An Optimization on Subway Vehicle Maintenance Using a Multi-Population Genetic Algorithm

In the area with subway engineering, maintenance optimization is an important issue for engineers and subway administrations because these actions allow maintenance costs to be reduced with acquired reliability. Aiming to optimize the maintenance strategy of subway vehicles, this paper presents a multi-objective maintenance model of cost effectiveness. Moreover, there is no doubt that subway vehicles have more than two statuses, rather than simple linear 0-1 statuses (working or failure). This paper develops a multi-status subway vehicle model to reflect the reality. A multi-population genetic algorithm is employed to solve the non-linear model. Additionally, to subsidize the optimized maintenance strategy, this paper studies a case of driven locking mechanism subsystem in subway vehicle. The result shows the advantages of optimized selective maintenance. The maintenance strategy proposed in this paper reveals a new method of subway maintenance.