Research On Joint Optimization Of Semiconductor Production And Maintenance Planning Based On Clearing Function

The complexity of the semiconductor wafer fabrication manufacturing process and expensive equipment make the production organization and equipment maintenance schedules become a very important activity.Most existing methods separately arrange these activities,so that production arrangements tend to lack the feasibility or hard to implement.With many re-entry times of wafers and the load fluctuating and breakdown,it is easy to excessive maintenance or lack of maintenance because of no consideration of the impact of production.And this will affect the feasibility and effectiveness of the production plan,or even affect product quality and increase production costs.Therefore,this thesis researches on the production of joint optimization for semiconductor wafer fabrication of equipment maintenance and production lot planning,in the full study of the characteristics of a semiconductor wafer fabrication process,designs production schedule method,combined with the relationships between production planning and equipment maintenance,sets the model and gets integrated optimization solution.In order to explore in semiconductor production systems,a series of experiments under different conditions of maintenance and equipment failure was taken and this paper designed a scaled-down production simulation model based on a real semiconductor production workshops.With the typical characteristics of semiconductor production: the uncertainty of production cycle,different equipment status,corrective maintenance situation,Clearing Function parameters were obtained by the simulation.Subsequently,the paper compares fixed lead time(FLT)linear programming model and Clearing Function(CF)model under different failure modes,equipment utilization of bottlenecks and different demand forms by the planning results.To inquiry clearing function model’s superiority compared to other models,the CF is able to capture the actual production of semiconductor production capacity is verified.It is a better solution to solve production planning problems because of the uncertainty of production in cycle the semiconductor manufacturing industry.Finally,considering the interaction of production planning and equipment maintenance,adding CF constraints,the model of joint optimization of production and maintenance planning is established for the semiconductor manufacturing system.The improved simulated annealing algorithm is designed for solving and the validity is verified.