Inverse DEA-R models for generalized restructuring analysis

Abstract A generalized restructuring refers to an action in a commercial environment in which a homogeneous set of units is restructured to create a new set of units in the same environment to achieve specific performance goals. Inverse data envelopment analysis (InvDEA) is a useful analytical tool based on multiple-objective mathematical programming to estimate the required levels of input/output to achieve given efficiency targets. This paper is devoted to develop a new InvDEA based methodology with the data in ratio form, called InvDEA-R, for modeling the generalized restructuring. The most important advantages of the proposed models are: i) Maintaining the confidentiality of data, which can strengthen the motivation of managers to participate in the process of improving efficiency through reconstruction; ii) In addition to using pure input and output data, it considers a combination of their ratios in the reconstruction process. This not only increases the more appropriate reconstruction analysis’s discrimination power, but also provides a reasonable endogenous weight restriction framework for using the expert view in the generalized units restructuring process. This paper also presents a DEA-R estimator model to determine the minimum efficiency aims that post-restructuring units can gain. The developed restructure theory is examined via a banking sector application.