Reserrch On Intelligent Scheduling Optimization Method Of Production Line For Cloud Service

As a pivotal role,it is a significant task for the 13 th Five-Year Plan to promote the upgrading and strengthen the high-quality development of manufacturing industry.With the proposal and development of intelligent manufacturing and cloud manufacturing,the intelligent manufacturing systems and the sharing and integration of manufacturing resources have become new methods in the transformation and upgrading of manufacturing industry.At this rising stage,manufacturing companies cannot break through industry barriers among enterprises,and the massive historical data of production scheduling cannot be effectively managed and utilized.The attempts and promotion of cloud service platform in the manufacturing industry have coordinated and shared various types,multi-dimensions and wide-time domain production control data and scheduling methods,laying a foundation for the optimization of jobshop-level production line scheduling methods.In a word,in order to improve the utilization of massive historical resources of equipment,production plans and knowledge methods,this paper studies production line intelligent scheduling optimization method based on cloud service.First,in the context of cloud service platforms and industrial big data,combining the advantages of the proposed intelligent scheduling model and the proposed "cloud-side-end" collaborative computational model,the "initial scheduling-rescheduling" model of production line scheduling for cloud services is proposed;ARENA production system simulation software is used for the idea of hierarchical modeling to establish a digital model of the manufacturing jobshop.Secondly,research on the cloud service method of scheduling resources in production planning tasks to solve the analysis and utilization of massive historical data under the cloud service platform.The data is expressed based on Extensible Markup Language(XML),which provides a unified and standardized way for integrating various types of enterprise data.In order to obtain the required target production resource attributes,a vector space model(VSM)based on text feature extraction,comprehensively integrating the equipment attribute,process characteristics and user needs of production scheduling resources,constructs a historical scheduling scheme selection model.An algorithm for semantic matching and similar optimization of cloud historical data and cloud platform user needs is designed to obtain a better production scheduling scheme.Thirdly,the problem of guiding production scheduling at jobshop level based on historical scheduling scheme of cloud platform is studied.Combined with the introduction of genetic algorithm in the historical scheduling scheme,an initial population optimization method for the scheduling problem of mixed-flow jobshop was proposed,and the model was solved with the goal of minimizing the maximum completion time.Reduce the impact of user requirements on the actual constraint deviation at the jobshop level,and improve the adaptability of the scheduling plan to guide actual production.Finally,a cloud service production scheduling system is constructed,and the production scheduling optimization method in the paper is embedded into it to realize the use of this cloud service production scheduling system from the user’s perspective to complete the production scheduling program that meets the corresponding needs,and guide the production line in the jobshop layers for actual production scheduling.