art learning algorithms applied to the real-time railway and metro rescheduling, identifying challenges and opportunities while making a parallel with other areas where learning algorithms led to breakthroughs.