An improved cross-ranking method in data envelopment analysis. Issue 1 (2nd January 2016)

• Record Type: Journal Article
• Authors: Jin, Minyue; Ang, Sheng; Yang, Feng; Bi, Gongbing
• Is Part Of: Infor. Volume 54:Issue 1(2016)
• Page Start: 19
• Page End: 32
• Publication Date: 2016-01-02
• Subjects: Data envelopment analysis -- interval cross-efficiency matrix -- cross ranking -- preference aggregation; Operations research -- Periodicals; Electronic data processing -- Periodicals; Systems engineering -- Periodicals; Systems engineering; Electronic data processing; Periodicals; 003.05
• DOI: 10.1080/03155986.2016.1149313
• Languages: English
• ISSNs: 0315-5986

Abstract

ABSTRACT: Cross-efficiency evaluation is an extension of data envelopment analysis methodology for fully ranking decision-making units (DMUs). Based on the idea that ranking order is much more meaningful than the individual efficiency score in several circumstances, a cross-ranking matrix was recently introduced. The matrix was built by replacing the efficiency score in the conventional cross-efficiency matrix with the ranking order of that efficiency score in each row. However, the non-uniqueness issue in the cross-efficiency scores in the cross-efficiency evaluation may result in different ranking orders in the matrix and may thus limit the usefulness of cross-ranking method. This study improves the cross-ranking method by introducing an interval cross-efficiency matrix in cross-efficiency evaluation. The interval cross-efficiency matrix can reveal all possible cross-efficiency scores through self-evaluation and peer evaluation. Therefore, the derived cross-ranking matrix based on the interval cross-efficiency matrix can show all information on possible ranking orders for all DMUs. The improved method is illustrated by two numerical examples.