AI for Engineering Optimisation

2025 UTS FEIT Research Excellence Guest Lecture Series

Prof. Pascal Van Hentenryck

Abstract
In many industry settings, the same optimisation problem is solved repeatedly for instances taken from a distribution that can be learnt or forecast. This in-personal talk explores how to accelerate these parametric optimisation problems to meet real-time constraints present in many applications. It reviews the concept of optimisation proxies that learn the input/output mappings of parametric optimisation problems, as well as novel techniques in "learning to optimise" for dimensionality reduction. These innovative methodologies are demonstrated on industrial problems in grid optimisation, end-to-end supply chains, logistics, and transportation systems. These results show that the fusion of AI and Optimisation can deliver outcomes that the two technologies cannot achieve independently.

Biography
Professor Pascal Van Hentenryck is the director of the US NSF AI Institute for Advances in Optimisation (AI4OPT), the director of Tech AI (the AI hub at Georgia Tech), and the A. Russell Chandler III Chair and Professor at Georgia Tech. He was a Professor of Computer Science at Brown University for over 20 years and led the optimisation research group at the National ICT Research in Australia. His current research focuses on AI for Engineering with applications in energy systems, supply chains and manufacturing, healthcare, corporate systems, and mobility. Van Hentenryck is a Fellow of AAAI and INFORMS, the recipient of two honorary degrees, and the recipient of numerous research and teaching awards. He has developed constraint programming and optimisation systems that have been in commercial use for several decades.