Joint optimization of inspection and spare ordering policy with multi-level defect information

Abstract

Condition-based maintenance (CBM) is addressing increasing attention in asset management, which makes both maintenance and spare decision-making based on system health state. This paper presents a joint inspection and spare ordering policy for a single-unit system with two levels of defective states, namely minor and severe defective. Inspections are executed irregularly to reveal the defective state, followed by separate spare decision depending on the severity level of defects. The identification of minor defective triggers a normal order, which checks system status more frequently by shortening the inspection interval. Preventive replacement (PR) is scheduled upon the identification of severe defective state, and corrective replacement (CR) is executed upon failure. The timeliness of PR/CR is determined by the availability of spare. PR/CR is immediate if the normal ordered spare is available, and emergency order is needed when facing shortage of normal order. We introduce a threshold level to decide whether to place an emergency order or wait for the normal order when the normal ordered spare hasn’t been delivered. The ultimate objective is to minimize the long-run expected cost per unit time via joint optimization of both inspection interval and threshold level. An optimization algorithm is presented to illustrate the applicability of the policy in the case study. The results show that it is cost effective to shorten the inspection interval from 20 days to 12 days, and the effect of some parameters on the optimal decisions, such as different defective levels, lead time of an emergency order and availability constraint, are also explored and analyzed.