Analysis of velocity calculation methods in binary PSO on maintenance scheduling

Abstract

Even though binary optimization in Genetic Algorithm has been studied extensively, Binary Particle Swarm Optimization (BPSO) methods are relatively new in the literature. There are several binary PSO methods proposed in the literature. Firstly, binary PSO was presented by the father of the PSO method. Then several other methods have been proposed claiming better results. In this work, we analyzed these BPSO methods on maintenance scheduling problem for condition based maintenance system. Evaluation of these methods revealed that their difference mainly focuses on the calculation of velocity vector. Thus, we focus on/compare different velocity calculation methods in BPSO on maintenance scheduling problem in Condition Based Maintenance, which has been presented with Genetic Algorithm in our early work. The tradeoff between maintenance and failure is quantified in risk as the objective function to be minimized.