近年來,隨著醫療產業競爭的逐漸白熱化,促使台灣的醫療機構必須在面對相當有限的醫療資源情況下,務實地思考如何提升其經營績效來提供看診民眾最佳的醫療服務品質。在本研究中,我們結合了資料包絡法、類神經網路及基因演算法,以台灣地區大型醫療機構為研究對象進行效率的分析,主要目在乃在於能提供台灣地區大型醫療機構在進行醫療服務效率分析時的一個參考模式。而為了要暸解影響大型醫院服務民眾效率的關鍵因素,本研究亦透過Tobit模式的建構來了解環境因素對執行效率之影響。根據實證結果我們發現,本研究所提的方法不但可以作為醫療機構在執行效率分析時的參考模式及程序,分析結果更可做為未來績效改善的思考方向。
The health insurance program in Taiwan was started on March, 1995. In order to precisely estimate the subsidization for each large hospital, the Department of Health in Taiwan has spend lots of effort to investigate the efficiency of large hospitals. This paper integrates data envelopment analysis (DEA) and neural networks (NNs) to examine the relative efficiency of large hospitals. In additions, the Tobit regression method was applied to find the ffects of environmental variables on efficiency scores. To demostrate the effectiveness of the proposed method, the estimated results are compared with the one made by normal DEA. On the whole they are comparable. Furthermore, the suggestion on how to improve the performance is given.