| Being a new form of economy, Knowledge Based Economy not only drives the development of economy itself but also emerges as a very import resource to evaluate the progress of modern society. However, the characteristics of high technologies, which poses highly integrated yet polarized features, leads to that no single organization is capable to have both knowledge and ability for Business Innovation. Therefore, employing the algorithms and models from knowledge transform and knowledge mining can effectively utilize computing power to speed up knowledge discovery, classification, extraction and recognition. These intelligent, dynamic and networking algorithms have broad prospects in many knowledge management research areas including customer management, Enterprise Resource Planning and supply chain.With knowledge transform, mining and management as the foundation, this thesis targets at knowledge discovery, classification, extraction and recognition in aspects of both performance and effects to provide an overall view in the following areas:review of systematic theory of knowledge management, combining management and information technology to discuss knowledge transform and mining. Furthermore, a few case studies are adopted to build specific knowledge management system. The thesis is organized as follows:First of all, it briefly describes the background, research objective and motivation of knowledge based economy. Then, it reviews the literatures of the field to analyze and compare the current research status and applications regarding to the knowledge management, transform and discovery.Secondly, we use background theory and prior knowledge on the basis of knowledge expression and algorithm efficiency to present an efficient global optimization searching algorithm based on individual immunity and group evolution. The proposed algorithm emphasizes the discovery of new knowledge and the prediction of advanced rules, which promotes the stability and overall performance of the group. In addition, this algorithm can also provide relative high precision during the rule extraction process.In a further step, regarding to the objective of knowledge discovery in customer knowledge management system, this thesis applies data mining technology over customer knowledge management through employing association rule mining. The built customer knowledge discovery system is presented in this section along with the relative algorithm and source code.In addition, based on Hall three dimensions structure, we compare ERP and KMS in the level of research methodology. Then, a new knowledge management system based on enterprise resource planning is proposed. By using the kinetic analysis of important knowledge factors of ERP and KMS, the system is then dynamically simulated and observed with various parameters and knowledge inputs to evaluate the system behavior and trends. As a result, the knowledge transform can be studied to choose optimal or second best solutions.In the fifth part of this thesis, it puts qualitative and quantitative analysis together to inspect supply chain risk management based on risk knowledge discovery and classification. Next, we apply fuzzy optimization to data mining methodology in order to build supply chain risk decision optimization model, which can be used to reduce the supply chain risk and enhance risk resistance capability.Last but not least, this section discusses the necessity and feasibility of overall system planning of ERP, SCM and CRM from the view of enterprise management and knowledge demand. Additionally, this section proposes an overall system framework, deployment plan, technology difficulty and priority. Moreover, it uses data warehouse technology to macroscopically build ERP, SCM and CRM system so that the real time data analysis, processing, management and decision can be achieved.The last section concludes and discusses step-wise future works. Some inadequate points of this study are also illustrated to expect further improvement.
|
PDF Full Text Request