An iterative approach for reducing the impact of infrastructure maintenance on the performance of railway systems

•Considering robustness when adapting the train schedule to maintenance actions is new.•By iteratively re-routing and re-timing trains, cancelations are minimized.•The developed algorithm improves the robustness of the service with more than 10 percent.•All this is a significant improvement compared to the state-of-the-art methods. Planned infrastructure works reduce the available capacity of a railway system and make it more vulnerable to conflicts and delay propagation. The starting point of this paper is a published timetable that needs to be adapted due to the temporary unavailability of some resources. Since the timetable is in operation, changed arrival or departure times and cancelations have an impact on the passengers who need to adapt their travel behavior. In the light of passenger service, a trade-off is made between these inconveniences and the delays that occur in practice due to the reduced capacity. Taking the robustness of the adapted railway timetable into account is a new approach to rescheduling in case of a planned infrastructure unavailability. In this paper, an algorithm that adjusts the train routing and the train schedule to the planned maintenance interventions and keeps the level of passenger service as high as possible is presented. To avoid large inconveniences, the developed algorithm tries to minimize the number of cancelations. Computational results show that by allowing small modifications to the routing and the timetable, the robustness of the resulting solution can improve by more than 10 percent and only few trains need to be canceled.