Optimal condition-based maintenance policy considering nested conditional value-at-risk and operational availability: A case study on semiconductor manufacturing equipment
提出一种基于马尔可夫决策过程的状态维修模型，通过最小化累积成本的嵌套条件风险价值并满足运行可用性约束，确定最优检查间隔和维修策略，并以半导体等离子刻蚀工艺验证其有效性。
To address concerns regarding economic risks and reliability issues in existing maintenance practices, this study introduces a novel condition-based maintenance model that considers failure risks in terms of both cost and availability. Utilizing a Markov decision process, this model determines inspection intervals and maintenance policies aimed at minimizing the nested conditional Value-at-Risk of cumulative costs while satisfying operational availability constraints. By applying this model to the plasma etching process, we demonstrate its effectiveness compared with existing maintenance models. Additionally, we found that higher risk levels do not necessarily lead to stricter maintenance policies, whereas achieving better operational availability incurs additional costs. These findings highlight the importance of balancing cost and availability risks when determining an optimal maintenance policy.