Study On The CBR Mechanism Of DSS For Enterprises’’Integration Of Itization And Industrialization’

The "Integration of Informatization and Industrialization"(IoII) meaning the deep integration of information technology and industrialization, refers to using information technology to stimulate industrialization, and promoting informatization with the development of industrialization, pursuing a sustainable development model. At present, China is exploring to promote the "Integration of Informatization and Industrialization"(IoII), confronting with a lot of problems and difficulties. The essence of loll is to promote the development of industrialization by the means of informatization. In addition to improve enterprise’s production efficiency and management level, the goal of loll is also to make industrial products more digital and intelligent, and to improve the industrial service level. Enterprise is the subject of this integration, and the core of integration solutions is to guide enterprises to achieve informatization, and it is difficult for enterprises to make achievements without reasonable schemes even with adequate technical supports. To solve this, Beijing Municipal Science&Technology Commission coperate with our university to develop a decision support system (DSS) for Enterprises’loll. Within the DSS, the reasoning subsystem is the core, and this paper studied the first development stage of this subsystem, namely case-based reasoning (CBR) mechanism.The heart of CBR Mechanisim is the case base and reasoning algorithm. Using the investigation method and literature research method, this paper firstly analyzed the decision-making process and case features of loll and the function requirement related to the CBR of this DSS. On this basis, with knowledge engineering method, according to the construction idea, this paper made a clear definition for the methods and principles of case selecting and maintenance, and constructed a hierarchical storage structure and two index catalogs for the case base in line with attributes. Meanwhile, adopting the method of system engineering, this paper designed two retrieval types for similar cases-basic and advanced, and there are two steps for each retrieval type-the first step to screen cases and the second step to calculate the similarition. Also, the AHP method was used to allocate attribute weights, and the standardization of the attributes and similarity calculation formula were given out. At last, from the view of the stability and the accuracy of retrieval, this paper examined the CBR mechanisim in this study.During the study, this paper has the following three aspects of innovation: 1) In allusion to the requirements of enterprises’ IoII decision-making, this paper designed a case base that can meet the demands of enterprises from four aspects:the case selection, case representation, case storage and case retention.2) Designed an improved standard normal distribution normalization algorithm for numeric attributes.3) Studied case retrieval and similarity algorithm especially for IoII cases.