Selective maintenance modeling for a multi-state system considering functionally significant items

Selective maintenance problem arises in many large multi-state systems which are required to perform multiple missions in succession. Maintenance work can only be carried out in finite maintenance breaks between any two consecutive missions. Only a selected set of multi-state components can be taken maintenance action under limited resources such as time and cost. Traditional selective maintenance strategies do not consider Functionally Significant Items (FSI) and determine which components need to be maintained only by using maintenance resources and performance state. However, in an actual industrial environment, such as safety will also affect maintenance decisions. Therefore, it is necessary to fully consider various factors to determine FSI. In such a case, Analytic Hierarchy Process (AHP) is used to determine FSI in this paper, and then the multi-state system selective maintenance model considering FSI is established with the goal of maximizing the reliability to perform next mission. An improved genetic algorithm (GA) is employed to solve the optimization problem instead of enumeration. Finally, an example is presented to illustrate the necessity of considering FSI.