It is widely recognized by experts that the computer-based design of chemical processes depends strongly on the correlated thermodynamic and transport properties, and the effect of property uncertainties should be incorporated into the design. The most significant source of property uncertainties on process design is from the correlations of phase equilibrium. Many approaches have been proposed, but uncertainty analysis is not a routine component of today's industrial practice, mainly because education and awareness is lacking, and the proposed methods are difficult to apply. The purpose of this paper is to report a new approach to uncertainty analysis that is intuitively appealing and easy to apply in process simulation. The proposed approach focuses on activity coefficients and is based upon the simplification that, for the purpose of perturbation, the liquid mixture can be treated as a set of pseudobinaries described by the Margules equation. The resulting perturbation in the activity coefficient of component i goes to zero when its mole fraction approaches unity, and also is relatively small when its estimated activity coefficient is close to unity (i.e., near-ideal systems are likely to be modeled more accurately than highly nonideal systems). In practice, the proposed uncertainty analysis is performed by varying a single parameter for each component in the mixture. The utility of the proposed uncertainty method is demonstrated by application to two problems: (1) a propylene-propane superfractionator for which small changes in correlated relative volatilities have a large effect on the design of the distillation column, and (2) a dehexanizer column that separates a mixture containing many close-boiling components. It is demonstrated that the proposed analysis provides quantitative insight into the effect of property uncertainties and helps to quantify the safety factors that need to be imposed upon the design. While the proposed method is applied to activity-coefficient models, the same idea is applicable to other models such as equations of state.