The transition toward Industry 5.0 introduces a new paradigm for manufacturing systems, shifting the focus from pure automation to human-centric, sustainable and resilient industrial ecosystems. In this evolving landscape, production environments are increasingly modeled as complex cyber-physical systems characterized by multi-scale interactions, nonlinear dynamics, uncertainty and strong interdependencies between physical processes, digital infrastructures and human operators.
Recent advances in Artificial Neural Networks and Deep Learning are redefining Advanced Manufacturing by enabling intelligent, data-driven and self-optimizing production processes. Among manufacturing technologies, Additive Manufacturing (AM) stands out as one of the most promising and challenging domains for the deployment of Artificial Intelligence.
Neural networks, Physics-Informed Neural Networks (PINNs) and hybrid physics–data-driven approaches are emerging as key enablers for high-fidelity surrogate modeling, learning under limited data regimes, real-time control architectures, autonomous defect detection and multi-objective optimization, including energy efficiency and sustainability metrics.
Topics of Interest
- Neural and hybrid models for AM process simulation
- Physics-informed and knowledge-guided learning
- Data-driven digital twins for AM systems
- Real-time monitoring, control, and anomaly detection
- Multi-objective and sustainability-aware process optimization
- Robotics integration in intelligent AM environments
- Robustness, interpretability and certification of AI in safety-critical contexts
By fostering an interdisciplinary and application-oriented dialogue, the workshop seeks to bridge the gap between methodological advances in neural networks and their reliable deployment in industrial AM systems contributing to the evolution of Additive Manufacturing toward intelligent, sustainable and resilient Industry 5.0 production paradigms.