A rice drink with low protein and fat contents was developed by response surface methodology. Effects of three factors, levels of white rice flour (3-8 % w/v) and xanthan gum (0.01-0.05% w/v) and process temperature (80-90 degrees C) on physical stability of the drink were evaluated. Viscosity, phase separation, sedimentation, color, particle size and zeta potential of the samples were measured as response variables. Models were developed from the data to predict response variables from the factors evaluated. Models for viscosity, particle size, color, zeta potential and phase separation of the drink were found significant while that of sedimentation was not significant. Significant models were then used in optimization to achieve selected response variables, minimum phase separation, particle size and viscosity and zeta potential in range of values measured in commercial samples of milk-like drinks. Optimum conditions were found to be 3% rice flour and 0.05% xanthan gum in formula and process temperature of 80 degrees C applied for 15 min. Sucrose addition at a level of (2.5% w/v) to the formula increased acceptability of the drink in sensory analysis but it had no significant effect on physical properties of the rice drink. The drink was fortified with a caseinomacropeptide (CMP) concentrate for development of a milk-like drink for consumers with phenylketonuria. Fortification with CMP concentrate resulted in partial neutralization of zeta potential, slight sedimentation and an increase in particle size of the drink. Color and zeta potential of the drink became closer to those of low fat milk by the addition of CMP concentrate. (C) 2020 Institution of Chemical Engineers. Published by Elsevier B.V. All rights reserved.