Maintenance modeling of serial-parallel multi-station manufacturing system with failure-induced damage and assembly parts

This paper studies the joint optimization of maintenance scheduling and spare part management for Serial-Parallel Multi-station Manufacturing Systems (SPMMS) with failure-induced damage, which implies the failure of one component may induce the damage of other structurally-connected components. In the proposed hybrid spare part usage policy, either the assembly part, which is a pre-assembled spare part module, or the combination of multiple unassembled single parts can be used to retore the failed station. On the basis of the introduced reliability model, the hybrid usage policy is combined into the joint analysis of spare part inventory and corrective maintenance actions. Then, an opportunistic maintenance policy with the greedy algorithm is developed to reduce the computing complexity. The numerical examples show that the cost advantage of assembly parts and the hierarchical inventory sharing of spare parts can jointly affect the cost effectiveness of the SPMMS. The proposed hybrid spare part usage policy is more cost-effective than the ordinary usage policy, the substitution policy and the component-level policy. The policy comparisons among different hierarchical sharing policies also show that both the division criterion of failure states and the usage priority of spare parts have significantly impacts on the cost effectiveness of the SPMMS.