碩士 / 明志科技大學 / 工業工程與管理研究所 / 99 / Semiconductor industry has the characteristics of continuous production. The revenue and profit rely on the increment of product output which are contributed by the advanced manufacturing technology. During mass production, abnormal events happened when one or more manufacturing factors drifted from normal. Certain product lots thus were affected. The purpose of this research is to minimize the mis-judgment of lot disposition and reduce the labor cost of meetings by Case-Based Reasoning (CBR) expert system. A man-machine interface which links to the database of historical abnormal event will be introduced.
This research investigate company I, in which the product be graded in 3 levels: Normal grade (QC), Secondary grade (NC) or Non-saleable grade (Scrap). We collect historical abnormal events and analyze the key factors in those cases. Based on CBR, an expert system is developed by the following 5 steps: Identification, Conceptualization, Formalization, Implementation, and Testing. Finally, we validate the expert system by testing abnormal events happened in company I during 2010.
| Identifer | oai:union.ndltd.org:TW/099MIT00030055 |
| Date | January 2011 |
| Creators | Lin YenHsi, 林延錫 |
| Contributors | Weng WeiTai, 翁偉泰 |
| Source Sets | National Digital Library of Theses and Dissertations in Taiwan |
| Language | zh-TW |
| Detected Language | English |
| Type | 學位論文 ; thesis |
| Format | 55 |
Page generated in 0.0017 seconds