Job Shop Scheduling Research For Intelligent Manufarturing

Intelligent manufacturing can partially replace the mental work of human beings in the manufacturing process,making production intelligent,efficient and personalized,which is the development trend of the future manufacturing industry.Intelligent manufacturing requires intelligent and transparent production processes.However,the manufacturing processes of many manufacturing companies in China today are far from meeting this requirement.Although many companies in the Chinese manufacturing industry use ERP to plan production and distribute production content to each shop,the production plan completed by ERP is still not specific enough.When the production plan is specific to the shop floor,many companies'production plans are still done manually,and this is especially true for manufacturing companies that do not use ERP.Such manually completed production plans are not efficient enough,resulting in high machine idle time,and the production plan is not transparent enough to cause the company to accurately estimate defects such as production schedule.In order to solve the above problems,enterprises urgently need a workshop-level scheduling technology.Therefore,the problem of shop scheduling has become a hot topic in the manufacturing industry.This paper first investigates the research status of job shop scheduling problems at home and abroad,introduces common models of job shop scheduling problems and common solutions to problems.Then learn about common machine learning methods.In the genetic algorithm for solving the scheduling problem of the shop,the problem of solving the large space of the shop scheduling problem and the defect of the genetic algorithm searching ability is not strong and easy to premature.By using neural network to record the information in the iterative process of genetic algorithm,the evolution direction of the population in the genetic algorithm is guided,the search direction of the genetic algorithm is optimized,and the quality of the solution is improved.For the two classifications of shop scheduling problems,two improved algorithms are used to compare the MT06 and Brandimarte data examples.Compared with the traditional algorithms,the algorithm improves the premature defects of genetic algorithms and has higher solution quality.Compared with the other 10 algorithms in the field,the best solution is obtained in 6 cases.