A RESEARCH ON THE INTERVAL ESTIMATION OF PAIRWISE WEIGHT RATIO IN DATA ENVELOPMENT ANALYSIS
Dr. Chiao-Ping Bao
Dr. Nai-Chieh Wei
Mu-Lin Chiou*
Department of Industrial Management,
I-SHOU University, Taiwan (R.O.C)
*Corresponding author: This email address is being protected from spambots. You need JavaScript enabled to view it.
Abstract
Traditionally, the Data Envelopment Analysis (DEA) efficiency evaluation model uses the mathematical linear programming to find a set of weights that can make their own efficiency value greatly for the Decision Making Units (DMU).Therefore, in the traditional model of DEA, there are generally unreasonable phenomena, including "Variable Weight is 0" and multiple "Efficient DMUs." In order to break through the irrational phenomena, this article uses the concept of AR Model (Thompson, 1986) and combines the Expert Weights Meth-od (EWM) (Bao, 2013) to develop a new Model, which is the Interval Estimation of Pair-wise Weight Ratio, EWM_AR Model. At the same time, this article proves a theorem which shows how the EWM_AR Model avoids the problem of generating the Formula of Weight Limit when dealing with the Interval Estimation of Pairwise Weight Ratio.
According to the research, after incorporating the Relevance of Each Variable, EWM_AR Model in this article produces two major achievements: (1) In terms of the weight limits, the shortcomings of the traditional "Variable Weight is 0" can be improved; (2) It can elim-inate the shortcomings of the traditional multiple "Efficient DMU."
Keywords: Data Envelopment Analysis; Region-Assurance Method; Expert Weighted Method
 0923 Final.pdf | [ ] | 157 kB |