Comparison of ranking methods in data envelopment analysis models

Standard data envelopment analysis (DEA) models split decision making units into two groups, efficient and inefficient. Efficiency score of inefficient units allows their ranking but efficient units cannot be ranked directly because of their maximum efficiency. That is why there were formulated several different models for ranking of efficient units. The aim of this paper is to compare several definitions of super-efficiency and other approaches for ranking decision making units in DEA models including new definitions based on distances from optimistic and pessimistic envelopes of the data set, cross evaluation models, our original super-efficiency definition based on slack models and goal programming, etc. Numerical experiments are realized on a simple example.