Research On Multi-machine Collaborative Flexible Job Shop Reconfigurable Scheduling

With the advent of a globalized economy,market demands are showing the characteristics of increasingly changing.To meet the individual needs of customers and quickly respond to changes in market demand,it is urgently required that the manufacturing system be able to adapt to market changes,adjust its own manufacturing structure and production capacity quickly and reliably to achieve high-efficiency production,which put forward higher requirements for the flexibility of the manufacturing system and the quick response ability of the enterprise.Therefore,it is extremely critical to research on flexible manufacturing systems.Reconfigurable manufacturing(RM)is a new manufacturing model that breaks through traditional scheduling constraints based on manufacturing cell reconfiguration or equipment reconfiguration.RM can satisfy the efficient production of rapidly changing products,and realizes the rapid response to the market.In addition,with the diversified development of processing equipment,such as the emergence of robotic arms,multi-machine collaborative processing has been realized,and it is widely used in various processing environments.However,the research on its task scheduling and resource allocation is still very less,which result in low production efficiency of relevant enterprise.Therefore,based on the flexible job shop scheduling problem(FJSP),this thesis takes low cost and high efficiency as the goal,and takes the resource adjustment and reconfiguration of RM cells as the starting point,and deeply studies the multi-machine collaborative flexible job shop reconfiguration scheduling(MFJRS)model of the manufacturing cell and its solution method under the production demand of new product tasks.The main research contents are as follows:(1)The research status of FJSP,multi-machine collaborative and RM are summarized and analyzed.It confirmed the positive effect of MFJRS technology on enterprise production that starting from the actual production status of the enterprise and market demand.Lastly,the research content of this article is determined.(2)The FJSP that considers multi-machine collaborative(MFJSP)is researched,the characteristics of multi-machine collaborative machining are analyzed,the mathematical model is established and the encoding and decoding are designed.At the same time,a self-learning genetic algorithm(SLGA)is proposed to solve the MFJSP according to the complexity of it,which combines reinforcement learning(RL)and genetic algorithm(GA)to realize the self-learning adjustment of parameters to ensure local and global search capabilities.Finally,the effectiveness of the algorithm is verified by the calculation analysis of the MFJSP instance,and the optimal solution of the model is realized.(3)In response to changes in the processing tasks of new products in the market,the MFJRS model is studied.The characteristics and problems of manufacturing cell reconfiguration are studied and analyzed,and the MFJRS of manufacturing cell is described.Based on the information of the new product-processing task and the original manufacturing cell,the optimization method of the MFJRS is proposed and designed.The reconfiguration and scheduling are deeply integrated,and the different situations of the reconfiguration scheduling of the manufacturing cell and the multi-machine collaborative processing mode are carried out.Finally,the mathematical model of MFJRS is established and its constraint conditions are specified.(4)Aiming at the MFJRS,its solution method and solution process are studied,and its encoding and decoding methods are designed.Aiming at the coupling of reconstruction and scheduling in this paper,a hierarchical search method including optimization of reconstruction layer and optimization of scheduling layer are designed.The Grey Wolf Optimizer(GWO)that used for large-scale NP-hard problems is improved to solving MFJRS,and the solution steps for calculating this problem are drawn up.Finally,through a calculation example from different perspectives and comparative analysis,the correctness and effectiveness of MFJRS model are verified.(5)Taking the production tasks of actual manufacturing enterprises as the background,using the MFJRS technology that studied in this article in actual processing cases,the application of the processing cases verifies that the MFJRS technology can meet the low-cost reconfiguration of manufacturing cells,reasonable allocation of manufacturing resources and high-efficiency production of the enterprise.