Research On Service Supply-demand Matching For Machining In Cloud Mode

Cloud manufacturing provides on-demand manufacturing resources and manufacturing capabilities to users through network and the cloud platform,which improves the utilization of manufacturing resources and also meets the diverse needs of users.However,it is difficult to match the service demands quickly with rich varieties and huge quantities of manufacturing services on the cloud platform.Therefore,in order to improve the efficiency of service matching and achieve a satisfactory cooperation effect between both parties,machining manufacturing services matching under the cloud mode was researched in this thesis.Mainly studied contents are as follows:The framework of service supply-demand matching for machining in cloud mode was established.The service unit for matching was determined and the specific content and process of service matching were illustrated firstly.Then,models of cloud machining service and service requirement were described.Moreover,clustering algorithm,the extension theory,variable precision rough set were introduced as theoretical bases,which were applied to different stages of matching respectively to achieve service matching efficiently in cloud manufacturing and meet demands of users and service providers.The similarity-based clustering method for cloud machining services was proposed.The traditional K-means algorithm with the initial cluster centers randomly and the number of clusters in advance was improved,and the similarity-based clustering algorithm was proposecd.Similarities between services were calculated through the attribute description of the service processing capability.Machining services with high similarity were clustered together finally to form different cloud service sets.The extension theory-based method for cloud machining service set selection was proposed The processing capability attributes of service requirement were determined.Then,service requirements and cloud service sets were formally described by the extension theory-based matter-element models.The correlation between the service requirement and the cloud service set was quantified by the correlation function,which was basic in the cloud service set selection.And the machining service candidate range that meets the user's demands was obtained ultimately through service filtering and service attribute values determination.The two-side evaluation method for cloud machining services was designed.Combining with the demands of service providers,the one-side evaluation based on service QoS was expanded by adding user evaluation indicators and the two-side evaluation model of machining cloud services was established.Then qualitative indicators were quantified and service QoS evaluation values and user evaluation similarities were calculated.The machining service that was the most suitable for users was selected by ranking candidate services.This method not only considers the demands of users,but also protects the benefits of service providers and enhances the satisfaction of both parties' cooperation.A prototype system of supply-demand service matching for machining in cloud mode was designed and developed.The functional modules of the prototype system were analyzed and the workflow diagram was given.The system was developed by using Java programming language,MySQL database,etc.,which implemented the whole process of cloud service matching and verified the validity of the models and methods proposed in this thesis.