A Robust DEA Model to Handle the Uncertainty in Production Trade-Offs

Abstract

Weight restriction based on production trade-offs as simultaneous changes of inputs and outputs is considered as a weight restriction approach that preserve the technological meaning of efficiency. However, estimating the production trade-offs depends on the expert’s judgments and the methodologies that will be applied, hence it may not be always possible to translate and estimate these production trade-offs into exact and precise data. In this article, robust optimization is applied to give a robust counterpart for the weight restriction DEA model which in some of the production trade-offs are subjected to uncertainty. Using the proposed robust weight restriction model (robust WR-TO), it is possible to evaluate efficiency of DMUs in the presence of uncertain production trade-offs to achieve more reliable solutions. The proposed robust counterpart is given using a combination of box and polyhedral uncertainty set. Moreover, an interval weight restriction model (interval WR-TO) is modified to handle the uncertainty in production trade-offs, therefore a comparison between the two proposed model is provided. A real case study of granted research projects is given to validate the proposed models and illustrate the potential application of the proposed robust weight restriction model. The results revealed that the proposed robust WR-TO model gives more reliable effiicency scores and also provides a better ranking of the DMUs compared to the conventional WR-TO model and interval WR-TO model.