Insights from DTE 2025 & AICOMAS 2025 in Paris
The 3rd IACM Digital Twins in Engineering Conference (DTE 2025) and the 1st ECCOMAS Artificial Intelligence and Computational Methods in Applied Science (AICOMAS 2025) brought together experts at the forefront of AI-driven engineering and digital twinning. We were thrilled to actively contribute with keynote inputs and engage in discussions, exchange ideas, and explore cutting-edge advancements in the domain.
With 500+ presentations across 47 minisymposia, the conference featured cutting-edge research and plenary talks from leading scientists, including:
- Laura De Lorenzis – Machine Learning, Data, and Physics for Constitutive Material Modeling
- George Karniadakis – Deep Neural Operators as Foundation Models for Digital Twins
- Eleni Chatzi – Physics-Enhanced Machine Learning for Monitoring & Twinning | An Exercise in Balance
- Gitta Kutyniok – Trustworthy and Sustainable AI: From Mathematical Foundations to Next Generation AI Computing
- J. Nathan Kutz – Modern Sensing and Physics Learning with Shallow Recurrent Decoders
Beyond the plenary sessions, our ETH Zurich team was actively involved in key thematic sessions, including:
✨ MS051A - Physics-Enhanced Machine Learning for Structural Health Monitoring, co-organized with Alice Cicirello (University of Cambridge) and Elizabeth Cross (The University of Sheffield), and co-chaired with Beatriz Moya:
- A Dual Updating Scheme for Damage Identification Based on Physics-Informed Neural Networks – led by Vasiliki Panagiotopoulou, in collaboration with Claudio Sbarufatti, Konstantinos Vlachas, and Eleni Chatzi
- Physics-Embedded Graph Neural Networks for Rapid Assessment of Electrical Substations in Extreme Events – led by Beatriz Moya and Huangbin Liang, in collaboration with Francisco Chinesta and Eleni Chatzi
✨ MS030 - Scientific Machine Learning and Uncertainty Quantification for Robust Digital Twins, co-organized with Dimitris Loukrezis (Siemens), Dimitris Giovanis (Johns Hopkins University), and Vissarion Papadopoulos (NTUA)
✨ Session on Digital Twins for Predictive Decision Making, organized by Andrea Manzoni and Matteo Torzoni, featuring:
- Adaptive Reduced-Order Modeling for Engineering Decision Support – led by Konstantinos Vlachas, in collaboration with Antonios Kamariotis and Eleni Chatzi
- Data-Driven Digital Twinning for Railway Network Optimal Maintenance Planning with Multi-Agent Reinforcement Learning – led by Giacomo Arcieri and Christophe Muller, in collaboration with David Häner (SBB CFF FFS) and Eleni Chatzi
- Structural Damage Detection Based on Aerodynamic Pressure Measurements – led by Philip Franz, in collaboration with Imad Abdallah, Gregory Duthé, Julien Deparday, Ali Jafarabadi, Alexander Popp, Sarah Barber, and Eleni Chatzi
A huge thank you to the IACM and ECCOMAS communities, the organizers, and all participants for making this event a success! Looking forward to continuing these discussions and collaborations.