On the Assessment of Performance and Robustness of Condition-Based Maintenance Strategies
Résumé
Many research works have shown in the literature that condition-based maintenance strategies are very efficient in keeping highest systems availability at lowest maintenance costs. The economic performance of such strategies are usually assessed through the well-known long-run expected maintenance cost rate criterion. However, ones do not often study their robustness in the sense that maintenance costs can significantly vary from a renewal cycle to another. Obviously, assessing a maintenance strategy according to the above cost criterion does not assure its robustness, because at the same time it may lead to over or under budgeting estimation. In this context, the present paper aims develop a cost model that allows quantifying both the economic performance and robustness to assess the condition-based maintenance strategies. Two representative strategies are investigated: a periodic inspection and replacement strategy, and a quantile-based inspection and replacement strategy. The comparisons of these strategies allows us to find out factors which mostly affect their performance and robustness, hence suggesting proper maintenance strategy.