The main objective of this work is to develop a predictive model for predicting the product quality of vacuum gas oil (VGO) hydrocracking process. Experimental data were obtained using a pilot plant hydrocracking catalytic reactor loaded with the same catalyst type used in a local refinery. Two sets of experimental runs were conducted under various operating conditions. The first one consisted of 18 runs and was used for parameter estimation, while the second set consisted of 29 runs and was used for model validation. Distillation curves of the cracked products were obtained using the simulated distillation (SimDist) test. A distribution model based on probability density function was used to develop the predictive model. The distribution model presents the boiling point as a function of the distilled weight fraction. Model parameters were estimated and related to the specific gravity of the cracked product. Model validation results showed that the proposed model is capable of predicting the distillation curves of the hydrocracked products accurately, especially at high operating severity. Simplicity and accuracy of the developed model makes it suitable for online analysis, to estimate the conversion as well as the product distribution of hydrocracking units in refineries. (C) 2010 Elsevier Ltd. All rights reserved.