Xiaohui Chen

Welcoming Dr Xiaohui Chen as a DAFNI Fellow!

Xiaohui is Associate Professor, Founder Director of Geomodelling and AI Group, and Deputy Director of Research and Innovation at the School of Civil Engineering at the University of Leeds.

“I am delighted to be selected as a DAFNI Fellow. This fellowship provides an exciting opportunity to advance next-generation AI-for-Science frameworks that combine physical modelling with data-driven approaches to support resilient national infrastructure systems,” Xiaohui comments.

“Through the DAFNI Fellowship, we hope to develop trustworthy and reusable digital tools for infrastructure and environmental risk prediction, while strengthening collaboration between academia, industry, and government,” he adds.

Xiaohui is one of 10 new DAFNI Fellows, selected from 116 applicants for the nationally competitive programme.

Each DAFNI Fellow has been awarded £10,000 to support staff time, travel, knowledge exchange, conferences, and dissemination activities. The programme also helps Fellows build networks across government, industry and academia through workshops, conferences and wider community engagement.

Xiaohui’s Fellowship will directly support DAFNI’s mission to enable trusted, reusable, and policy-relevant infrastructure science.

He explains, “I develop physics-informed AI and digital-twin frameworks that improve the reliability, transparency, and interoperability of models used for national infrastructure resilience, particularly for soil–water–infrastructure interactions under climate stress.”

Xiaohui is founder and director of the Geomodelling and Artificial Intelligence Group, which includes 2 postdoctoral researchers, 11 PhD students, and 9 MSc/MEng students. The group has secured over £2.7 million in funding from research councils (including NERC, EPSRC, Horizon EU, and CSC) and industry partners. Focusing on cutting-edge theoretical, numerical, mathematical, and AI modelling, the group has proven impacts in various industries such as petroleum, nuclear waste, carbon capture and storage, landfill, tunnelling, water, and agriculture.

His research interests are focused on Physics-Informed Machine Learning (PIML) and Coupled Multiphysics Modelling. His research has broad industrial applications, including geotechnical engineering, water research, subsurface disposal, energy, sustainable agriculture, and environmental engineering. Additionally, he is a Fellow of the Institution of Environmental Sciences and a Chartered Environmentalist (UK).

Xiaohui has expertise in providing international academic leadership in physics-informed machine learning (PIML) for infrastructure and environmental systems, position as Task Force Leader on PIML within the International Society for Soil Mechanics and Geotechnical Engineering (TC309), and service as the UK Representative on international technical boards for machine learning and environmental geotechnics. He also contributes to UK standards development through the British Standards Institution.

He concludes, “My research can benefit significantly from the DAFNI Platform by enabling the integration, testing, and reuse of physics-informed AI models for complex infrastructure–environment systems at scale. Much of my work focuses on coupling physical laws with data-driven methods to model soil–water–infrastructure interactions, digital twins for buried infrastructure, and climate-driven environmental risks. DAFNI provides a unique environment to deploy these models alongside national datasets, enabling cross-project comparison, benchmarking, and

Xiaohui has authored 70+ peer-reviewed publications and a UKRI-funded research monograph, with research spanning geotechnical engineering, environmental systems, and artificial intelligence.
Recent publications on Physics-Informed Machine Learning include:

1: Chen, L, …, Chen, X. and Guo, H., (2025). Unveiling the Role of Wetland Strategies in Antibiotic Risk Reduction across China by Machine Learning. Environmental Science & Technology. Open Access Link
2: Yuan, B., Choo, C. S., Yeo, L. Y., Wang, Y., Yang, Z., Guan, Q., … & Chen, X. (2025). Physics-informed machine learning in geotechnical engineering: a direction paper. Geomechanics and Geoengineering, 1-32. Open Access Link