Comparison of alternative fleet-level selective maintenance models

Selective maintenance refers to the process of identifying the subset of maintenance actions to perform from a set of desirable maintenance actions. In this work, existing selective maintenance optimization models are extended to include fleets of systems that share limited maintenance resources. Three performance metrics are defined to evaluate the effectiveness of the selective maintenance actions: (1) fleet reliability, the probability that all missions successfully complete the next mission, (2) the expected number of successful missions, and (3) minimum system reliability. Three corresponding discrete, nonlinear, selective maintenance optimization models are developed. The third model is easiest to solve since its objective function can be linearized. Based on extensive numerical experimentation, it is shown that for small fleets, minimum subsystem reliability provides a good approximation for both fleet reliability and the expected number of successful missions.