An Extended DDF -Based Approach for Efficiency Measurement under Data Uncertainty

The directional distance function DDF -based efficiency is a non-oriented efficiency measure that incorporates features of both input-oriented and output-oriented DEA models and gauges the deficiency both in inputs and outputs together. The next advantage of the DDF model is its flexibility in selecting a directional vector that, in a scenario, can be chosen by the decision maker. On the other hand, one of the pivotal challenges in efficiency evaluation, in application, is the uncertainty of the data on the basis of the degree of belief. From a logical point of view, applying a deterministic DEA model in such a case may lead to unreasonable results. To address such drawbacks, in this study, we propose a new non-radial DDF model in the presence of uncertain data. This paper provides a model-based non-oriented DDF that simultaneously considers uncertain data and undesirable (bad) outputs for evaluating efficiency DMUs. To determine the radius of stability, the sensitivity and stability of the proposed method are analyzed, and the incidence of data variations does not change the classification of DMUs. Furthermore, the concept of uncertain DDF and uncertain SBM efficiency was utilized to measure a new uncertain directional mix efficiency, which is a direction’s ability to determine the type of inefficiencies for all inefficient DMUs.

کلیدواژه ها:

Data Envelopment Analysis ، Directional distance function ، Uncertainty ، Mix-efficiency ، Undesirable outputs