Overhaul Scheduling Approach for Metro Vehicles Based on Genetic Algorithms

Abstract

It is necessary to schedule overhaul of metro vehicles in consideration of many factors, such as maintenance positions, equipment and manpower. In order to reduce the idle time of maintenance equipment and workplace, and improve the throughput of maintenance workshops under the available resource conditions taking the overhaul workshop of an enterprise as an example, this paper analyzes and summarizes the overhaul process of rail transit vehicles, and proposes an overhaul scheduling approach based on the genetic algorithm to optimize the allocation of maintenance resources from the perspective of workplace, equipment and time. In this scheduling approach, an evaluation function is applied to transform the train overhaul scheduling into the tasks to optimize the resource allocation for the overhaul process, and the resource allocation to the overhaul process is optimized leveraging the inherent parallelism and global optimization ability of the genetic algorithm, and the optimal chromosome is deemed as the final result of the overhaul scheduling. This paper also analyzes the complexity and scope of application of the proposed scheduling algorithm, and comprises the overhaul scheduling experiment. The experimental results show that a schedule can be generated within one hour using this algorithm, which greatly reduces the labor intensity compared with the manual scheduling that takes several days, and effectively improve the equipment utilization rate; moreover, by reasonably allocating the maintenance resources of the maintenance workshop, the total time spent on vehicle overhaul can be reduced by about 40%.

Keywords

• urban rail transportation
• genetic algorithms
• metro vehicle overhaul
• comprehensive scheduling