A Memetic Algorithm for High-Speed Railway Train Timetable Rescheduling

Abstract

This study addresses a high-speed railway train timetable rescheduling (TTR) problem with a complete blockage at the station and train operation constraints. The problem is formulated as a mixed-integer linear programming (MILP) model that minimizes the weighted sum of the total delay time of trains. A memetic algorithm (MA) is proposed, and the individual of MA is represented as a permutation of trains’ departure order at the disrupted station. The individual is decoded to a feasible schedule of the trains using a rule-based method to allocate the running time in sections and dwell time at stations. Consequently, the original problem is reformulated as an unconstrained problem. Several permutation-based operators are involved, including crossover, mutation, and local search. A restart strategy was employed to maintain the the population diversity. The proposed MA was compared with the first-scheduled-first-served (FSFS) algorithm and other state-of-the-art evolutionary algorithms. The experimental results demonstrate the superiority of MA in solving the TTR through permutation-based optimization in terms of constraint handling, solution quality, and computation time.