Extended DFM Models in DEA

Abstract

This chapter aims to present a newly developed and extended distance friction minimization (DFM) model in the context of Data Envelopment Analysis (DEA), in order to comply with plausible and real-world circumstances. The DFM model is generally able to calculate either an optimal input reduction value or an output increase value in order to reach an efficiency score of 1.000, even though in reality this might be hard to achieve for low-efficiency DMUs. Most DEA models and also the DFM model have intrinsic limitations or weaknesses. Therefore, we need a method that allows for the presence of reference points that remain below the efficiency frontier. In this chapter we propose successively a Goals-Achievement model, a Stepwise Improvement model, and a Target-Oriented model based on the DFM framework. These models are categorized as “Target approaches.” On the other side, in many cases, input or output factors may not be directly flexible or adjustable due to the indivisible nature or inertia in some input or output factors. Usually, the original DEA model and the DFM model do not allow for such a non-controllable or a fixed input factor. Therefore, we need a method that may take into account a flexible or adjustable factor in a DFM model. In this chapter, we propose an Adjusted-Improvement model and a Fixed-Factor model based on the DFM framework. These models are categorized as “Adjustment approaches.”