<p>In Network-Data Envelopment Analysis (NDEA), the values of all data are to be known deterministically, whereas there are many real-world situations in which some uncertain data is available. This paper develops the NDEA models to include uncertain data of interval type. Since the data are uncertain of interval type, it is natural to obtain the ranges of the efficiency measures over the data. The intermediate activities play a central role in measuring the efficiency measures of units with a two-stage structure. The scholars introduced six scenarios to model the intermediate products in NDEA. As the intermediate products have the dual role of outputs of the previous stage and inputs of the next stage, obtaining the efficiency bounds faces some difficulties. This paper provides an explicit formula for computing the lower bound and the upper bound of efficiency measures under different scenarios. The results show that solving only two linear programming problems is needed to derive the bounds of the efficiency measures under some situations, while the computational efforts increase in other cases. Finally, a case study has been used to illustrate the proposed method.</p>