Safety-critical systems require rigorous analysis to guarantee their safe, reliable, and correct operation. In this talk, I will discuss the importance of formal modeling and model checking in the design and analysis of such systems. Our approach is based on probabilistic models, specifically Markov chains, to explicitly capture uncertainties, such as the probability of component failures. These models can then be analyzed using probabilistic model checking, enabling the evaluation of a wide range of quantitative metrics and supporting a variety of analysis techniques.

The talk will focus on (dynamic) fault trees, a prominent reliability model that describes how failures occur and propagate through a system. Despite their intuitive graphical representation, dynamic fault trees have intricate semantics that require careful formalization. We translate the fault trees into Markov chains to enable efficient and automated analysis via model checking. I will illustrate the benefits of this approach with an industrial case study on autonomous vehicles.

Finally, I will present ongoing research on automatically learning fault tree structures from failure data. By combining genetic algorithms with structural information, we aim to synthesize fault trees that are both accurate and concise - paving the way toward data-driven, model-based safety analysis.

Matthias Volk is an Assistant Professor in the Formal System Analysis group at Eindhoven University of Technology (TU/e). His research goal is to improve the safety and reliability of complex, safety-critical systems using formal methods. In his research, Matthias Volk develops rigorous, automated techniques for modelling and analysing safety-critical systems, with a focus on probabilistic model checking. He is interested in providing optimised implementations in mature tools for use by other researchers, as well as applying the approaches in industrial collaborations.

Matthias Volk received his B.Sc. and M.Sc. degrees in Computer Science from RWTH Aachen University. He received his Ph.D. from RWTH Aachen University under the supervision of Joost-Pieter Katoen. He then worked as a postdoctoral researcher at the University of Twente. Since 2023, he is an assistant professor in the Formal System Analysis group at Eindhoven University of Technology.