Semi-Plenary Lecture 9

Room: Audimax
Time: Thursday, 1. October 2026, 10:20 - 11:00
Chair: Ioannis A. Kougioumtzoglou

Matthias Faes

Prof. Dr.-Ing.
Reliability Engineering
TU Dortmund, Germany

Uncertainty quantification and reliability analysis for complex dynamical systems

Uncertainties are ubiquitous in the simulation of the structural dynamic behaviour of structures and machines, particularly when the effects of natural phenomena such as earthquakes or wind loads must be taken into account. Stochastic processes provide a rigorous framework for representing uncertainties and the associated space and time correlations of uncertain loads, grounded in the well-established principles of probability theory. Accounting for the stochastic nature of external loads, as well as for uncertainties inherent in the system itself, enables the approximation of structural failure probabilities through stochastic dynamic simulations. In practice, however, estimating the failure probability of complex dynamical systems typically relies on computationally intensive simulation models. Moreover, the presence of stochastic processes often leads to problem formulations involving hundreds or even thousands of random variables. Together, these aspects pose substantial challenges for the reliable simulation of realistic engineering systems.

In this talk, I will present practical numerical schemes that aim to address these challenges. The presentation will draw on recent work in the reliability analysis of stochastic dynamical systems, with a focus on advances in surrogate modelling, including functional dimensionality reduction and Bayesian active learning, as well as on efficient sampling strategies such as multi-domain line sampling and importance line sampling.

Short Bio

Matthias Faes, a full Professor in Reliability Engineering at TU Dortmund (appointed in 2022 at age 30), served as a post-doctoral fellow of the Research Foundation Flanders (FWO) at KU Leuven and was affiliated with the Institute for Risk and Reliability at the University of Hannover as an Alexander von Humboldt Fellow. Before, he obtained his PhD from KU Leuven in 2017. His research addresses theoretical and numerical approaches to uncertainty quantification and reliability analysis, focusing on inverse and data-driven methods, surrogate modeling, stochastic fields, design optimization, and imprecise probabilities. He has receved multiple awards and honors, including the 2017 ECCOMAS European PhD award, the 2019 ISIPTA-IJAR Young Researcher Award, and the 2023 EASD Junior Research Prize. He has published over 90 journal papers, with an H-index of 27. Currently, he is associate editor of Mechanical Systems and Signal Processing, Reliability Engineering and System Safety and Associate Managing Editor of both part A and B of the ASCE – ASME Journal of Risk and Uncertainty in Engineering Systems.