AI-Powered Two-Phase Method for Microscopic Periodic Railway Operation Diagrams
Keywords:
AI Optimization, Two-Phase Method, Microscopic Modeling, Time-Discretized Extended Space-Time NetworkAbstract
In the actual organization of railroad transportation, the periodic train schedule of the railroad provides for the arrival, departure or passage of the train at each station in a fixed period, and these time points will be fixed and repeated in each period, so that the periodic train schedule has a significant predictability. This fully demonstrates the notable advantages of high-speed railways in terms of speed and comfort, allowing passengers to conveniently transfer at interchange stations within the railway network, achieving a ‘public transit-like' operation. This paper conducts a microscopic modeling of the railway network based on track circuit sections, and constructs a 0-1 integer programming model for the optimization of periodic train timetable compilation using a time-discretized extended space-time network approach. The model is decomposed according to the solution idea of train decomposition by adopting a grouped sorting method, optimizing only the optimal space-time path of one train route at a time, and the sub-model is solved by calling commercial optimization software. An integer linear programming model is established using operations research optimization methods, and an efficient decomposition algorithm is designed to solve the model, effectively improving the utilization rate of railway line capacity and the quality of transportation services. It innovatively applies a time-discretized extended space-time network method, integrating artificial intelligence (AI) optimization algorithms to construct a 0-1 integer programming model for compiling periodic train timetables.
Downloads
Published
Issue
Section
License
Copyright (c) 2025 Xingyu Zhou

This work is licensed under a Creative Commons Attribution 4.0 International License.