Multi-objective Scheduling Simulation of Flexible Job-shop Based on Multi-population Genetic Algorithm

In view of the difficulty of obtaining the optimal solution to the multi-objective scheduling of flexible job-shop by the general genetic algorithm, this paper takes into account the shortest processing time and the balanced use of machines, and puts forward the multi-population genetic algorithm based on the multi-objective scheduling of flexible job-shop. The method attempts to minimize the longest make-span of workpieces, the load of each machine, and the total machine load through the overall process scheduling of the job-shop. Research results reveal that the proposed method is highly efficient in seeking the optimal machine allocation chain, and effective in avoiding the complex process of intermediate assignment, making it easier to obtain the said optimal solution. The feasibility and effectiveness of the proposed method are also validated by two instances. Compared with the conventional flexible job-shop scheduling algorithms, the proposed algorithm boasts better population quality, algorithm starting point, and initial expression. Besides, it is far better than other algorithms in terms of the initial solution quality and the convergence rate. Despite the local fluctuations in the early phase of the genetic process, the total machine load and the machine load variance are gradually declining and the curves start to converge after the 50 generation. (Received, processed and accepted by the Chinese Representative Office.)