abstract: Nowadays, thanks to the rapid increase in computational resources, highly resolving computational approaches are applied to increasingly complex flow configurations of interest in technological or environmental applications. In this context, the assessment of the quality and reliability of computational results has become a topic of increasing interest. For complex simulations this task is particularly difficult, because different sources of uncertainty may have comparable effects and may interact in a nonlinear way leading to counterintuitive results. Uncertainties can also derive from a lack of knowledge of the considered problem set-up and another critical issue is, indeed, the sensitivity of the results to the different simulation parameters. Systematic sensitivity studies are difficult to be carried out when the costs of each single simulation are huge. A possible approach, is Uncertainty Quantification (UQ), in which the uncertain quantities are modeled as input random variables with a given probability distribution. These uncertainties can thus been propagated through the computational model to statistically quantify their effect on the results. Since this propagation process implies once again large computational costs, a computationally inexpensive surrogate model is usually adopted to build continuous response surfaces in the parameter space starting from a few deterministic simulations. In the lecture, some examples of application of the UQ methodology to computational fluid dynamics are presented.