Research On Discovery And Integration Of Manufacturing Resource Based On Joint Mechanism

Along with the development of information technology and network information resources, the total technology information provided by the Internet reaches more than 20TB. And its increment rate is more than 25% per year. The Internet brings us a mass of information of manufacturing resources, but it brought us some negative issues also at the same time. There are good and bad information on the net. It takes us much time to access to the valuable information. The resources utilization rate is still low, the necessary information resources can’t be found timely and accurately by demanders.One of the aims of network manufacturing is the sharing of manufacturing resources. The basis process of manufacturing resources shared is releasing and finding effectively manufacturing resources.At present the description of information of manufacturing resources lack of efficient finding and integration mode.This work was partially supported by State Key Lab of CAD & CG of Zhejiang University （Grant No.A0702） and Innovation Programme of Regular Institutions of Higher Education for graduate in Jiangsu province （Grant NO.CX09B₁91Z）. This work expounded the principle of finding and integrating manufacturing resources and the related problems. The dissertation is organized as follows:Firstly, based on the analysis of the current research results at home and abroad, for the disadvantages of modeling of current networked manufacturing resources, according to the current requires of modeling, meta-model of network manufacturing resources was put forward. The meta-model can support an uncertain environment of network manufacturing resources. After introduction and analysis of modeling language of the current network manufacturing resources, the semantic Web markup language is selected for modeling language. Ontology technology and information coding technology are introduced into modeling at the same time. Finally an example is given.Secondly, the shortcomings of the model of the existing manufacturing resources discovery are analyzed, combining UDDI （Universal Description, Discovery and Integration）,WSIL （Web Service Inspection Language） with manufacturing resources discovery model, a joint discovery mechanism are introduced first in the uncertain network enviroment. So far the mechanism has not been found in the literature. The model frame and working principle are given and the characteristics are investigated. Then the key technology in discovery mechanism, domain ontology, semantic Web Service language, ontology similarity algorithm and matching algorithm is discussed.Thirdly, through the manufacturing resources discovery model, a candidate of manufacturing resources can often be attained, so it is necessary to evaluate and select one of manufacturing resources. The demand on the evaluation of current manufacturing resources is analyzed. According to evaluation principles, corresponding evaluation index of manufacturing resources are presented. Based on evaluation indexes system of manufacturing resources, an evaluation model combined fuzzy theory with analytic hierarchy process and law of comparative judgment was presented for the problem of evaluation and selection of distributed manufacturing resources. The method solves the problem how we can decide the weight factors when there are many evaluation indexes, enhancing scientificalness and reliability of evaluation. An example is given to demonstrate the effectiveness of the method.Fourthly, based on the joint discovery mechanism, different needs in all kinds of personnel are analyzed. They have both the general and special characters. Active pushing service can improve greatly the accuracy of manufacturing resources discovery and save lots of time. Service framework is introduced. The representation methods of document text are compared. Vector space model is selected to represent text. The method of gaining text term is introduced and the theme weights are amended. Interested model and updating model are established. Various active push methods are summarized and the mail delivery service is used to push information in the article.Fifthly, according to the corresponding theory of the preceding chapters, a prototype system and main module are introduced. The development environment and tools, database design, discovery module, personalized realization technology are analyzed. Through the operation and analysis of the prototype system verify the correctness of the system. The system can improve the accuracy of manufacturing resources discovery and has hommization also. The research results have reached the leading level in China.