OPTIMAL COMBINATORIAL REPLACEMENT POLICY UNDER A GIVEN MAINTENANCE INTERVAL FOR THE COMBINED GOVERNOR IN DIESEL LOCOMOTIVES

Combined governor is one of key components in diesel locomotives, as a subcomponent it must meet the existing maintenance periodic of the diesel locomotive, while it is passively replaced or maintained in a midway because of over/under-maintenance in practice. In this paper, four reliability models of a sequential PM cycle are developed using three years of maintenance data of combined governors in one Chinese Railway Bureau to determine its reliability distribution, in which some zero-failure data and censor data are used. Meanwhile, a novel combinatorial replacement (CR) policy is proposed to optimize its preventive maintenance (PM), in which a component is replaced several times using a preventively maintained one in a given operational interval. After that, necessary optimizations are introduced based on the determined reliability models and the maintenance interval given by the PM criterion of diesel locomotives (23000km similar to 25000km), and then the genetic algorithm is also used to solve the constraint optimization function. Results show that the proposed CR policy is the best policy among the existing policy and the general (T, delta) policy, and other results can be viewed as an optional policy when spare components are limited.