A Novel Job-Shop Scheduling Strategy Based on Particle Swarm Optimization and Neural Network

This paper innovatively introduces particle swarm optimization (PSO) and neural network (NN) to solve the job-shop scheduling problem (JSP). Each particle in the swarm was treated as a connection in the NN. Then, the connection weight was iteratively updated according to the latest position of the corresponding particle. In this way, the NN no longer falls into the local optimum trap. Then, the PSOoptimized NN was applied to solve the JSP with a single objective: minimizing the maximum makespan. Through experiments on benchmark problems, it is confirmed that the proposed strategy outperforms the other scheduling methods in fulfilling the optimization objective.