Research On Multi-objective Flexible Job-shop Scheduling Problems Under Uncertain And Dynamic Environment

Manufacturing industry has occupied a more and more important market share in the national economy.In the practical manufacturing environment, however, optimize multiple production objects, uncertain workpiece processing time,uncertain delivery date, frequent random arrival of workpiece and machine maintenance and so on all attract the attention of scholars.Therefore,our article start a research to the multi-objective flexible job-shop scheduling problem,uncertain flexible job-shop scheduling and dynamic flexible job shop scheduling problem into consideration.Firstly, we introduce the source, the research background and the significance of the subject.Outline the classification, main characteristics and major research methods of the flexible job shop scheduling problem. For the multi-objective, uncertain processing time and the dynamic features of FJSP problem,we systematically and deepth analysis the existing primary research methods and research status at domestic and oversea.Then introduce the related optimization theory of multi-objective,build the multi-objective optimization model under statistic conditions and put forward the corresponding optimization algorithm for solving the problem.Secondly, we study the flexible job-shop scheduling problem under the random processing time. First of all, the FJSP problem model containing random time is introduced, then convert the FJSP scheduling problem under uncertainty to determine robust equivalen one with the help of robust optimization theory. Applying the improved differential algorithm to the robust equivalence problem. Through example test, we can verify the effectiveness of our algorithm in solving the flexible job-shop scheduling problem under uncertainty.Aimed at the complexity of dynamic flexible job shop scheduling problem, we build the flexible job shop dynamic scheduling model,which aims to optimize scheduling robustness and stability. Then,a kind of effective differential algorithm with variable neighborhood search method is designed. In this method, during any scheduling cycle, generator would generate dynamic events for the next phase,and at each reschedule point,complete reschedule method and right-shift reschedule method are applied to optimize the problem and produce a new scheduling scheme. In addition, experimental method would be used to explore the influence of the arrival rate and number of new jobs on scheduling system. The results show that the arrival rate and the quantity of new job all have statistically significant impact on scheduling robustness and scheduling stability.Finally, we summarize the work we have done, and prospect the future research direction of this area.