2572. Cross Efficiency and DEA
Invited abstract in session TA-60: DEA methodological developments II, stream Data Envelopment Analysis and its applications.
Tuesday, 8:30-10:00Room: Western LT
Authors (first author is the speaker)
| 1. | sara hosseini
|
| Business, Aberdeen university |
Abstract
We propose a novel approach to cross-efficiency evaluation in Data Envelopment Analysis (DEA) that is applicable to various DEA models and accommodates any returns to scale. A key advantage of our method is that it ensures the diseconomy of scale of one Decision-Making Unit (DMU) is never used to assess another, addressing a critical limitation in existing methods. We demonstrate that our approach offers stronger theoretical justification while maintaining consistency with the standard cross-efficiency method under constant returns to scale (CRS). To support our methodology, we introduce a unique measure for the diseconomy of scale of a DMU and define the concept of preferred scope. While preferred scope is not uniquely determined, we establish centroid preferred scope as a well-defined, average measure to facilitate analysis. Using variable returns to scale (VRS) DEA, we compare our method with existing approaches and illustrate how hierarchical cluster analysis based on centroid preferred scope can reveal meaningful differences in performance. A major challenge in cross-efficiency analysis has been its limited application beyond CRS models, with only a few exceptions (Lim & Zhu, 2015; Su & Lu, 2019). Previous methods primarily focus on ensuring nonnegative cross-efficiencies by translating inputs and outputs or imposing additional constraints, both of which lack strong justification and have not gained widespread adoption. Our approach addresses this gap by offering a gener
Keywords
- Data Envelopment Analysis
- Optimization Modeling
Status: accepted
Back to the list of papers