Research On Master Plan Scheduling Based On Improved Multi-objective Particle Swarm Optimization For Parallel Casting Workshops

The master plan scheduling of parallel casting workshop is a process for the group multi workshop foundry enterprise to complete the optimal order scheduling decision of parallel casting workshop with the same process and different benefits under the fuzzy production environment.The efficiency of the current manual scheduling is low,and the results are not scientific and reasonable,which is easy to casue serious order delay,low production efficiency,unbalanced workshop load and other problems.Therefore,this paper studies the modeling and solution method of parallel casting workshop master plan scheduling based on Pareto discrete particle swarm optimization algorithm,and verifies the effect of the proposed multi-objective optimization scheduling scheme through multiple scale simulation experiments.The main work is as follows.Firstly,the characteristics of master plan scheduling problem in multi workshop foundry are analyzed,the order-process-workshop information model and the black box model of casting workshop are established.A representation method of determining the order production time and due date with triangular and trapezoidal fuzzy numbers is constructed to reduce the impact of fuzzy production environment on scheduling results.A multi-objective integer programming mathematical model is established with the penalty cost of order advance / delay,the maximum completion time and the load balance degree of the workshop as the evaluation indexes.The feasible solution space scale formula of the parallel workshop master plan scheduling model and its influence on the solution algorithm are analyzed.Secondly,a Pareto discrete particle swarm optimization algorithm for master plan secheduling of parallel workshops is proposed.Through the production process information,the batch of orders to be scheduled and the division of the parallel workshops are completed,and the improved multi-objective particle swarm optimization algorithm is used to solve the optimal scheduling for each parallel workshop,so as to realize the algorithm auxiliary scheduling.The improvement of the algorithm mainly includes: adopting discrete particle encoding and decoding method,reinforcing particle information exchange based on twopoint crossover operator,improving the generation method of individual and global optimal solution set,designing the particle update based on the fuzzy optimization method and second-order heuristic rule,introducing simulated annealing mechanism based on Pareto to control the iteration.The comparison test results show that the dominance and distribution of Pareto optimal solution obtained by the proposed algorithm are significantly improved compared with the other three multi-objective optimization algorithms.Finally,a number of parallel workshop scheduling simulations with different problem scales are designed.By comparing the solution based on the improved multi-objective particle swarm optimization algorithm with the weighted coefficient method and the restrict transformation method of genetic algorithm,the feasibility of the proposed master plan scheduling scheme and the improved algorithm's solution effect under different scale problems are verified.The results show that the solution of the proposed scheme has obvious advantages over the above two single-objective optimization schemes.The penalty cost of order advance / delay and the maximum completion time of the final recommended scheduling determined by the fuzzy optimization method are effectively reduced,and the load balance degree is significantly improved.The average optimization rate of the recommended scheduling's objective value on each problem scale is as high as 81.86%.