RRAP-CM: A new reliability-redundancy allocation problem with heterogeneous components

Abstract

In this paper, a new version of the reliability-redundancy allocation problem (RRAP) is introduced. In RRAPs, the reliability of the components used in a subsystem is considered as a decision variable and it is also assumed that all the redundant components in each subsystem have the same reliability value. The latter assumption is a restricting and unrealistic one. In this paper, the RRAP is for the first time investigated with the assumption of component mixing, which changes the traditional RRAP model to a more complicated heterogeneous one. A new mathematical model is developed and implemented on different benchmark test problems. Moreover, a powerful meta-heuristic algorithm, called stochastic fractal search (SFS), is used to solve the new mixed integer non-liner programing (MINLP) models. The advantages of the new RRAP model are demonstrated by comparing the results with those reported in the literature. Results reveal that in all the test problems, better structures with higher reliability values are obtained. The improvement values are also compared with improvements achieved in recent studies.