Research On System Reliability Management For Manufacturing Grid

With the development of the technology of computing and network, and the constant advance of manufacturing informationization, Manufacturing Grid emerges as the results of combining the information technology and manufacturing, to provide a general platform for Networked-Manufacturing. As a grid application in manufacturing, the MG is a distributed isomeric system, whose complexity lies in the heterogeneous and various resources and the complex actions. In the process of grid application, the system reliability of Grid is focus on the network connectivity. The runtime status of resources was ignored. The problem becomes more important as the expanding of the grid application. The integrated system reliability management for Manufacturing Grid was proposed to keep the MG system running continually and stably with high performance.First, a scope of the reliability management was given by analysising the system reliability of MG, the connotation of the MG system reliability was probed, and the reliablity characteristic variables of MG were defined. According to the polymorphism of the MG system, the layering reliability index was presented. Based on the analysis of MG reliabilty, the reliablity block diagram and the reliability evaluation for MG were proposed, which provides the theoretical direction to put the reliability management into action effectively.The reliability management strategy was determined based on the reliablity analysis of MG. The architecture of MGRM was developed, which was described from its concept model, process model and fuction hierarchical model. And the critical technologies were discussed as well.The reliabiltiy management and evaluation goes unpon the data indicating the runtime state of system. So, it is important for the effective management of MG reliablity that how to cellect the data from the resources dispersed in the MG. The Data Collection and System Recovery Based on Agents was proposed by adoping the agent technology. At the same time, the structure model, communication model, communication strategy and coding scheme of MGRM-Agents were presented. The collaberation of static agents and mobile agents enhanced the scability of MGRM, reduced the system consumption of the managed object and improved the reusability of MGRM-Agents. We studied the realization mechanism of intelligent agents as well.The critical point of MGRM lies on the perfect data management and the system reliability analysis and evaluation. We adopted Data Warehouse into the data management in the MGRM to manage the sponge data generated in the process of the MG reliability management effectively and support the high effcient data retrieval. To gain more useful information and knowledge from the sprone and sophisticate data, which are helpful for optimizing or determining the monitoring and managing strategy,...