Research On Scheduling Problem For Mixed Production Of Controller For Electric Vehicles Of Company A

In recent years,the sharing economy has emerged,and the cross-border product of shared electric vehicles has emerged at the historic moment.With the support of capital,various companies have entered the market and the supply chain of electric vehicle products has begun to reorganize.As a manufacturer of electric vehicle controllers,company A has also become a member of the shared electric vehicle supply chain in the past two years.The order products are divided into four types: X1,X2,X3,and X4.With the intensification of competition among sharing companies,company A faced the challenge of large fluctuations in customer orders and the gradual decrease in order prices,and various problems still existed in the company's mixed production site.Faced with these situations,the company urgently needs to optimize the mixed-flow production line without adding additional costs.Based on the actual situation of the electric vehicle controller produced by company A,this thesis studies the scheduling problem of the company's mixed-flow production method.First,analyze the problems of the mixed production line of the electric vehicle controller of company A,determine the two production stages that need to be optimized,and give the overall scheduling optimization plan: first split the order;then optimize the production scheduling process of the four types of controllers in the second production stage to clarify the production time of the four types of controllers at this stage,especially the X4 controllers with multiple varieties and small batches and process constraints;finally,based on the production time of the four types of controllers,the production sequence of the four types of controllers in the two production stages is scheduled and optimized to reduce the total completion time and the waiting time of semi-finished products.Then,taking the production process of the X4 controller in the second production stage as the research object,randomly select an order from the company as an example,first split the order,and then establish a scheduling optimization model based on the TOC: identify bottlenecks,establish a two-stage mathematical model to use bottlenecks as much as possible,make other processes cooperate with the bottleneck,and finally use order data to solve the two-stage mathematical model to verify the validity of the model.The results show that: using TOC and DBR method to schedule the production line can effectively identify the bottleneck,make full use of the bottleneck,and make other processes cooperate with the bottleneck.For multi-variety small batch products with process constraints,when the mold resources are insufficient,the optimization of mixed production methods can reduce the constraints of mold resources.Finally,based on the production time of each product,a hybrid flow shop scheduling model in two production stages is established.According to the specific problems of the company,with the minimum objective of minimizing the maximum completion time and minimizing the waiting time of semi-finished products,a 0-1 integer programming model is established,and a real-coded genetic algorithm is designed based on the weighted sum method to solve the model,and the impact of target weights is analyzed.It is found that the real-coded genetic algorithm has good searchability and the scheduling optimization result is sensitive to the target weight.The results show that: Based on the current status and demand of the production line,the A company should determine the importance of the objective function and draw a scheduling plan to reduce the waiting time of semi-finished products or completion time.The research results of this paper enrich the research on production scheduling problems facing process constraints,increase the application scenarios of hybrid flow shop scheduling problems,and provide a new scheduling ideas for A company and other companies producing similar electronic products in mixed-flow production methods and product scheduling.The thesis has 18 figures,17 tables,and 70 references.