Bayesian Importance Measure-Based Approach for Optimal Redundancy Assignment

摘要

The present work resolves a reliability optimization problem, using assignment of redundancy via Bayesian importance. If a system fails, it is of interest to know which component has caused the failure of the system and how important the different components are for the functioning or falling of the system. Bayesian importance of a component is a measure that reflects its role in system failure, and hence, it might be a good idea to strengthen a system component with a high value of Bayesian importance. Thus, this measure can serve as a basis for making a decision with regard to selecting the right system component for maximizing system reliability through adding redundancy to the said component. Some general results have been proved toward optimal assignment of redundancy in complex coherent systems. The method developed here is capable of accommodating any structure function and any component-life distribution. The procedure has been illustrated through numerical examples in the context of some complex coherent systems.