A four-phase algorithm to improve reliability in series-parallel systems with redundancy allocation

In general, reliability is the ability of a system to perform and maintain its functions in routine, as well as hostile or unexpected, circumstances. The Redundancy Allocation Problem (RAP) is a combinatorial problem which maximizes system reliability by discrete simultaneous selection from available components. The main purpose of this study is to develop an effective approach to solve RAP, expeditiously. In this study, the basic assumption is considering Erlang distribution density for component failure rates. Another assumption is that each subsystem can have one of coldstandby or active redundancy strategies. The RAP is a NP-Hard problem which cannot be solved in reasonable time using exact optimization techniques. Therefore, an approach that combines an Ant Colony Optimization (ACO) algorithm as a meta-heuristic phase, and three other heuristics, is used to develop a solving methodology for RAP. Finally, to prove the effciency of the proposed approach, some well-known benchmarks in the literature are solved and discussed in detail.