碩士 / 開南大學 / 企業與創業管理學系 / 97 / To estimate the performance of decision making unit (DMU) within the same group, a hybrid algorithm combined K-means clustering technique with data envelopment analysis (KDEA) was proposed in this study. KDEA improved the similarity in the same cluster, and proved to be easily to derive the relative efficiency of each DMU more accurately.
Both K-means clustering and data envelopment analysis (DEA) have similar features in dealing with multidimensional abstract concept with little limitation in heterogeneous data type and in need of derivative. In this study the novel approach KDEA has exploited these features and been proved by its outstanding outcomes. A comparison between K-means clustering, DEA and KDEA indicates that KDEA performs superior than the others in terms of ranking and predicating operation achievement of each DMU in a classified group.
Finally, a real case in studying of the managing performance of listed transportation companies in Taiwan was employed to verify the proposed algorithm in this paper.
| Identifer | oai:union.ndltd.org:TW/097KNU00121015 |
| Date | January 2009 |
| Creators | Yen Chi-Fang, 嚴奇芳 |
| Contributors | Huang Kuang-Chung, 黃光中 |
| Source Sets | National Digital Library of Theses and Dissertations in Taiwan |
| Language | zh-TW |
| Detected Language | English |
| Type | 學位論文 ; thesis |
| Format | 71 |
Page generated in 0.0013 seconds