Uncertainty quantication (UQ) is the process of quantitative characterization and prop-agation of input uncertainties to the response measure of interest in experimental and com-putational models. The input uncertainties in computational models can be either aleatoryi.e. irreducible inherent variations or epistemic i.e. reducible variability which arises fromlack of knowledge. Previously, it has been shown that Dempster-Shafer Theory of Evidence(DSTE) can be applied to model epistemic uncertainty in case of uncertainty informationcoming from multiple sources. The objective of this paper is to model and propagatemixed uncertainty (aleatory and epistemic) using DSTE. In specic, the aleatory vari-ables are modeled as Dempster-Shafer structures by discretizing them into sets of intervalsaccording to their respective probability distributions. In order to avoid excessive compu-tational cost associated with large scale applications, a stochastic response surface basedon point-collocation non-intrusive polynomial chaos has been implemented as the surro-gate model for the response. A convergence study for accurate representation of aleatoryuncertainty in terms of minimum number of subintervals required is presented. The mixedUQ approach is demonstrated on a numerical example and high delity computational uiddynamics study of transonic ow over RAE 2822 airfoil.