Research On Process Oriented Production Scheduling In Cloud Manufacturing Environment

The proposal of "Made in China 2025" has deepened the integration of information technology and manufacturing technology and formed the commanding heights of a new round of industrial competition.As a typical representative of the integration of information and industrialization,cloud manufacturing has become one of the important contents of the "Made in China 2025" strategic planning.However,there are still many problems in the application of cloud manufacturing related theories and research in scheduling.In consideration of this situation,this dissertation is based on the relevant research results at home and abroad,combined with the production process,to explore the process oriented production scheduling problem in the cloud manufacturing environment.Through research on the key technical issues such as production task decomposition,inter-enterprise production scheduling,and workshop-level scheduling in the cloud manufacturing environment,a scheduling optimization framework has been established to meet the scheduling requirements of the entire production process in the cloud environment.The specific research work and results were as follows:Based on the key technologies of cloud manufacturing and the characteristics of cloud manufacturing scheduling,this paper constructed a production scheduling optimization system for downstream processes in cloud manufacturing environment based on the attributes of scheduling in the cloud manufacturing environment,and studied the key technical issues in the established optimization system.For the cloud manufacturing task decomposition problem,established a cloud manufacturing task decomposition optimization model.The idea of BOSS tree was introduced into the task decomposition of production scheduling in cloud manufacturing environment,and an optimization algorithm of task decomposition based on boss tree was proposed.Firstly,the related measurement methods of cloud manufacturing tasks were studied.Secondly,based on the principle of cloud manufacturing task decomposition,the decomposition optimization model considering the interaction between cloud manufacturing tasks was established.Thirdly,according to the BOSS tree idea,a task decomposition optimization algorithm including the whole life cycle of production task was proposed,and the heuristic rules were used to optimize the algorithm,which effectively solved the disjunction problem between cloud manufacturing task decomposition and service task matching.Finally,the feasibility and effectiveness of the algorithm were verified through an example.An improved genetic algorithm(NSGA-II),which was based on non-dominated sorting with elitist strategy and crowding degree,was proposed to solve the multi-objective production scheduling problem of inter-enterprise in cloud manufacturing environment.Firstly,the cross enterprise production mode was analyzed;secondly,the mathematical model of inter enterprise scheduling in cloud manufacturing environment was established;thirdly,the hierarchical coding of multi-layer two-dimensional matrix was designed,and the NSGA-II algorithm was improved by adaptive evolution strategy based on congestion degree;finally,the effectiveness of the improved model and algorithm was verified by an example.For the production of workshop level,the workshop scheduling problems of simultaneous produced cloud task and local task and that of only produced cloud task were respectively studied.For workshops with simultaneous produced cloud task and local task:firstly,the mixed-flow workshop scheduling problem encountered in this production mode was analyzed and modeled;secondly,the characteristics of differential evolution algorithm and genetic algorithm were compared,and the advantages of genetic algorithm for effectively dealing with discrete variables and differential evolution algorithm for effectively dealing with continuous variables were merged,and a new method based on differential evolution algorithm was proposed according to the production status.Finally,the feasibility and validity of the hybrid genetic algorithm were verified by an example.For workshops with cloud tasks only: firstly,a hybrid workshop scheduling model aiming at minimum total completion time was established according to production problems;secondly,an improved hybrid immune cloning selection genetic algorithm was proposed,the process of antigen recognition,antibody coding and decoding of the algorithm was reconstructed,the affinity function was also reconstructed,and the hybrid operations of cloning,mutation,crossover and selection of antibody groups in the algorithm were also carried out.At last,two simulation experiments were used to verify the reliability of the new algorithm.According to the scheduling optimization system in cloud manufacturing environment proposed in this paper,the system development environment,MATLAB program integration and data service process were researched.Combining with the research on three key technical issues of cloud task decomposition,inter-enterprise scheduling and workshop scheduling,the functional modules of the system were designed.Finally,the prototype system is developed.