Research On Networked Integrated Scheduling Algorithm For Cooperative Manufacturing Of Flexible Machines

The traditional job shop scheduling method is a manufacturing scheduling scheme that separates the processing and assembly of products.In the face of the need for a single and complex product,due to the feature that the separation of processing and assembly,the processing and assembly of the product cannot be paralleled,resulting the increase of the processing time.Integrated scheduling regards the assembly of the product as a special process for the production of the product,so that the processing and assembly of the product components are scheduled together,to make them parallel,thereby improving the production efficiency of a product.Integrated scheduling has researched so far,there are some mature solutions for various problems in single workshop,double workshop or a small number of workshops.However,with the further development of globalization and the deepening of cooperation among enterprises,the integrated scheduling faces the following challenges: the processing machine involved in product production is often not in the same workshop or in a few workshops,and the transportation cost between machine cannot be ignored;The existence of flexible machine,the current selected processing machine for the process will have an uncertain impact on the final completion time of the product.This paper studies on the networked integrated scheduling for collaborative manufacturing of flexible machines problem.With the goal of minimizing the completion time,the artificial bee colony algorithm is used to solve the flexible networked integrated scheduling problem.In order to apply the ABC algorithm to this problem,encoding the solution in double-chain form,and combine the process-chain crossover as the mutation of the ABC algorithm solution,and randomly allocate machines in the machine-chain to jump out of the local solution is proposed,thus the ABC algorithm of FNISP is proposed.For FNISP,the order of process processing and machine selection will have an impact on the product's completion time.In the employed bee phase,a dynamic perturbation step strategy is proposed.In order to make the following bees have more probability to choose the solution which is not similar to the optimal processing structure to search,the processing structure coefficient is proposed.As a parameter of profit,this coefficient affects the probability of neighborhood search of the solution selected by the following bees,thereby improving the efficiency of FNISP's ABC algorithm.For FNISP,flexible network integrated scheduling software is designed and implemented.The software can solve the provided instances and get a scheduling Gantt chart.Several calculations were performed on each instance,and the average time of the solutions and the degree of dispersion of the obtained results were analyzed.