Methods of Uncertainty Analysis and Quantification

Course content

Students can formulate and name elementary rules of probability theory and different ways to describe probability distributions. They can model technical/physical systems in a stochastic way using random variables. The students are further able to apply Monte Carlo and stochastic spectral methods to quantify uncertainties and also to assess the impact and propagation of uncertainties in models through global sensitivity analysis. Moreover, they are able to evaluate the numerical efficiency of the aforementioned methods. The students are also able to outline the principles of data-driven approaches to uncertainty analysis.

Probability and random variables, advanced Monte Carlo methods, stochastic quadrature, stochastic spectral methods, global sensitivity analysis, data-driven uncertainty quantification