Joint optimization of condition-based opportunistic maintenance and spare parts provisioning policy in multiunit systems

Abstract

This study investigates the joint optimization of the strategy for periodic condition-based opportunistic preventive maintenance (CBOM) and a safety policy for the provision of spare parts for multiunit systems with a known number of identical units. Under this policy, maintenance activities as well as the ordering and holding activities for spare parts at each decision point are determined according to the maintenance requirement determined by the state of deterioration of the entire system and the state of the spare parts inventory. On the basis of the derivation of all possible maintenance and spare-parts-related activities at each decision point and their corresponding probabilities, an expected long-run cost rate model is formulated for a general, identical multiunit system to determine its joint optimal strategy. This is accomplished by using semiregenerative process theory that takes into account the cost of maintenance and that of the management of the spare parts inventory. In this model, all possible maintenance requirements and their probabilities are derived using the deterioration state-space partitioning method combined with the derivation of the stationary law of the system state affected by opportunistic maintenance. Further, maintenance activities, the ordering and holding actions for spare parts, and their corresponding probabilities are all determined on this basis. Finally, a case study is carried out to test the proposed optimization method. This involves a wind power farm consisting of a large number of identical main bearings in the turbines. Furthermore, a sensitivity analysis was conducted to analyze the effects of the model.