Towards partial autonomy of operation and maintenance of unreliable equipment

This study considers the case of unreliable equipment subjected to random failures that induce high maintenance and environmental costs. We consider situations where the equipment is located in remote areas, which are difficult to access, and situations where there could be confinement linked to a pandemic, making it impossible to perform the replacement of the failed equipment. In such situations, the objective is to explore the possibility of providing a system with self-maintenance capabilities to a certain extent by adding redundant (stand-by) identical modules. Both the designs (with and without passive redundancy) are considered. A mathematical cost model is developed for each alternative to help decide whether to adopt redundancy and determine the optimal number of redundant modules, which minimises the total expected cost. The latter includes the costs related to the acquisition, maintenance, and recycling of failed modules. A numerical example is presented, and a sensitivity study is performed to investigate the effect of variations in relevant input parameters on the optimal design.