The Design and Evaluation of a Knowledge Management System

• Main Authors: I-Chun Ou, 歐伊純
• Other Authors: Shy-Ming Ju
• Format: Others
• Language: zh-TW
• Published: 2003
• Online Access: http://ndltd.ncl.edu.tw/handle/18170044600334069119

碩士 === 國立高雄第一科技大學 === 資訊管理所 === 91 === To improve competitiveness a business relies on information technology to make sound business decision. Furthermore, able to use knowledge management system to compile, store, derive, enquire, propagate and apply organizational knowledge will certainly be a key to success for enterprises in the age of knowledge economy. Knowledge can be classified as explicit or tacit. To human, explicit knowledge is recorded on paper or electronic media in the form of picture, text, sight or sound, thus facilitating sharing. And tacit knowledge is in individual’s mind, thus allowing derivation and application. While explicit knowledge to knowledge management systems is still the same as that to human, tacit knowledge is metadata created and encoded during reification of explicit knowledge. The effectiveness of a knowledge management system in supporting knowledge enquiry, derivation and application is closely related to the completeness of metadata. Nonaka stated that, depending on context and needs, there are four types of transitions between explicit and tacit knowledge: combination of different explicit knowledge, internalization of explicit knowledge into tacit knowledge, externalization of tacit knowledge into explicit knowledge, and socialization of tacit knowledge. To individuals, transition of knowledge takes place all the time. To organizations, compilation of organizational knowledge still requires human intervention, but due to large amount of knowledge, large number of users, complex structure and frequent changes, having a knowledge management system to help in organizational knowledge transition will certainly reap a great benefit. The knowledge management system proposed in this research is built on the World Wide Web. It adopts XML Topic Maps as knowledge encoding standard. Encoded knowledge is stored in a native XML database, and the combination and enquiry of knowledge fully utilize the semantics represented by XML Topic Maps. It also uses the induction and deduction capabilities of an inference engine to achieve knowledge derivation. The knowledge encoding standard, the knowledge database, and the inference engine are all generic, and the system can interface to decision support system, expert system and artificial intelligence system through XML-based protocols, therefore, this system is not limited to a particular application domain. We have chosen a company in surface mount technology industry for case study, and used a popular criteria, K = (P+I)s , to evaluate the effectiveness of the proposed knowledge management system.