evaluating the efficiency score of two-stage network dea in the presence of shared resources using a multiplier model

In recent years, several network dea models have been developed. of these, amirteimoori et al. (2016) proposed an additive model to measure the efficiency of the two-stage network dea model when stages (sub-dmus) consume shared input sources in both operation stages. however, the method proposed by amirteimoori et al. (2016) does not obtain efficiency scores for each stage in the network system, as well as, for the whole operation of the system, and just by applying an additive model the efficiency and inefficiency of the dmus is determined. however, it can be argued that this is a weakness of this method. to overcome this deficiency, we propose an approach for estimating the efficiency score of the two-stage network dea model in the presence of shared input sources. numerical examples show the applicability of the approach.