Research And Application Of Stamping Shop Scheduling And Assembly Line Sequencing Method In Automobile Mixed-model Production

In order to satisfy the diverse requirements from the market,the kinds of products in the automobile mixed-model production lines become more and more.However,product transition in the mixed-model production often leads to setup time,production time and the change of the material requirements,the complex production process makes the workshop need more meticulous production plan.The automobile production mainly includes the stamping,welding,painting and assembly shop.Based on the adjacent workshop production plan,each shop needs to develop a suitable production plan to improve the production efficiency and resource utilization,reduce the manufacturing cost and shorten the order completion time.The traditional manual production plan based on the engineer's experiences is hard to achieve the reasonable vehicle production plan.The vehicle production enterprises urgently need a scientific and efficient production planning optimization system for vehicles.Therefore,this dissertation takes the vehicle production shop in an automobile company as application background,the major production plan in each shop as core research and the intelligence optimization algorithm as supporting technology,the optimization issues of final vehicle production planning and scheduling optimization problems are researched.Based on the overall research of the stamping shop scheduling and assembly line sequencing method in the automobile mixed-model production,three kinds of planning and scheduling problems are investigated as follows: the stamping shop scheduling problem,the car sequencing problem in welding,painting and assembly shop,the associated workshop car sequencing problem.Finally,this paper implements the construction and validation of prototype system for automobile mixed-model planning and scheduling,and part of the models and algorithms have been applied in the MES project for a typical enterprise,which has obtained the good effect in practical application.The dissertation focuses on vehicle production processes and analysis the problem characteristics,mathematical models,optimization algorithms,practical application,system construction and validation etc.for each production shop in detail,which includes:(1)According to the overview of the vehicle production process,this dissertation discusses the characteristics and difficulties in automobile mixed-model production.Based on in-depth study of automotive stamping shop scheduling problem,the car sequencing problem and the associated workshop car sequencing problem,after making some improvements and innovations for the models of workshop planning and scheduling and optimization algorithms,this paper puts forward overall research scheme for the stamping shop scheduling and assembly line sequencing method in the automobile mixed-model production.(2)Aiming at the characteristics of flexible production in stamping shop,set up the scheduling model of stamping shop based on flexible job-shop scheduling and select the minimum of workshop production cost and the minimum of makespan as the optimization target.Based on the improvement on the standard artificial bee colony algorithm,this dissertation presents a multi-object hybrid artificial bee colony to solve the multi-object flexible job-shop scheduling problem and uses the flexible job-shop scheduling problem example to test the MHABC algorithm.In the case study of the stamping shop production plan in an automobile company,this paper employs MHABC algorithm to find the Pareto optimal solutions.The final optimal stamping shop plan can be achieved by selecting the machines for each process of each workpiece,which reduces the workshop production cost and ensures the efficient production of stamping shop.(3)In allusion to the characteristics of welding,painting and assembly production process,the mixed-model production planning problem can be conducted to the car sequencing problem and the assembly shop is selected to research.According to the actual production needs of the automobile assembly line,after subdividing the assembly production load balance,this paper establishes a multi-objective car scheduling problem with the production load balance constraints,which regards the workstations load balance and the similarity of assembly products as the target.In combination with the characteristics of small-world network structure,this paper first proposes a new fast scheduling method for mixed-model production: the discrete small-world optimization(DSWO)algorithm.Then this issertation applies the standard flow shop problem to validate the correctness and effectiveness of the DSWO algorithm.After designing a Pareto discrete small-world optimization(PDSWO)algorithm to solve this multi-object car sequencing problem,the proposed approach is employed to optimize the assembly vehicle sequence in an automobile company to keep the assembly workshop production load equilibrium.(4)The model of associated workshop planning problem is proposed based on the limited storage between the adjacent workshop,which combines the continuity characteristics of the production process in the vehicle manufacturing.After studying the painted body storage(PBS),the multi-object model of PBS sequencing problem is proposed in order to minimize the consumption rate of options and minimize the total overtime and the idle time of workstations.In the issue of single PBS sequencing problem,a two-phase optimization algorithm combined with the DSWO algorithm is presented.At the same time,this paper has carried on the detailed discussion for the multi-stage PBS sequencing problem.In the solving of two-stage PBS sequencing problem,a three-phase optimization algorithm combined with the DABC algorithm is proposed.Then,the proposed algorithms are applied to the single PBS and two-stage PBS sequencing problem in a case company,the results validate the correctness and effectiveness of both methods.These methods optimize the PBS sequencing process and ensure the smooth of load balance and the homogenization of material consumption for each workstation.(5)According to the vehicle production process and the planning optimization model of each production workshop,the vehicle production planning and scheduling prototype system is designed and the functions are divided.The system takes into account the characteristics of each production workshop planning optimization and ensures that each production workshop can build the configurable planning model under the premise of considering the related workshop plan optimization.The prototype system is applied into an automobile company to validate the effectiveness of the vehicle production planning system by scheduling the production plan with the real production data.Finally,part of the models and algorithms are embedded into the planning and scheduling module in the Manufacturing Execution System and the on-site deployment in an automobile enterprise has obtained a satisfied effect in practical application.