Robust Train Routing Optimization for Railway Stations

The continuing increase in railway traffic leads to higher capacity usage and reduces available time supplements to handle disruptions. Particularly, busy stations act as critical junctions where even a small delay can spread to multiple lines. Traditional planning approaches, which rely heavily on manual expertise, are time-intensive, making it difficult to explore different options and account for the complex inter‑dependencies. Supporting planners through a systematic method that can anticipate and mitigate delay propagation can enable operators to improve punctuality, reduce costs, and enhance customer satisfaction.

We have developed a robust routing‑optimization methodology that leverages advanced mathematical optimization to construct routing and timetabling plans at the station level that minimize the expected propagation of delays and fix local infeasibilities of a desired network timetable. This can be used to offer decision support to railway planners, both in the tactical phase of making yearly or construction timetables, as well in strategical planning scenarios like infrastructure dimensioning, where multiple potential future scenarios and timetables must be explored.

The algorithm can be integrated with broader network‑planning approaches, as was experimentally done in the project MOTIONAL in collaboration with European research partners. The routing plans produced by our algorithm have been experimentally validated by railway planners on Dutch and Norwegian test instances and deemed to be very promising. Further development and refinement is planned in the continuation of the project.

Publications:

Widmann, Philipp and Weik, Norman, Robust Routing in Railway Stations – Generalizing and Assessing Robustness Concepts Using Absorption Graphs.