A simulation study of predictive condition-based maintenance strategy for items purchased with extended warranty

This paper studies predictive condition-based maintenance (CBM) and failure-based replacement for a degrading machine protected by an extended warranty policy. After the warranty expires, replacement is made based on the number of major failures, and preventive actions are scheduled based on condition monitoring. The aim is to develop a simulation model for system life cycle cost (LCC) analysis with respect to decision variables, including the length of the warranty period, the maximum number of failures allowed after the warranty period, and the degradation threshold for triggering PM action in CBM. Moreover, the CBM system may be affected by two types of error: prognosis error and maintenance human error. The proposed model incorporates human and prognosis errors into the global system optimization model. Optimal warranty and replacement policies, along with optimal CBM implementation policy, are derived to ensure minimum long-run LCC. This study also discusses the effects of the error parameters on customers’ decisions and costs.