CROSS-EFFICIENCY EVALUATION UNDER NON-DECREASING RETURNS TO SCALE AND WEIGHT RESTRICTIONS

Data Envelopment Analysis (DEA) is a powerful technique for estimating the efficiency of decision-making units (DMUs) by assigning appropriate weights. To enhance the discriminative capabilities of DEA in evaluating DMUs, the cross-efficiency evaluation method has been introduced, based on peer evaluations. However, a significant challenge in cross-efficiency analysis lies in the non -uniqueness of optimal weights. In this paper, we propose a novel approach to cross -efficiency evaluation, emphasising the selection of weighted profiles within non -decreasing returns to scale (NDRS) production technologies. This approach introduces conditions to ensure non-zero weights and reduce weight disparities. Additionally, we present two optimisation models from benevolent and aggressive viewpoints as secondary goals. To illustrate the effectiveness of the new approach, we provide numerical examples and an empirical application involving the performance evaluation of 14 airline companies.