Integrated degradation-based burn-in and maintenance model for heterogeneous and highly reliable items

In modern industries, manufacturers employ burn-in procedures to detect and remove faulty units before they are put into use. However, with the increasing prevalence of highly reliable items, traditional lifetime-based burn-in approaches may not be effective. So, the degradation-based approach screening units based on their degradation levels becomes more prevalent. Ensuring the safety and reliability of a system requires maintenance, making it crucial to find optimal burn-in settings and maintenance time simultaneously. This paper presents an integrated degradation-based burn-in and maintenance model, assuming a population comprising several heterogeneous subpopulations. The model has four decision parameters, three for the burn-in settings, and one for maintenance. To determine the parameters optimally, two optimization problems are presented. The first problem minimizes expected cost function by considering availability and safety as constraints. The second problem maximizes a joint function of cost and availability given a safety constraint. A comprehensive numerical analysis is conducted to offer valuable insights for practitioners. The proposed approach is compared to a stand-alone planned maintenance model, as well as models that neglect availability and safety constraints. The results of the comparative studies show the integrated model outperforms. The proposed model is illustrated using a case study on high-power semiconductor lasers.