A stepwise benchmarking approach to DEA with interval scale data

Abstract

The conventional DEA models assume that all variables are measured on a ratio scale. However, in many applications, we have to deal with interval scale data. In Dehnokhalaji, A., Korhonen, P. J., Köksalan, M., Nasrabadi, N., & Wallenius, J. (2010). Efficiency analysis to incorporate interval scale data. European Journal of Operational Research 207(2), 1116–1121, we proposed a model for efficiency analysis to incorporate interval scale data in addition to ratio scale data. Our proposed model provides efficiency scores for each unit, but does not suggest target unit(s) for inefficient ones directly. In this paper, we investigate the concept of benchmarking in Dehnokhalaji et al. s (2010) model. We propose an algorithm which results in a path of targets for each inefficient unit. All units on this path are better than the unit under evaluation in terms of efficiency scores defined for interval scale data. The intermediate targets belong to sequential layers obtained from a layering algorithm and the final unit on the path is an efficient unit.