AI for Engineering Systems
Within the AI for Engineering Systems theme, this track highlights AI and data-driven methods for engineering analysis, design, optimization, and operational decision-making. We encourage contributions that pair methodological innovation with rigorous validation, robustness analysis, and deployment-ready considerations.
AI adoption in engineering increasingly depends on trustworthy performance, interpretability, and integration with physics, experiments, and sensor data. We invite research on physics-informed and hybrid AI approaches, inverse design, digital twins, and predictive maintenance, as well as data pipelines and evaluation protocols that support reproducible outcomes. Submissions that address uncertainty-aware decision-making, model governance, and responsible AI practices in safety-relevant settings are particularly encouraged.
Topics include (but are not limited to):
- Machine learning for prediction, classification, anomaly detection
- Physics-informed learning and hybrid (AI plus physics) modeling
- Inverse design and optimization, including multi-objective methods
- Digital twins for engineered systems and processes
- Uncertainty-aware modeling, robustness, interpretability
- AI for process control and smart manufacturing systems
- Predictive maintenance and reliability analytics
- Data pipelines, feature engineering, model deployment practices
- Integration of AI with experimental or sensor data streams
- Benchmarking, validation protocols, reproducible AI workflows
- Responsible AI, model monitoring, and auditability for high-stakes engineering