Utilizing dea and preference relation for comparing efficient decision making units : an application for ranking bank branches

Abstract EN

The aim of this paper is to present an integrated method using data envelopment analysis (DEA) and preference relation for full ranking efficient decision making units (DMUs).

One of the main imperfections in using DEA to construct the preference relation is the presence of alternative optimal solutions in the related DEA models.

In this situation, ranking DMUs may be varied by changing optimal solution.

This fact is shown by an example.

In this paper a model based on DEA to acquire the components of the preference relation is derived and then a new method to modify the proposed model in order to obtain unique optimal weights is suggested.

The performance of the new method and other methods in the literature are compared by a numerical example.

In addition, a numerical example is provided to illustrate the geometrical interpretation of the proposed method.

A case study about bank branches in Iran is outlined to assess the validity of the proposed approach.