The 4th International Symposium on Project Management (ISPM 2016)

216 / 2016-06-30 11:06:42

A Model of Mutually Acceptable Cross Efficiency in Data Envelopment Analysis

10719

This study summarizes the concept of cross efficiency derived from data envelopment analysis (DEA) and analyzes the cross efficiency related issues. The cross efficiency is a generalization of the traditional CCR model and widely applied to many fields. It is based on peer-evaluation, linking the performance of one decision-making unit (DMU) with others. Many scholars have studied cross efficiency from different aspects, but the non-uniqueness of the DEA optimal weights may still suppress the effectiveness of cross-efficiency evaluations. To address this uniqueness problem, this paper generalizes the original DEA cross-efficiency concept to a mutually acceptable cross-evaluation efficiency. Two numerical examples illustrate the power and suitability of the proposed algorithm to improve the cross-efficiency scores of the DMUs.