Research On Distributed Flow Shop Scheduling Optimization Based On Hybrid Swarm Intelligent Algorithm

Shop scheduling problem is an important issue in the industrial production process.An excellent scheduling method can improve resource utilization,shorten production time,and improve production efficiency.Distributed flow shop scheduling problem is a typical problem of shop scheduling.This problem is more in line with the current production mode and has important guiding significance for modern manufacturing.Due to the high complexity of the distributed flow shop scheduling problem,the existing heuristic algorithms are difficult to achieve the expected objectives when solving this type of problem.This thesis proposes different hybrid strategies to improve the particle swarm optimization algorithm and form a hybrid swarm intelligence algorithm to solve the distributed flow shop scheduling problem.The main research contents are as follows:Firstly,a hybrid multi-objective swarm intelligent algorithm based on multi-directional update is proposed to solve the flow shop scheduling problem in the case of a single factory.The hybrid multi-objective swarm intelligent algorithm includes a multi-directional update strategy and a particle’s sequence-based local search strategy.The multi-directional update strategy enables the algorithm converge to the Pareto front quickly.The local search strategy fine-tunes the particle sequence to further improve the particle quality.The proposed algorithm is compared with five algorithms on the standard flow shop scheduling problem proposed by E.Taillard.The results show that the convergence of the proposed algorithm is significantly improved,and the searched solutions have good distribution performance.Secondly,a multi-objective swarm intelligence algorithm based on hybrid difference strategy is proposed to solve the distributed flow shop scheduling problem in the multi-factory situation,and the particles are encoded by double vectors.The hybrid difference strategy includes a multi-directional update strategy and a particle sequence difference-based local search strategy.The multi-directional update strategy can speed up the convergence of the algorithm toward the Pareto front.The local search strategy can enhance the connection between particles within the particle swarm,and guide the poorly performing particles to move closer to the better performing particles,which strengthens the local search capability of the algorithm.In the experiment part,the proposed algorithm is compared with six algorithms,and the results show that the proposed algorithm has better performance in solving the distributed flow shop scheduling problem.The hybrid multi-objective swarm intelligence algorithm proposed in this research uses multi-directional update strategy and different local search strategies.The multi-directional update strategy improves the algorithm framework,improves the algorithm’s convergence performance in solving multi-objective problems,and the searched solutions have good distribution performance.The designed two local search strategies adjust the particle sequence in different ways to enhance the particle’s ability to explore the neighborhood and further improve the particle quality.Experiments prove that the proposed algorithm and strategy can better deal with the distributed flow shop scheduling problem,and at the same time provide a useful reference for the hybrid swarm intelligent algorithm to solve other complex multiobjective optimization problems.