Ranking DMUs by using interval DEA cross efficiency matrix with acceptability analysis

• Published in: European journal of operational research 2012-12, Vol.223 (2), p.483-488
• Main Authors: Yang, Feng, Ang, Sheng, Xia, Qiong, Yang, Chenchen
• Format: Article
• Language: English
• Subjects: Acceptability; Applied sciences; Cross efficiency; Data envelopment analysis; Data envelopment analysis (DEA); Decision making; Decision making models; Decision making units; Decision theory. Utility theory; Efficiency; Exact sciences and technology; Intervals; Mathematical models; Mathematics; Matrix; Multiple criteria decision making; Nonparametric inference; Operational research; Operational research and scientific management; Operational research. Management science; Operations research; Probability and statistics; Ranking; Sciences and techniques of general use; Statistics; Stochastic multicriteria acceptability analysis (SMAA); Strategy; Studies
• ISSN: 0377-2217; 1872-6860
• DOI: 10.1016/j.ejor.2012.07.001

► We propose an interval cross efficiency matrix. ► We extend the traditional DEA cross efficiency to a SMAA-based technique. ► The decision-maker does not need to select aggressive or benevolent strategy. ► Uniform and Normal distribution are assumed to represent the weight distribution. Cross efficiency method is an extension of data envelopment analysis (DEA), and has been widely used for ranking performance of decision making units (DMUs). To eliminate the non-uniqueness of cross efficiency scores, the aggressive and benevolent strategies have been proposed as secondary goals to determine the unique cross efficiency score. The current paper aims to propose an alternative strategy which does not consider the preference of the decision maker in choosing aggressive or benevolent strategy. Instead, the paper considers all possible weight sets in weight space when computing the cross efficiency and each DMU is given an interval cross efficiency. By using the stochastic multicriteria acceptability analysis (SMAA-2) method, all DMUs in the interval cross efficiency matrix (CEM) could be fully ranked according to the acceptability indices. A numerical example about efficiency evaluation to seven academic departments in a university is illustrated.