Decomposition weights and overall efficiency in a two-stage DEA model with shared resources

Abstract Data envelopment analysis (DEA) is a non-parametric approach for measuring the relative efficiencies of peer decision making units (DMUs). The additive efficiency decomposition approach expresses the overall efficiency of a two-stage network DEA (NDEA) model as the weighted average of the efficiency scores for the two stages. To determine the weights, there are two main approaches; constant weights and an approach in which the weights are expressed as the portion of total resources devoted to each stage. The current paper provided an examination of the monotonicity of the decomposition weights in a two-stage DEA model with shared resource flows and found that the weight in such a model was not biased towards the second stage. The usage of constant weights in such a model is able to improve the discrimination of the efficient DMUs. Finally, the overall and individual efficiency variations were also studied by varying the weights in the model. The findings were verified using banking industry data and the results were compared with other models.