An improved particle swarm optimization algorithm for reliability-redundancy allocation problem with mixed redundancy strategy and heterogeneous components

The reliability–redundancy allocation problem (RRAP) has been intensively studied to identify the optimal system design under several constraints, so as to maximize the system reliability as much as possible. Recently, the mixed redundancy strategy where both active redundant components and cold standby components can be simultaneously used by a subsystem has been introduced into the RRAP, leading to a more superior result than that of a subsystem with either active redundancy strategy or cold standby strategy. Nevertheless, in the existing studies, all the components deployed in a subsystem have to be homogeneous even if the mixed redundancy strategy is chosen. The strong premise is released in this paper by introducing heterogeneous components into the RRAP. The resulting optimization problem is resolved by proposing an improved particle swarm optimization (PSO) algorithm with stochastic perturbation nature. The efficiency of the proposed algorithm has been demonstrated by conducting a set of comparative studies on three benchmark problems. As observed in our study, the design flexibility introduced by our method can always yield a greater system reliability than that of the state-of-the-art methods. Most importantly, we also reveal that the switching reliability has significant influence on the optimal system configuration.