Optimizing corrective maintenance for multistate systems with storage

Production-storage systems abound in different industries, where the storage is used to store surplus product generated by the production subsystem (PS) and compensate the deficiency when the PS performance cannot meet the demand. The existing models mostly failed to consider the effects of random external shocks and the further mitigation actions. This paper makes contributions by modeling shocks and the mitigating corrective maintenance policy (CMP) that aims to restore the PS performance to a higher level, alleviating negative impacts from the shocks and enhancing the mission success probability (MSP).The proposed CMP defines the condition triggering the maintenance action, which encompasses the shock occurrence time, the PS state, the number of shocks since the previous maintenance, and the amount of product in the storage. A new numerical algorithm is put forward to assess the MSP of the considered multistate production-storage system under any given CMP. The genetic algorithm is then implemented to determine the optimal CMP that maximizes the MSP. A case study of a six-state power generating system subject to voltage surges caused by external shocks is conducted to illustrate the proposed model. Impacts of several model parameters on the MSP and optimization solutions are also examined.