Imperfect maintenance modeling and assessment of repairable multi-component systems

Industrial systems are subject to several maintenance tasks during their operational life. To ensure optimal performance of these systems, it is crucial to build maintenance models that describe the real behavior of the system. In this scope, the herein paper provides an original approach for maintenance efficiency modeling of multi-component systems. The proposed approach seeks to model the behavior of multi-component system after the replacement of one or more of its components and evaluate this effect. As a matter of fact, from maintenance efficiency point of view, the replacement of one component can be considered as perfect maintenance for the component itself but imperfect for the whole system. To this end, a hybrid model, which considers geometric adjustment of failure intensity, proportional adjustment of failure intensity and proportional adjustment of age is provided. Then, an efficient procedure based on Bayesian networks is provided for the assessment of the imperfect maintenance model parameters. Finally, the proposed model is compared to the most common imperfect maintenance models.