Research On Batch Planning And Scheduling Of Rare Earth Magnetic Materials Based On Nsga-Ⅲ Algorithm

As an important industrial core material,the demand for rare earth permanent magnetic materials in terminal applications is increasing with the deepening of the "double carbon" goal and the continuous deepening of intelligent manufacturing.Rare earth permanent magnet materials are an order oriented "multi variety,small batch" production mode,in which smelting is the first process,with high cost and closely related to the performance of the final product.Therefore,the production organization method in the smelting process usually adopts the grouping-furnace production technology based on the combination of multiple production orders.The results of furnace assembly are quite important for the production of rare earth permanent magnet materials,which directly affects the production scheduling optimization of products in multiple stages and processes.If the result of furnace assembly is not ideal,it will lead to excess inventory,occupation of working capital,reduction of on-time delivery rate,and increase production costs.Reasonably organizing and optimizing the scheduling of furnace production while meeting customer needs is not scientific based on experience alone.How to make full use of production resources and develop reasonable and effective furnace production plans has become an urgent issue for such enterprises to solve,and is also the key to achieving cost reduction and efficiency increase and improving industry competitiveness.Based on this,the batch planning and scheduling of the rare earth permanent magnet material melting section is studied,and the specific work is as follows:(1)Firstly,the research background and significance of the batch planning and scheduling problem of rare earth permanent magnet melting furnace are described,and the importance of optimizing the batch planning and scheduling problem is emphasized.(2)Secondly,aiming at the problem of batch planning in the smelting section of rare earth permanent magnet materials,combined with the production process constraints and the characteristics of the production process,an optimization model of the work order group furnace considering the total number of heats,brand deviation and delivery date difference was established.At the same time,the improved NSGA-III algorithm based on adaptive penalty distance is used to solve the problem,and two coding methods and genetic operations are designed.Finally,the coding methods and the improved algorithm are compared to verify the effectiveness of the algorithm.(3)Finally,in order to avoid the problem that batch plan and scheduling scheme sequential decision-making can’t achieve the global optimization,the objectives of the two stages of batch plan and scheduling are comprehensively considered,and a two-level algorithm is designed to solve the batch plan and job scheduling of the furnace.The outer layer uses NSGA-III algorithm to determine the batch plan of the furnace,and the inner layer is equivalent to the decoding function,The improved NSGA-II algorithm is used to optimize the heat plan and generate the corresponding scheduling scheme.From the obtained non-inferior solution set,the SMAA-2 method is used to select the most acceptable scheduling scheme as the optimal scheduling result of the heat plan.The non-inferior solution set is obtained through comprehensive optimization by calling the inner algorithm from the outer layer to achieve the overall optimization.Finally,an example is given to verify the effectiveness of the proposed two-level algorithm.