Font Size: a A A

Research On Flexible Job-shop Scheduling Optimization Method Based On Co-CEM

Posted on:2023-08-15Degree:MasterType:Thesis
Country:ChinaCandidate:X Z JiangFull Text:PDF
GTID:2558306902980489Subject:Computer Science and Technology
Abstract/Summary:PDF Full Text Request
At present,the production and manufacturing processes of most domestic shipping companies are still in the semi-automated scheduling stage.In most cases,dispatchers schedule production according to the current idle state of resources and experience,which causes unreasonable arrangement of scheduling plan,and affects the production and manufacturing efficiency,critically.So that solving the Flexible Job-shop Scheduling Problem(FJSP)in shipbuilding is the key to improving shipbuilding efficiency.First of all,Evolutionary Algorithms are mainly used to solve FJSP currently,which have deficiency in local search and inefficient improvement of individual.Secondary,Cross-Entropy Method(CEM)is rarely used to solve FJSP,and most studies have not combination of related neighborhood functions.In response to this problem,this paper designs a Cooperative coevolution-based CEM(Co-CEM),and the experiment proves that it has high efficiency in solving FJSP.The main work of the thesis is as follows:(1)For FJSP,The Cooperative Coevolution-based CEM is proposed.First of all,the coevolution strategy makes up for the problem that CEM has strong global search ability but inefficient in local search.Next in importance,The Active Scheduling-based Genetic Decoding Algorithm is proposed,in which the active scheduling operation guarantees that the solution obtained is optimized,and the genetic operation saves the relevant scheduling information in the gene,which improves the search ability of the algorithm,effectively.(2)Aiming at the search abilities of Co-CEM in MOFJSP,a hierarchical multi-objective search strategy is proposed,which enables of further cooperation between the global search and local search.Furthermore,this strategy considers the optimization of multiple targets during local search phrase,which enhances the population’s diversity and convergence.Finally,SPEA2 based environment selection operator and external memory strategy are introduced to ensure the feasibility of Co-CEM under MOFJSP.
Keywords/Search Tags:Combinatorial optimization, Flexible job-shop scheduling, Evolutionary algorithm, Cross-entropy method, Multi-objective optimization
PDF Full Text Request
Related items