Scheduling For Cluster Tool Based On The Harris Hawks Optimization Algorithm

With the rapid development of science and technology,people's demand for semiconductor components is becoming more and more diverse and the requirements are getting higher and higher.The basic component of semiconductor components is the wafer,Due to the advantages of high efficiency and flexibility,cluster tool is widely used in wafer processing.At present,in the research on cluster tool,more scholars focus on the research on the vacuum side of cluster tool,and there are fewer researches on cluster tool that includes both the atmospheric side and the vacuum side.With changes in demand,the processing mode of wafers has changed from single-variety,large-scale production to today's multi-variety,small-scale production.In the actual production process,it often occurs that multiple varieties of wafers are processed simultaneously in the cluster tool.Therefore,it is very necessary to optimize the scheduling of the processing of multi-variety wafers on the cluster tool with atmospheric side.In this paper,the vacuum side and the atmospheric side are cluster tool with a single manipulator,and the scheduling problem of production efficiency optimization,production efficiency and multidimensional optimization of wafer quality is studied.The main research contents are as follows:(1)In order to better describe the dynamic production process of cluster tool,the production process of cluster tool is described by using Petri nets,and the switching problem of multi-variety wafers in cluster tool,as well as the connection between the atmosphere end and the vacuum side,are analyzed.The time characteristics of the process were analyzed.(2)In order to obtain a feasible and efficient scheduling scheme,the Harris hawks optimization(HHO)is used to optimize the production efficiency of cluster tool.Based on the description of the machining process by Petri nets,considering the optimization problem of the production efficiency of the cluster tool,a nonlinear mathematical model is established,and the model is solved by the HHO algorithm,and the effectiveness of the scheduling scheme is verified through simulation experiments.(3)For more efficient scheduling schemes,the improved Harris hawks optimization(IHHO)optimizes the productivity of cluster tool.The HHO algorithm is easy to fall into local optimum,and the development and search ability of the algorithm are difficult to balance.Therefore,the non-linear convergence factor and tabu search are introduced to improve the HHO algorithm,and an improved hawks optimization is proposed.In the simulation experiment,the effectiveness and efficiency of the IHHO algorithm are verified by comparing the IHHO algorithm with the HHO algorithm and Ant Colony Optimization(ACO).(4)In order to optimize the production efficiency of cluster tool and wafer processing quality at the same time,a multi-objective optimization model of cluster tool scheduling is established,and the multi-objective improved Harris hawks optimization is used to solve it,and finally,the effectiveness of the scheduling scheme is verified by simulation experiments.The above studies show that the cluster tool of the vacuum side and the atmospheric side is a single robot arm,and an improved Harris hawks algorithm is proposed,which can be applied to the mixed-flow wafer production scheduling in the multi-dimensional optimization of production efficiency,production efficiency and wafer quality.