Diego Galar

As industrial systems transition from being simply "smart" to becoming truly "genius," we are witnessing the birth of hybrid intelligence—where data-driven AI models work in concert with physics-based and expert knowledge. This convergence is enabling a new generation of assets that are not only connected and predictive, but context-aware, adaptive, and capable of self-preservation. 

In this presentation, we will explore how the integration of machine learning with domain-specific models—such as physics-informed neural networks and hybrid digital twins—empowers systems to anticipate failures, orchestrate responses, and optimize performance in real time. These hybrid architectures blend the agility of AI with the robustness of engineering knowledge, ensuring reliable decisions even in complex or uncertain industrial environments. 

We’ll share practical examples from large-scale European implementations and discuss how this paradigm is revolutionizing maintenance, energy efficiency, and operational resilience across critical industries.