A Computational Fluid Dynamics model based on a two-fluid Eulerian description of the mass and momentum equations, two-equation mixture turbulence model and energy equation for the liquid phase is developed to predict the void distribution in a BWR assembly specified in the BFBT benchmark study. The effects of uncertainties in operating conditions, (pressure mass flow rate, inlet temperature, and power input) and model parameters, (momentum and heat exchange between the phases, heat transfer from the rods) has been analyzed. The results of both analysis indicate that the model predictions are more sensitive to experimental and modelling uncertainties in the low void fraction regions of the computational domain. The model predictions are in agreement with the experimental data with experimental and model uncertainties taken into account. Finally a sensitivity analysis of the model parameters has been performed and parameters of wall heat flux and heat transfer between the phases models have been identified as the major parameters that influence the phase distribution. (C) 2009 Elsevier B.V. All rights reserved.