Mining Of Resource-Service Temperal Composition In Cloud Manufacturing

Cloud manufacturing is a new pattern based on network, and “Manufacturing is Service” is its central idea. Select the combination of resource-service means that in the cloud manufacturing systems, for a multi-resource service demand task, the systems must select resource-service that meet the needs of performing this task from a large number of resources. As the amount of manufacturing resources in cloud manufacturing is huge and widely distributed, for business process, reasonable resource-service composition can improve the efficiency of resource selection and utilization ratio. In Cloud manufacturing, a business process is completed by a number of organizations that work together, therefore it becomes more difficult to analyze resource composition. It is more reasonable to analyze the service time series combination of resources if taking into account the overall efficiency of the business process, because it represents the scheduling and service of tasks in the workflow model. Our resources are called service order composed of temperal sequence known as a resource service chain. And resource services invoked in sequential order are called the resource-service chain(RSC). The method proposed composition mainly analyze from the spatial and temporal of resource-service, considering the dependencies between resource-service and timing of resource services, then, to combine resource-service reasonably.According to the research background and study purpose mentioned above, this paper studies the following aspects:1. The formal definition of manufacturing resource service and resource classification is given in this paper. And resource service model is proposed through the analysis of manufacturing resources classification.2. The resource classification based on correlation degree in cloud manufacturing environment. Analysis the correlation degree between resource-based tasks in the build-time stage of workflow model, and divide the resource-service into a number of resource candidate sets according to the degree of correlation between the resources. Combined task timing relationship workflow model, Excavate initial resource service chain based model phase.3. The combination of resource-service timing when the workflow is running. In the runtime stage of the workflow, mining resource service chain through the analysis of workflow historical data. First of all, put forward three evaluation index of resource service chain similarity, according to the characteristics of the resource service chain. Then, calculating the similarity according to the three indexes betweenall the resource-service services chain, combined with initial resource chain obtained in the first phase, and clustering historical data.Finally, taking product design, manufacture and assembly as example, the proposed method is used by research and development of small and medium-sized enterprise cloud manufacturing service platform, the practical application is used to verify the correctness and effectiveness of the proposed method.