Research On Many-objectives Schduling Optimization In Network Manufacturing System Based On Improved PCA-NSGA? Algorithm

Network technology and information technology are developing steadily ceaselessly under the background of economic globalization.Enterprises begin to cooperate and share resources to improve their competitiveness in the market.This has led to the emergence of the network manufacturing mode that enterprises will cooperate and share their sources and tasks in the network environment.There are many enterprises in the network manufacturing system,and the requirements of each enterprise are not exactly the same.In order to meet all enterprises' production requirements at the same time,it do need to be considered many objectives in scheduling optimization,so most scheduling problems in the network manufacturing system relate to many objectives scheduling optimization.So,it is necessary in both theory and practice to study the optimization problem of many objectives scheduling for the advanced manufacturing mode.Besides,as green sustainable manufacturing is being paid more and more attention,it has great significance for environment and manufacturing industry to consider low carbon emissions in the many objectives scheduling problem.So it's explored how to solve many objectives scheduling optimization and reduce carbon emissions in network manufacturing enviroment in this paper.After a comprehensive overview of network manufacturing and then the comprehensive demand of all enterprises,the objective of scheduling optimization and the importance of green manufacturing have been considered,a many objectives scheduling optimization model is proposed to minimize total production cost,makespan,earliness and tardiness(E/T)penalty and carbon emissions in manufacturing processes.A dominant strategy based on principal component analysis(PCA)is presented to deal with the problem that the selection and search ability of the traditional multi objective evolutionary algorithm based on Pareto sorting will be greatly weakened with the increase of the objective when the algorithm is used to solve the many objectives scheduling optimization.At the same time,an external population is added to improve the algorithm's elite strategy for enriching the structure of the population and increasing the convergence rate of the algorithm,.An improved PCA-NSGA ? algorithm is designed to solve the proposed many objectives scheduling optimization problem.On the basis of the proposed model and the actual manufacturing processes,a case study is designed and the improved PCA-NSGA ? algorithm is adopted to get the results.Results of the proposed model are compared with results obtained from the traditional NSGA ? algorithm and the NSGA? algorithm based on the epsilon dominant.The conclusion is to verify the effectiveness and rationality of the proposed algorithm for the model.And It showed that the PCA technology can alleviate the problem that the large number of non-dominated solutions leads to the reduction of algorithm selection ability when the traditional multi-objective evolutionary algorithms are used to solve the many objectives scheduling problem.Then the results of the improved PCA-NSGA ? algorithm for case study are compared with the results of the traditional PCA-NSGA ? algorithm,which shows that the convergence rate of the improved PCA-NSGA ? algorithm is better.