The use of weight restrictions in data envelopment analysis (DEA) is considered an appropriate method to avoid zero inputs and outputs weights. Due to a number of problems associated with the use of weight restrictions, the production trade-off method in DEA is often preferred. An important issue is that in most common DEA models, the internal structure of the production units is ignored, and the units are often considered as black-boxes. The current study aims to estimate the production trade-offs in two-stage network data envelopment analysis (NDEA) to observe its likely impact on efficiency evaluation and discrimination of units and subunits. Also, the probable effect of such the trade-offs are shown on the overall efficiency decomposition to divisional efficiencies and production possibility set (PPS) in two-stage NDEA. Finally, a numerical example is used to explain the results and compare the possible effects of different production trade-offs scenarios in two-stage NDEA with standard DEA models.
