Network DEA by Multi-Objective Programming Methods and its Applications

• Main Authors: Chieh-Yu Chan, 詹婕妤
• Other Authors: Han-Ying Kao
• Format: Others
• Published: 2013
• Online Access: http://ndltd.ncl.edu.tw/handle/71168821935406624532

碩士 === 國立東華大學 === 資訊工程學系 === 101 === This study proposes the multi-objective programming (MOP) method for solving network DEA models. In the proposed method, the efficiency of each division (within a DMU) and the overall efficiency of the DMU are formulated as different objective functions. By the fuzzy approach, the overall as well as the divisional efficiencies can be computed in a cohesive framework. We adopt information entropy to estimate the discriminating power of various DEA models. A case of the solar energy industry in Taiwan is designed and implemented. We collect the operational data from a set of public listing companies in the industry to explore possible problems regarding difference levels of efficiencies. The results show that this study obtains satisfactory and discriminating efficiency scores and extracts meaningful management implications for the indusrtry.