By now, there is a vast array of uncertainty quantification methods, but they often rely on regularity assumptions on the physical model that can be difficult to verify in practice. We offer expertise in selecting the best uncertainty quantification algorithm for the selected problem, with a special emphasis on Monte Carlo, quasi Monte Carlo and multilevel Monte Carlo. Furthermore, we can post-process (or in-situ process, where relevant) the data generated to compute derived quantities such as the probability density function.
As complementary expertise, we can also incorporate observations through data assimilation, and hardware-accelerated numerics for dealing with low regularity problems.