Pablo G. Morato
School of Engineering and Design, Technical University of Munich
Theresienstr. 90
80333 Munich, Germany
I am a Senior Researcher in engineering risk analysis and artificial intelligence, @ ERA, TUM.
My research focuses on the development of probabilistic and data-driven decision-making methods for engineering systems operating under uncertainty. I work at the intersection of engineering sciences and artificial intelligence, with an emphasis on sequential decision-making, risk-aware optimization, and life-cycle management of infrastructure systems.
A central theme of my work is the use of Markov decision processes, partially observable decision models, and reinforcement learning to design adaptive inspection, monitoring, and maintenance strategies. I combine these methods with Bayesian inference, uncertainty quantification, and reliability analysis to enable informed decisions in settings characterized by partial observability, model uncertainty, and long time horizons.
My research is motivated by applications in large-scale and networked engineering systems, particularly in offshore wind energy and critical infrastructure, where scalability, interpretability, and robustness are essential. More broadly, I am interested in bridging classical reliability theory and modern AI methods, and in developing computational tools that translate probabilistic models into practical, risk-aware decision support.
news
| Dec 2025 | Co-organizing the MS139 Optimization under Uncertainty at WCCM-ECCOMAS 2026. 19 to 24 July 2026 @ Munich, Germany. |
|---|---|
| Dec 2025 | Joined the Engineering Risk Analysis Group at TUM as a Senior Researcher. |
| Dec 2023 | Presenting IMP-MARL at NeurIPS2023. |
| Sep 2023 | Joined the AiDAPT Lab at TU Delft as a Postdoctoral Researcher. |
| Jun 2023 | IMP-MARL released. |
selected publications
2024
- Farm-wide virtual load monitoring for offshore wind structures via Bayesian neural networksStructural Health Monitoring, 2024
2023
- Inference and dynamic decision-making for deteriorating systems with probabilistic dependencies through Bayesian networks and deep reinforcement learningReliability Engineering & System Safety, 2023
- IMP-MARL: a Suite of Environments for Large-scale Infrastructure Management Planning via MARLarXiv preprint arXiv:2306.11551, 2023
- Interpretation and analysis of deep reinforcement learning driven inspection and maintenance policies for engineering systemsIn 14th International Conference on Applications of Statistics and Probability in Civil Engineering (ICASP14), 2023
2022
- Optimal inspection and maintenance planning for deteriorating structural components through dynamic Bayesian networks and Markov decision processesStructural Safety, 2022
- Managing offshore wind turbines through Markov decision processes and dynamic Bayesian networksIn 13th International Conference on Structural Safety & Reliability (ICOSSAR), Shanghai, China, 2022