Special Session Organized by

Aldo Dagnino, North Carolina State University, USA, Marcel Dix, ABB Corporate Research Center, Germany, Franz C. Kunze, Ruhr University Bochum, Germany, Gianluca Manca,, Ruhr University Bochum, Germany, and Mehmet Mercangöz, Imperial College London, UK.

---

Download Call for Papers

Click here to download the special session cfp.

---

Focus

The deployment of Industrial AI systems in manufacturing and process industries demands robust solutions capable of maintaining consistent performance despite variability and disturbances. Robustness ensures reliable operation under diverse conditions, minimizing risks. However, uncertainties from incomplete data, model limitations, or unexpected scenarios must also be quantified and communicated effectively. This session addresses strategies to enhance AI robustness, integrate uncertainty quantification, and convey insights through human-machine interfaces. It targets not only operators but also engineers and professionals relying on AI for planning, optimization, and decision-making, supporting trust and informed use of AI in industrial contexts.

---

Topics under this session include:

Robust machine learning methods for industrial applications

Robustness and uncertainty quantification in cyber-physical systems

Statistical process monitoring and anomaly detection under uncertainty

Out-of-distribution detection in industrial settings

Explainable AI (XAI) techniques for industrial applications

Uncertainty visualization and risk communication in human-machine interfaces (HMIs)

AI-driven decision support systems for operators and engineers

Addressing hallucinations and increasing robustness of LLMs in industrial AI

Data quality analysis and monitoring for training and inference in AI models

Data augmentation and synthetic data strategies for industrial AI

MLOps practices for maintaining robust and reliable AI system

Hybrid AI approaches combining data-driven and physics-based models

Incorporating domain knowledge into AI systems for enhanced reliability

Benchmarking and evaluation of uncertainty estimation methods in industrial contexts

Case studies demonstrating robustness and uncertainty management in industrial AI