Optimal condition-based maintenance with general repair and two dependent failure modes

• Published in: Computers & industrial engineering Vol. 141; p. 106322
• Main Authors: Zheng, Rui, Makis, Viliam
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.03.2020
• Subjects: Gamma process; General repair; Policy iteration; Proportional hazards model; Semi-Markov decision process; Policy iteration; Gamma process; Semi-Markov decision process; General repair; Proportional hazards model
• ISSN: 0360-8352; 1879-0550
• DOI: 10.1016/j.cie.2020.106322

•New CBM policy with general repair, minimal repair and replacement is proposed.•Two dependent failure modes represent failure mechanism better.•New, effective computational algorithm is developed using matrix-based approximation in SMDP framework.•The proposed policy outperforms other widely used CBM policies. This paper proposes a condition-based maintenance (CBM) policy for a system subject to dependent soft and hard failures. When the deterioration exceeds a predetermined level, the system experiences a soft failure that can be rectified by corrective general repair or corrective replacement. Three possible actions, namely no maintenance, preventive general repair, and preventive replacement, are considered at each inspection epoch. The hard failure rate depending on both age and deterioration is described by a proportional hazards model with the covariate characterized by a gamma process. The hard failure can be corrected by minimal repair, general repair, or by replacement. Different from previous CBM policies, no thresholds are used to define maintenance actions. The objective is to determine the optimal maintenance policy that minimizes the long-run expected average cost per unit time. The optimization problem is formulated in the semi-Markov decision process (SMDP) framework and solved by the policy iteration algorithm. Two practical numerical examples are provided to illustrate the proposed policy. The comparison with other CBM policies shows an outstanding performance of the proposed policy, as well as the significance of minimal repair and general repair in CBM decision making.