A novel boundary swarm optimization method for reliability redundancy allocation problems

A new methodology called boundary simplified swarm optimization (BSO) is proposed by integrating a novel self-boundary search (SBS) and a two-variable update mechanism (UM2) to improve simplified swarm optimization (SSO) in solving mixed-integer programing problems that include both discrete and continuous variables. To balance the exploration and exploitation ability, the proposed SBS is implemented to update the current best solution (called gBest) based on the boundary conditions and analytical calculations to enhance the exploitation ability of gBest, the UM2 updates the solutions (called non-gBest) that are not gBest to fix the over-exploration of the SSO, in which all variables need to update without exploiting the information of the neighborhood area. The performance of the proposed BSO is ascertained by comparing the results with existing algorithms using four reliability redundancy allocation benchmark problems in the existing literature.