Applied Research Of Improved Ant Colony Optimization In Flexible Job-shop Scheduling Problem

As the global manufacturing industry gradually develops towards intelligent manufacturing,the issue of production scheduling has become a key content for modern enterprises to realize advanced manufacturing technology.Compared with the traditional Jobshop Scheduling Problem(JSP),the Flexible Job-shop Scheduling Problem(FJSP)is more in line with the actual situation of the production shop,it has become a research hotspot in the academic and engineering fields in recent years.Aiming at the weak global optimization ability and slow convergence speed of Ant Colony Optimization(ACO)in solving the FJSP,this thesis makes improvements by analyzing the characteristics of ACO.First,a machine selection strategy is designed,which prioritizes the machines of each process based on the processing time,and selects them with different probabilities,so as to ensure the quality of the initial solution and improve the global optimization effect of the algorithm.Then,according to the phased characteristics of ACO,a phased state transition rule is proposed to change the selection state of ants.In the initial stage,random search is used to reduce the possibility of the algorithm falling into local optimization;When the richness of the initial solution reaches a certain degree,the positive feedback mechanism is triggered to make the algorithm evolve towards the optimal solution;After several iterations,the greedy search method is used to speed up the convergence speed of the algorithm in the later stage and improve the execution efficiency.Moreover,local pheromone updating and global pheromone updating are combined to expand the search space of the algorithm,learn from the idea of Max-Min Ant System(MMAS),the maximum and minimum values of pheromone concentration are set on each path to control the heuristics of prior knowledge to ants,and ensure the possibility of ants exploring the optimal solution.Finally,the mathematical model of ACO is established to solve the FJSP,and choose to minimize the maximum completion time,the bottleneck machine load and the total machine load as the solution goals.Through the experimental simulation of classic examples and the comparative analysis with the research results of other related literatures in recent years,the feasibility and superiority of the proposed algorithm are verified.A job shop scheduling management system is designed and developed for a manufacturing enterprise,the improved ACO is applied to it,which verifies the feasibility of the new algorithm in practical application.