Data Envelopment Analysis is a nonparametric method based on the mathematical model. The concept of return to scale is one of the most important issues in economics and also in the data envelopment analysis, which includes a large part of the studies. Determining the type of return to scale (increasing, decreasing, and constant) will provide information to the manager through which he will be able to decide on how to achieve the optimal unit level under evaluation. Despite the fact that in the majority of the studies the concept of return to scale (RTS) has been investigated in radial models, this paper, in order to recognize the type of return to scale, has expressed and proved a method based on a non-radial additive model. We also developed this method for a two-stage network and in addition to inputs and outputs; we have introduced new entry intermediate measures in the intermediate products to the system. Then, we estimate and prove the type of return to scale for this network model. At the end, examples are given to examine the proposed method. پرونده مقاله