Rescheduling of Railway Rolling Stock with Dynamic Passenger Flows

Traditional rolling stock rescheduling applications either treat passengers as static objects whose influence on the system is unchanged in a disrupted situation, or they treat passenger behavior as a given input. In case of disruptions however, we may expect the flow of passengers to change significantly. In this paper we present a model for passenger flows during disruptions and we describe an iterative heuristic for optimizing the rolling stock to the disrupted passenger flows. The model is tested on realistic problem instances of NS, the major operator of passenger trains in the Netherlands.