Font Size: a A A

Research On Job Shop Scheduling System Of MES Based On Hybrid Genetic Algorithm

Posted on:2010-11-03Degree:MasterType:Thesis
Country:ChinaCandidate:C H CaoFull Text:PDF
GTID:2178360275954749Subject:Power electronics and electric drive
Abstract/Summary:PDF Full Text Request
Manufacturing Execution System(MES) is an information technical system designed for the workshops management.It helps to integrate the systems of administration and the workshops and to improve the efficiency of the enterprises.For discrete manufacturing,the shop scheduling problem is the core of MES.The optimization of shop scheduling problem is the most valuable technique for enterprises to reduce expenditure and promote the efficiency.Therefore,the study of MES,which is based on advanced theory of scheduling,would have significant effect on improving the competitive power of the enterprises.First of all,this dissertation provides an overview on the basic concepts of the shop scheduling problem,including the problem definition,the performance metrics,and the problem classification,etc.The Job-shop Scheduling Problem(JSP) and Flexible Job Shop Scheduling Problem(FJSP) are studied thoroughly,and the mathematic models of JSP and FJSP with different scheduling objects are presented.Then,to overcome the premature convergence and the lower convergence speed,a novel Self-adaptation Cultural Genetic Algorithm(SCGA) bases on the evolutionary idea of Cultural Algorithm is proposed.According to the experience knowledge in knowledge space,the individuals in population implement genetic process,which can reduce the randomicity of the algorithm and improve the convergence speed.To solve the JSP which is a typical NP-hard problem,SCGA adopts the operation-based representation,the adaptive crossover operator and mutation operator. Meanwhile,the update scheme of knowledge is designed.For the more complex FJSP,the two-layer encoding scheme based on operation and machine distribution is designed to improve the crossover operation,the mutation operation,and the update scheme of knowledge space.The simulations are conducted for a set of FJSP and JSP benchmark problems,where Makespan is as optimal object.Experiment results show that SCGA is effective to solve JSP and FJSP.Finally,taking a mechanical manufacture workshop of an elevator company in Shandong province as example,the manufacture characteristic of the workshop is analyzed deeply,and the function frame of MES is studied.Then,according to the practical situation of the workshop,a shop scheduling system is developed.The system adopts Visual Basic 6.0 as the tool for foreground application development and SQL Server2000 as the background database,and integrates the proposed SCGA.The system implemented has effectively solved the shop scheduling problem.
Keywords/Search Tags:MES, Job-Shop Scheduling, Flexible Job-Shop Scheduling, Genetic Algorithm, Cultural Algorithm
PDF Full Text Request
Related items