A condition-based and opportunistic maintenance model for a two-unit deteriorating system

Abstract

This paper studies condition-based and opportunistic maintenance for a two-unit deteriorating system in series. The system's condition is observed at equidistant time intervals. We suppose each unit may have random failures and fail only once within an inspection interval. The random failure probability of each unit is exponential and the failure rate is dependent on the condition. When one unit fails, corrective maintenance action is performed and another unit's condition will be inspected. If the condition of the inspected unit reaches a preventive maintenance threshold, opportunistic maintenance (OM) action will be performed simultaneously. Considering discounted optimality criteria, we construct an infinite horizon cost model via dynamic programming. Based on the model, a condition-based and opportunistic maintenance policy is determined through minimizing discount total maintenance cost using value iteration algorithm. A numerical example is provided to illustrate how the policy proceeds and cost-savings are analyzed to validate the superiority of this maintenance policy. An approximate policy is proposed and its integrity relative to optimal policy is analyzed through simulation. Finally Sensitive analysis is discussed by using the approach of two-level factorial analysis.