Production Management of Multi-Objective Flexible Job-Shop Based on Improved PSO
Cấn Văn Mão
Abstract
It is of great practical significance to improve the traditional particle swarm optimization (PSO) for the production management of multi-objective flexible job-shop. However, the current studies have not solved the problem that the traditional PSO cannot apply to the production and processing environment with numerous uncertain changes. Therefore, this paper improves the PSO for the production management of multi-objective flexible job-shop under different conditions. Firstly, the author modelled the production management of high-dimensional dynamic multi-objective flexible job-shop, and explained the pre-reaction dynamic rescheduling method for production management. Then, the PSO was improved in terms of inertial weight, learning factors, global search ability, and local search ability, and the dynamic response mechanism was presented for the improved algorithm. The feasibility of the improved PSO was demonstrated through experiments. The research provides a reference for applying the improved PSO to the optimization in other fields.