Strengthening Supply Chain Management with AI and Mathematical Insight
Recent global shocks, such as the COVID-19 pandemic and geopolitical disruptions, made it abundantly clear how important and fragile modern supply chains truly are. Their management ensures the seamless flow of goods and resources across interconnected networks. “While AI already plays a significant role in this field, most solutions rely too heavily on vast historical datasets, which can become obsolete or a single, highly detailed simulation, which might not reflect reality,” explains Nicola Serra.
While AI already plays a significant role in this field, most solutions rely too heavily on vast historical datasets, which can become obsolete or a single, highly detailed simulation, which might not reflect reality.
Serra, a physics professor at the University of Zurich (UZH), has received a BRIDGE Discovery Grant worth CHF 1.7 million, together with ETH mathematics professor Alessio Figalli. Their goal is to chart a different course for supply chain management by combining optimal transport theory – a mathematical framework for finding the most efficient way to move mass from one configuration to another – with advanced AI methods.
Improve competitiveness
“We want to employ an ensemble of simplified simulations, each informed by real data and constrained by optimal transport principles,” says Figalli. The project will culminate in an AI-driven platform that enhances adaptive robustness by learning from and generalizing across a wide range of scenarios. Since this approach only requires limited data to be effective, the barrier to adoption is lowered, enabling small and medium-sized enterprises to use the method. This will improve the competitiveness of these companies, leading to significant economic and societal impact.
We want to employ an ensemble of simplified simulations, each informed by real data and constrained by optimal transport principles.
The two investigators bring complementary expertise that is critical to the project’s success: Nicola Serra has extensive experience applying advanced AI solutions to large-scale, data-intensive experiments at CERN, while Figalli is a leading mathematician and Fields Medalist with seminal contributions in the field of optimal transport and robust optimization. They plan to launch a spin-off to facilitate technology transfer through licensing and early-stage collaborations, which will bring the solution to market quickly.