Cost-Variability-Sensitive Preventive Maintenance Considering Management Risk

Abstract

Traditional preventive maintenance policies, such as age replacement, periodic replacement under minimal repair, and replacement policy N, are all studied based on the expected cost criteria without considering the management risk due to the cost variability. As a result, these policies could be significantly beyond the anticipated maintenance budget allocation and lead to crisis. In order to solve this problem, a new analysis methodology is proposed in this paper to consider the effects of both cost expectation and cost variability on the optimal maintenance policy. A new concept of the long-run variance of the cost is defined to represent the maintenance management risk, and then the objective function is revised accordingly to achieve an optimal cost-variability-sensitive maintenance policy. Based on the proposed framework, three traditional preventive maintenance policies have been reinvestigated and the effect of the variability sensitivity on the optimal policies is further analyzed, which reveals general management insights and explicates the search bound of the optimal solution. An example is given to illustrate the importance and the effectiveness of the proposed methodology. Compared with the traditional optimal maintenance policy, the numerical solution shows that the proposed variability-sensitive optimal policies can significantly reduce the maintenance management risk with only a small increase in the expected cost.