ABSTRACT
Maintenance is an essential activity in any production facility. In this context, numerous research works have provided multiple relevant approaches to optimize maintenance decision based on different characteristics of the system in order to reduce the associated costs and in the same time to maximize the availability and the safety of the considered system (Rausand and Hoyland 2004). Condition-Based Maintenance (CBM) approaches (Wang 2002), where the decision is directly driven by a measurable condition variable which reflects the deterioration level of the system, have proved their efficiency compared to classical time-or age-based policies (Gertsbakh 2000) in terms of economical benefits and also in terms of system safety performance. Most of the maintenance decision framework does not take into account some potential variations in the deterioration mode. The classical approaches in maintenance optimization do not explicit the relationship between the system performance and the associated operating environment.
