A reliable mathematical model to predict the performance of double-deck screens from an industrial urea granulation circuit is provided in this work. The classification in each deck was described by the empirical model reported by Karra [Karra, V. K. Development of a Model for Predicting the Screening Performance of a Vibrating Screen. CIM Bull. 1979, 72, 804.] for the mining industry. The estimation of model parameters was based on industrial data collected from two large-scale double-deck screens belonging to a urea granulation plant. Prior to parameter fitting, data reconciliation of the solid streams was performed, with the aim of determining the unmeasured mass flows and the reconciled particle size distributions that fulfill the material balances for every size class. The results indicate that the fitted model reproduces in an accurate way the performance of the urea granules classification device. Thus, the screen mathematical representation can be used as a module of a plant simulator for many existing urea granulation circuits, which (independently of the technology) have similar industrial vibrating double-deck screens to attain the marketable product granulometry. In addition, the fitted model allows finding the optimal screen apertures to achieve the desired product attributes and recycle ratios.