The application of hybrid Particle Swarm Optimization in job shop scheduling problem

The background of this paper is pursuing the shortest total processing time of job shop scheduling problem. The research is inspired by the Migrating Bird Optimization (MBO) algorithm. In order to improve the local search capability, the Particle Swarm Optimization (PSO) algorithm is combined with the MBO algorithm to optimize the efficiency of the PSO algorithm. MBO algorithm is a new Neighborhood Search algorithm, which simulates the V formation in birds during migration, and optimizes the algorithm by reducing energy consumption. The algorithm starts with a certain number of parallel solutions, so individuals in the population can not only find the better solution from their own neighborhood, but also find the better solution from the previous individual neighborhood, which makes it quick to find the optimal solution. In this paper, we add the MBO algorithm in the PSO algorithm to update the individual and global extremes of the particles by increasing the information of other particles in the neighborhood of the common particle, adjusting the flight state to achieve a larger search range and global finest solution. Compared with the PSO algorithm in the literature [9], the convergence of this algorithm is more obvious, the number of trapped in local optimal solution is decreased significantly. Paper is divided into four parts, the first is the introduction, the second one introduces the mathematical model of job shop scheduling, the third part mixes the MBO algorithm with the PSO algorithm, and the last is the experimental analysis of the algorithm.