Biography
With a PhD focused on integrated mathematical modeling, Computational Fluid Dynamics (CFD), and Artificial Intelligence applied to environmental hazards, Dr Monegaglia has since then built a cross-industry career spanning AI-powered quality inspection and defect detection in wood manufacturing and food industry, as well as in remotely managed logistics. Since 2022, he has been working at GlassFORM.ai, a Joint Venture between Bottero and Tiama, where he focuses on transforming glass manufacturing through integrated CFD and Physics-AI Digital Twins. His work aims to reduce carbon footprint while improving operational efficiency and competitiveness in glass packaging. Federico also holds an MBA in General Management. He currently serves as R&D Manager at GlassFORM.ai.
Presentation
The hollow glass industry is facing increasing pressure from rising energy costs and shrinking margins, despite its strong positioning as a sustainable packaging solution. Energy demand remains a key factor limiting competitiveness compared to alternative materials. To address this challenge, we developed Physics-AI Digital Twin-based solutions with a dual objective: reducing energy consumption while increasing productivity through reduction of defects, waste, and downtime.
This work presents a physically consistent modeling framework based on Physics-informed Artificial Intelligence, where Computational Fluid Dynamics (CFD) models are accelerated and approximated in real-time using AI techniques. These hybrid CFD-AI models are deployed as Digital Twins of key process units, including the forehearth conditioning system, the feeder, and the gob forming process.
The proposed approach enables real-time prediction of process states and optimization of control setpoints under varying operating conditions. By embedding these Digital Twins within advanced control strategies, the system continuously identifies energetically optimal configurations while ensuring target glass temperature and homogeneity.
Experimental validation on industrial use cases demonstrates that simultaneous improvements in energy efficiency and process stability are achievable. Results indicate measurable reductions in energy consumption alongside improved production consistency and reduced defect rates.
The presented methodology represents a step change in process control for the glass industry, demonstrating that “more with less” is not only theoretically feasible but practically achievable. This approach provides a scalable pathway to enhance competitiveness and sustainability, supporting the transition of hollow glass manufacturing toward more efficient and adaptive production paradigms.