Multi-objective Flexible Job-shop Dynamic Scheduling Based On Grey Wolf Optimization Algorithm

How to do a good job of production scheduling is a key issue for the survival of a company.A good production scheduling solution makes a company more productive and the rationality of the scheduling solution has a great impact on the stability and profitability of the company's production.In the past,experts and scholars have mostly focused on static shop floor scheduling issues.However,many unexpected conditions often occur on the production site,e.g.the arrival of new workpieces,mechanical breakdowns,etc.Taking these disruptions into account,these dynamic events inevitably have an impact on the current scheduling plan,so that the delivery dates agreed between manufacturers and customers are often ambiguous and,due to the different perspectives they are in,there are often contradictions between manufacturers and customers in terms of delivery dates.From the company's point of view,the final completion time needs to be controlled within a suitable processing time window to ensure that orders are delivered on time without increasing production costs;from the customer's point of view,the delivery period should be within a suitable time frame to ensure that the next production plan is launched on time,otherwise the early/delayed delivery of workpieces will have an impact on the satisfaction of both the company and the customer,causing losses to both parties,therefore in the production system It is therefore necessary to take into account the overall satisfaction of the company and the customer in the production system.This is why it is necessary to take into account the overall satisfaction of both the company and the customer in the production system.It is a common concern of companies and experts to reduce production fluctuations caused by these dynamic disruptions in order to satisfy the customer and at the same time guarantee the satisfaction of the company.The research and practical value of considering fuzzy lead times in shop floor scheduling is very important.With the refinement of fuzzy theory,the dynamic parameters of the production process can already be expressed using fuzzy mathematical methods.The use of suitable mathematical algorithms in solving the scheduling model to match the scheduling plan with the production plan of the order has become an important research direction in the field of scheduling.The Grey Wolf Optimizer(GWO)has the advantage of simple parameters and rapid convergence.However,there are also disadvantages such as local optimality and poor stability during the operation of GWO.In this paper,the grey wolf algorithm is improved to promote a balance between local optimality and global search while compensating for the stability of the algorithm itself.Therefore,based on theories such as flexible job shop scheduling,fuzzy theory and dynamic scheduling,this paper addresses the occurrence of equipment failures and the re-scheduling strategy is used in the scheduling window according to the joint influence of cycles and events;as the workpiece delivery period has fuzzy characteristics,this paper represents it using trapezoidal fuzzy numbers and uses dictionary order expressions in the multi-objective optimisation model,setting the maximum completion time as the first level objective,customer and enterprise The overall customer and enterprise satisfaction is set as the second-level objective,and a flexible job shop dynamic scheduling model is established.Then,the improvement process of the grey wolf optimisation algorithm is designed to optimise the initial population of the algorithm and improve the overall algorithm convergence speed;change the convergence factor from linear to non-linear to enhance the ability of the algorithm to expand the search range,avoid falling into local optimum and balance the local optimum with the global optimum;assign different decision weights to the grey wolf leadership at different positions to improve the dynamic weighting of the accuracy of the algorithm.Finally,the convergence performance and global search performance of the improved GWO were verified by combining a production example at the processing site of the DS acrylic sheet factory,and the example was solved to provide a comparative analysis of the production situation with different equipment failures,comparing the maximum completion time and overall satisfaction,and filtering out the critical equipment for key maintenance.Feasible recommendations are also made for the future production operations of the business.