Research On Scheduling Problem Of Semiconductor Wafers Distributed Green Manufacturing

As the world’s largest semiconductor consumer,China accounts for more than 50%of the global demand for chips.But at present,Chinese enterprises are still in the low-end field in the chip industry chain,such as design and manufacturing,and can only complete the independent manufacturing of some low-end chips,but cannot achieve the self-sufficiency of high-end chips.In the face of foreign "core sealing" incidents,the Chinese government attaches great importance to promoting the transformation,product upgrading and technological innovation of domestic semiconductor enterprises.Compared with the shortcomings of traditional centralized manufacturing,such as single product variety,high cost,easy inventory backlog and lack of cooperative innovation,distributed manufacturing undoubtedly set off a wave of manufacturing model reform and innovation.Distributed manufacturing can help companies better fulfill their commitments to customers in terms of product quality,cost,and delivery time,and strive for more competitive advantages for themselves,which has become the development trend of many companies.According to statistics,China’s carbon emissions will still rank first in the world in2019.The Central Economic Work Conference in December 2020 proposed to do a good job in carbon peak and carbon neutral.This has pointed out the goal and direction for China to save energy and reduce emissions and take the green and low-carbon development path.As a pillar of the economy,the semiconductor industry is a big energy consumer,and reducing energy consumption is also an effective way to reduce costs and enhance its competitiveness.Production scheduling can effectively optimize product manufacturing process,reasonable allocation of resources,improve the efficiency of equipment utilization,thereby reducing energy consumption emissions accordingly.As one of the important combinatorial optimization problems,it is difficult to solve.The semiconductor wafers manufacturing enterprises pursue the balance of economy,benefit and green indicator,which is actually a multi-objective optimization problem of reentrant hybrid flow shop scheduling.At present,in addition to considering the impact of different scheduling schemes on optimization indicators,it is also a challenge for enterprises to explore more energy saving and emission reduction methods.Therefore,the scheduling problem of semiconductor wafers distributed green manufacturing is studied in this paper.The main work of this paper is as follows:(1)Aiming at the optimization problem of semiconductor wafers distributed green manufacturing,a mathematical model of reentrant hybrid flow shop scheduling is established to minimize the makespan,carbon emissions and total tardiness.According to the problem characteristics,the assignment rule of jobs-to-factories is designed,and an improved multi-objective grey wolf algorithm is proposed to solve the problem.Finally,the reentrant standard test cases verify that the proposed algorithm is superior to other comparison algorithms in solving this problem.(2)The joint optimization problems of single-stage and two-stage semiconductor wafers distributed green manufacturing and preventive maintenance is discussed successively.A reentrant hybrid flow shop scheduling model is established to minimize the makespan,total carbon emissions and total maintenance costs.A synchronous scheduling and maintenance strategy suitable for reentrant hybrid flow shop manufacturing environment is designed,and the strategy is also extended to consider the flexible constraints of workers.Through simulation experiments of randomly generated large,medium and small scale test cases,it is verified that the proposed hybrid multi-verse optimizer algorithm and grey wolf algorithm have certain advantages in the distribution,convergence and diversity of the obtained Pareto solution set.The effectiveness analysis of the synchronous scheduling maintenance strategy shows that it is in line with the actual production situation of semiconductor wafers manufacturing.Also it is suitable for the multi-factory distributed manufacturing mode,which is conducive to the realization of energy saving and emission reduction goals.The case analysis experiment of the actual semiconductor production line HP24 model shows that the application of distributed manufacturing in semiconductor wafers manufacturing enterprises has certain practical significance and value,and can effectively optimize carbon emissions and makespan indicators.(3)The distributed green manufacturing scheduling problem of semiconductor wafers considering distributed energy resources(DERs),energy storage system(ESS)and multi-energy complementary cooperative power supply are studied respectively.Firstly,it is verified that the use of renewable energy can effectively reduce the carbon emissions of production workshop through comparative experiments of three energy strategies under time of use(TOU),which are only consider the power supply from ordinary main grid,the cooperative power supply between ESS and ordinary main grid,and the cooperative power supply between DERs,ESS and ordinary main grid at the same time.On the basis of this research,uncertainty constraints and the PM operation of the equipment have been added.An improved interval number probability calculation method is proposed to solve the problem,and an energy allocation strategy considering renewable energy,natural gas energy and ordinary main grid is designed at the same time.Through different energy allocation schemes comparative experiments,it shows that the multi-energy complementary cooperative power supply mode can effectively optimize the energy consumption and energy consumption costs.Through the randomly generated different scale test cases and the case analysis experiments of the actual semiconductor production line Mini-Fab model,it shows that the proposed hybrid salp swarm algorithm and the improved whale swarm algorithm can effectively solve the corresponding problems,and has a certain practical value and significance.The main innovations of this paper:(1)To expand and enrich the research ideas of this type of problem by improving the design of the new intelligent optimization algorithms and applying them to the distributed semiconductor wafers manufacturing RHFS scheduling problem that considers the manufacturing stage and the inspection and repair stage at the same time.(2)Under the background of distributed semiconductor wafers manufacturing,the collaborative optimization of enterprise efficiency,economic benefits and green indicator among heterogeneous factories is carried out.By discussing the joint optimization of jobs scheduling and equipment preventive maintenance,the equipment degradation and forgetting effect caused by long-term continuous production can be alleviated,the extension of jobs processing time can be avoided,and the energy consumption and carbon emissions can be reduced accordingly.(3)By discussing the multi-energy complementary cooperative power supply mode such as using renewable energy under the TOU electricity price in distributed manufacturing environment,the goal of energy saving and emissions reduction has been achieved effectively,which provides an important reference value for enterprises to transform to green manufacturing mode.