Research On Multi-Objective Shop Scheduling Based On Game Artificial Bee Colony Algorithm

As market competition becomes more and more fierce,and customer needs become more and more diverse,the production method of firm has also changed from the past single-variety and large-volume to the multi-variety and small-batch.Therefore,in the MO-FJSP,if the influence of production assistance time in batch processing of work is considered,the results will be closer to the actual situation and have greater guiding significance for production.Aiming at the MO-FJSP with multiple varieties and small batches,a GABC is proposed to minimize the completion time、the total processing cost,and the machine load.The algorithm uses the ABC as the framework and embeds game theory to solve multi-objective problems.The specific improvements are as follows:First,in the initialization process,in order to ensure that the initial population has a certain diversity and better quality.this paper uses three strategies to generate the initial solution.Second,in order to optimize the quality of the population,in the process of solving,the IABC is used to continuously optimize the population by designing crossover,mutation and local search mechanisms.The catfish effect is integrated into the ABC to remove the bees with the worst food source quality and replace them with the same number of catfish bees,thereby enhancing the algorithm’s global search ability.Third,use game theory to solve the MO-FJSP.Through game theory,the strategies between the three optimization objectives are optimized to achieve the optimal combination of sub-games to refine the NE and obtain the optimal solution of the problem.For the purpose of verifying the capability of the algorithm,this paper carried out simulation tests on benchmark examples and actual cases,and through a series of comparisons with other algorithms,the GABC algorithm shows good solution performance in solving the MO-FJSP of multi-variety and small-batch.