Major Maintenance Schedule Optimization for Electric Multiple Unit Considering Passenger Transport Demand

TL;DR

This paper presents a novel optimization model and genetic algorithm approach for scheduling major maintenance of electric multiple units, aiming to minimize costs while considering passenger demand and operational constraints.

Contribution

It introduces a time-space network and a 0-1 integer programming model combined with a genetic algorithm with simulated annealing to improve maintenance scheduling efficiency and effectiveness.

Findings

Reduces maintenance costs compared to manual scheduling

Improves solution quality and computational efficiency

Effectively incorporates practical constraints and passenger demand

Abstract

It is an important objective pursued in a railway agency or company to reduce the major maintenance costs of electric multiple unit (EMU). The EMU major maintenance schedule decides when to undergo major maintenance or undertake transportation task for train-set, based on practical requirements, such as passenger transport demand, workshop inspection capacity, and maintenance requirements. Experienced railway practitioners can generally produce a feasible major maintenance schedule; however, this manual process is time-consuming, and an optimal solution is not guaranteed. This research constructs a time-space network that can display the train-set status transformation process between available and major maintenance status. On this basis, a 0-1 integer programming model is developed to reduce the major maintenance costs with consideration of all necessary regulations and practical constraints. Compared with the manual process, the genetic algorithm with simulated annealing survival mechanism is also developed to improve solution quality and efficiency. It can reduce the complexity of the algorithm substantially by excluding infeasible solutions when constructing the model.