Stochastic Schedule–Based Optimization Model for Track Allocations in Large Railway Stations

Abstract

This paper develops a track allocation model for large railway stations based on stochastic schedules, which considers balanced utilization as well as minimum occupation time of station tracks. The stochastic schedule is represented by the mean and variance of both train arrival and departure times. Due to the nondeterministic polynomial completeness of the track allocation problem, a tailored simulated annealing algorithm is developed to solve the proposed model. In addition, a numerical case from the Baoji railway station is studied that demonstrates the model’s performance. After solving the model, the results of track allocation schedule can detect the limited carrying capacity area in the station, and the capacity relationship of two station bottleneck sections can be obtained. The capacity of the left station bottleneck area is about 1.48 times higher than that in the right station bottleneck area in Baoji railway station.