New dynamic heuristic for the optimization of opportunities to use new and remanufactured spare part in stochastic degradation context

This research investigates the case of a machine subject to a stochastic degradation. A condition-based replacement policy is adopted to keep the system running. The replacements are carried out with spare parts of a given quality determined by their degradation level. The replacement spare parts are either new or used: the used ones are the parts recovered from the machine at the time of previous replacements. Before the replacement, a remanufacturing process can be applied on the recovered part to improve its quality, by reducing its degradation level. The purpose of this paper is to propose an optimal maintenance policy, by determining dynamically (i.e. at each moment of decision making): the decision thresholds, inspection dates and spare parts qualities to be installed in the machine. Two components approach is proposed to achieve these objectives, the first component is executed offline and the second component is run online. Numerical examples are presented to illustrate the efficiency of the proposed approach.