Defining Optimal Maintenance Scope for Multiple k-out-of-n Load-Sharing Production Systems Connected in Series Based on RUL Predictions

Abstract

This paper presents a method to define the optimal maintenance scope of a production system consisting of multiple k-out-of-n systems connected in series. Maintenance recommendations are based on Remaining Useful Life (RUL) predictions obtained from a Prognostics and Health Management (PHM) system for each production unit within the production system. Defining the techniques applied in order to estimate the degradation level of production units is out of the scope of this paper. It is assumed here that a PHM system is available and provides the degradation level and RUL estimates for each production unit. The goal is to find the maintenance scope that minimizes the expected total cost per cycle until the next maintenance activity. A k-out-of-n load-sharing system is assumed, which means that the failure of a production unit results in a higher load (and consequently a higher degradation rate) on the surviving production units. The total cost comprises the production cost and the maintenance cost. Production cost of each k-out-of-n system is also affected by the number of surviving production units. A preventive maintenance cost is incurred to maintain a degraded but still functional production unit. A corrective maintenance cost is incurredto maintain a failed production unit. An Ant Colony Optimization (ACO) approach is adopted, which allows the proposed method to deal with large instances of the problem. A numerical example is presented to illustrate the application of the proposed method.