Abstract: The remaining useful life (RUL) can be predicted using the information related to the equipment health state, and the decisions of maintenance and spare ordering are made to reduce the cost associated with maintenance and spare parts. A joint policy of maintenance and spare ordering for single-component systems is proposed based on RUL, in which the control limit policy is used to maintenance decision-making, i.e. judging whether a preventive replacement or a failure replacement is performed depending on the degradation level. Meanwhile, the RUL at each monitoring time is estimated based on the historical degradation information, and the order threshold is introduced to decide whether an order is made or not. All possible renewal scenarios are obtained by analyzing the spare state when the replacement is required, and the corresponding occurrence probability is derived to calculate the expected cost and length. Then, the joint policy model is established to minimize the expected cost per unit time, and a discrete event simulation algorithm is designed to address the proposed model. Finally, a case study is given to demonstrate the proposed model and algorithm, and the minimal expected cost per unit time is 14.656 3 with regard to the optimal preventive replacement threshold 8 and the optimal spare ordering threshold 1 000.

Key words: remaining useful life, preventive replacement, spare ordering, Wiener process, control limit policy, the renewal award theorem