Research On Multi-flexible Job Shop Scheduling Problem Based On Genetic Algorithm

With the advancement of technology and information technology,thecompetition between enterprises will be more drastic. Many sequences andsmall batch manufacturing model,which reflect changes of marketdemand,become the trend of business development. Accordingly, the advancedmanagement concepts and efficient scheduling systems have becomeincreasingly important.Compared to job shop scheduling(JSP), the Flexible Job Shop Scheduling(FJSP) has important theoretical and practical significance which is more inline with actual production conditions. The genetic algorithm(GA) is animportant method to solve the flexible job shop scheduling, showing theefficient performance. In this paper, single and multi-flexible FJSP are studiedin depth espectively by using improved GA.Firstly, the classification, characteristics, significance and research status athome and abroad of JSP are briefly introduced, and the workshop evaluationand research methods commonly used of JSP is given.Secondly, the model ofthe partial path FJSP is established. According to the characteristics of themodel, an improved GA is applied, a new twin substring encoding is designed,and the corresponding crossover and mutation methods are given.Thirdly,according to the characteristics of the process flexibility in the actualmanufacturing process， the problem model is presented based on themulti-flexible JSP. The solution of the model is designed by using processposition and the follow-up process constraint set. On this basis, two piecesencoding scheme is proposed based on machine selection and flexible process,and the original individual is constructed at random. An order crossover modelof GA is proposed, meanwhile the corresponding crossover and mutationstrategy is designed. At the last,through the simulation results，the feasibilityand efficiency of the model and the algorithm proposed in the paper can beverified．...