RESEARCH ON THE WEIGHT RESTRICTION OF DATA  ENVELOPMENT ANALYSIS - TWO-STAGE OF PAIRWISE WEIGHT RATIO INCORPORATING THE EXPERT WEIGHTING METHOD

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

The traditional DEA model generally has an unreasonable problem of "variable weight is 0," and this problem will seriously affect the discrimination of the DEA efficiency evaluation; corresponding to this shortcoming, we have also proposed a model that can solve the problem. This model is called EWM_AR (Bao, 2019). Nonetheless, in the follow-up study of "How to Expand the Application Scope of the EWM_AR Model," we found that it is indeed easy to get the lower limit of negative values when dealing with the related issues of "Interval Estimation of Pairwise Weight Ratio". Therefore, this article is re-presenting a model with a non-negative lower limit characteristic. This model is called "Two-Stage Weighted Proportional Interval Estimation"- mEWM_AR Model.

The research results of this article found that after implementing the "Normalization" and incorporating "the Importance of the Correlation Between the Variables" method, there are two main results produced, including: (1) It can indeed solve the unreasonable problem that the traditional DEA model generally has "variable weight is 0"; (2) It can indeed make the"Interval Estimation of Pairwise Weight Ratio" have the characteristic of non-negative lower limit. This means that the mEWM_AR Model will provide a more objective and a more unbiased efficiency evaluation mode.

Keywords: Data Envelopment Analysis; Expert Weighted Method; EWM_AR Model

Attachments:
Download this file (1045 Final.pdf)1045 Final.pdf[ ]201 kB
Go to Top
JSN Solid 2 is designed by JoomlaShine.com | powered by JSN Sun Framework