efficiency measurement and frontier projection identify of two-stage network dea in the presence of non-discretionary inputs and shared discretionary inputs between two stages

Conventional data envelopment analysis (dea) is a method forevaluating the performance of a homogeneous set of decision making units (dmus),without considering the internal structure of dmus. in recent years, considering theinternal structure of dmus, the network dea has been developed, henceforth, in themodels of network dea assumes that all inputs are discretionary, and it can becontrolled by the management of each sub-dmu and dmu, and also can be changed atits discretion. in any realistic situation, however, there may exist non-discretionaryinputs that are beyond the control of the management of each sub-dmu and dmu, andcan be said that weakness of each sub-dmu and dmu to product maximal outputlevels with minimal input consumption results in a decreased efficiency score, whichnumber of several approaches have been developed to overcome this weakness. thisstudy proposes an approach can be considered a standard model for the inclusion ofnon-discretionary inputs, and also shared discretionary inputs between two stages, formeasuring the efficiency and identifying the frontier projection in network dea. twoexamples of measuring the efficiency of the heart surgery hospitals in iran, that hasbeen discussed in this study, illustrates the idea presented in this study.