分方法DEA-系统工程学报.PDF

28 6 Vol.28 No.6 2013 12 JOURNAL OF SYSTEMS ENGINEERING Dec. 2013 ( DEA ) 1 2 1 , , (1. , 100190; 2. , CT2 7PE) : . , : , , , , . , DEA , . DEA , DEA . , . : ; ; : C934 : A : 1000−5781(2013)06−0840−21 Review of data envelopment analysis 1 2 1 Yang Guoliang , Liu Wenbin , Zheng Haijun (1. Institute of Policy and Management, Chinese Academy of Sciences, Beijing 100190, China; 2. Kent Business School, University of Kent, Canterbury CT2 7PE, UK ) Abstract: The data envelopment analysis (DEA) is a multi-factor productivity analysis tool for measuring the relative efficiencies of a homogenous set of decision-making units ( DMU s), which is firstly proposed by Charnes, Cooper and Rhodes in 1978. There are several key elements in DEA models, including production possibility set, measure, preference, type of variables, level of the evaluation problem, and explicit data or not. The combination of these elements can form different DEA model, which can be used to solve different problems. This paper aims to summarize a number of important studies in recent years based on the above six key elements of the DEA models. Also, this paper briefly describes some applications of DEA models. Finally, this paper presents some possible future research directions. Key words: data envelopment analysis; relative efficiency; key elements 1 (data envelopment analysis, DEA) [1]. Farrell[2] , (decision making units, DMU ) , . , (productivity) (efficiency). DEA [3] . DEA DMU . DMU