An ACO framework for single track railway scheduling problem

Abstract

This work focus on application of ant algorithms to railway scheduling problem. The railway scheduling problem especially on a single track is considered to be NP hard problem with respect to number of conflicts in the schedule. The train scheduling is expected to satisfy several operational constraints, thus making the problem more complex. The ant algorithms have evolved as more suitable option to solve the NP hard problem. In this paper, we propose a mathematical model to schedule the trains that fits into ACO framework.The solution construction mechanism is inspired by orienteering problem. The proposed methodology has the capability to explore the complex search space and provides the optimal solution in reasonable amount of time. The proposed model is robust in nature and flexible enough to handle additional constraints without any modification to the model. The model assumes that set of trains will be scheduled in a zone, that covers several cities and they are optimized with respect to number of conflicts. © 2013 Springer.