Data Envelopment Analysis (DEA) 資料包絡分析

-- DEA was established by Charnes et al. [1], in terms of a set of inputs and outputs, in 1978 as a methodology used in relative efficiency evaluation (of operating units with the same goals).

-- DEA has been regarded as a powerful and effective new methodology for organizing and analyzing data, identifying the best practice frontiers and inefficiency sources, and providing improved performance, for example [4] on financial institutions.

 [1] Charnes A, Cooper WW and Rhodes EL, “Measuring the efficiency of decision making units,” European Journal of Operational Research, 1978, 2(6): 429-444.

 [2] Andersen P, and Petersen NC, “A procedure for ranking efficient units in data envelopment analysis,” Management Science, 1993, 39(10):1261-1264.

 [3] Hadad Y, Friedman L, Sinuany-Stern Z, and Mehrez A, “DEA super efficiency multistage ranking,” Computer Modelling & New Technologies, 2003, 7(1):37-46.

[4] European Journal of Operational Research, April 1997 special issue, 98(2).

Example 1:

Input 1 = capital (in 100s of hospital beds)

Input 2 = labor (in 1000s of labor hours/month)

Output 1 = 100s of patient-days during month for patients under age 14 (or17)

Output 2 = 100s of patient-days/month for patients between 14 (or 17) and 65

Output 3 = 100s of patient-days during month for patients over 65

Table: Input & Output

(1)   Efficiency = (Value of output)/(value of input) <= 100%

Value of Input >= Value of Output

(2)   CCR (Charnes, Cooper, and Rhodes) Ratio Form:

(3)   Linear Model:

Comments on DEA:

(1)   Popular quantified approach;

(2)   Single index;

(3)   Problematic Input/Output choices;

(4)   Not absolute efficiency;

(5)   Unable to rank:

(a)    efficiency－(1993) Andersen and Petersen “Super Efficiency”

(b)   in-efficiency－(2003) Hadad et al. “Multi-stage Ranking”

Reference Website參考網站:

1.      http://www.deazone.com/

Ali Emrouznejad's DEA Home Page, Warwick Business School, UK. This page has been created in honor of A. Charnes, W. W. Cooper, E. Rhodes and Rajiv D. Banker who originally developed DEA

2.      www.uq.edu.au/economics/cepa/index.htm

The Centre for Efficiency and Productivity Analysis (CEPA) is in the School of Economics at the University of Queensland. The main purpose of the Centre is to provide a focal point for research, consultancy and training in efficiency and productivity analysis in Australia and the Asia/Pacific Region.

3.      www.deafrontier.com

This page consists of a series of DEA softwares which are Add-Ins for Microsoft® Excel developed by Joe Zhu, DEAFrontier^{TM .}

Book中文書:

1.

績效是描述組織運作結果之整體觀念，績效評估則為管理控制之必要程序。經由績效評估建立有效的迴饋系統，可達成良性的管理循環，進而改善資源之使用效率，提高組織績效。本書作者對DEA的研究頗有心得，由簡入深有系統的介紹DEA的各種模式及其應用。尤其書中所舉案例，都是作者以台灣為應用對象並發表於國際學術期刊的創作，兼具有本土化與國際化的特色。

2.

資料包絡分析法－理論與應用

       本書強調理論與應用，適合對績效評估與管理有興趣的讀者閱讀。本書深入淺出介紹DEA基本理論與最新發展模式，並採個案方式，為讀者說明DEA在管理實務上之應用。本書共二篇，理論篇與應用篇。理論篇介紹DEA基本觀念、發展近況、理論模式；DEA特性、限制、應用程序、軟體、書籍、網站、文獻目錄(1978-2003)與績效評估方法。應用篇採8個本土化個案（醫院、車床、航空公司、聯營公車、工廠、大學系所、製造業與地檢署），來說明理論之應用。

3.

對於學習「管理」學門的學子而言，「績效評估」是一門無可避免、必須學習的學問。評估績效的方法有很多種，學者們常用的有比例法、迴歸分析法。但是這些方法各有侷限性，且都不如「資料包絡分析法」的客觀和應用範圍較廣。本書之內容並不是以「深奧」為導向，所敘述的都是「基本的」方法，不強調太多深奧之理論發展，而是從最基本的「線性規劃」開始介紹。