Passenger service-oriented timetable rescheduling for large-scale disruptions in a railway network: A heuristic-based alternating direction method of multipliers

Unpredictable disruptions arising in railway operations can cause significant inconvenience for passengers, including missed connections or deviations from their travel plans. To address this issue, this paper proposes a service-oriented timetable rescheduling approach to assist dispatchers in managing major disruptions in a railway system that incorporates a seat reservation mechanism. The railway network is abstracted at a near-microscopic level, and the actual passenger service is explicitly modeled at the station platform. An integer linear programming (ILP) formulation is then constructed based on the space-time network, which considers train retiming, reordering, and local rerouting to minimize the adverse effects on passengers. A compact form of an incompatible set is formulated to describe the coupling connections of train headway and station capacity, to minimize weighted passengers' inconvenience. To overcome computational barriers and practical complexities, the paper proposes an Alternating Direction Method of Multipliers Heuristic, i.e., ADMM-H. The ADMM-H decomposes the original problem into a series of train-specific subproblems that are easy to solve. An importance-based heuristic sorting strategy is adopted while individually solving train-level path subproblems based on their duality and feasibility. The proposed methods were tested using numerical instances based on real-world data to verify their effectiveness and efficiency. The results showed that the service-oriented train rescheduling approach significantly improves the passenger experience during disruptions, with an improvement (up to 19.86 %) compared to traditional operation-oriented approaches. The ADMM-H algorithm can reduce the computational time (up to 8.41 %) and obtain a sub-optimal solution with a lower gap compared with the naive ADMM. The proposed approach and algorithm can be valuable for railway operators to mitigate the negative effects of disruptions and enhance passenger satisfaction.