Physics-Informed Machine Learning for Corrosion Prediction

About the Project

The global infrastructure landscape is dominated by ageing assets, many of which were not designed with today’s loading and environmental challenges in mind. Extending the service life of existing structures has become a vital strategy for minimising the environmental and economic costs associated with demolition and new construction. However, one of the most pressing threats to structural integrity is the corrosion of reinforcing steel, which compromises safety, durability, and sustainability.

Current corrosion prediction models often fall short because they rely on oversimplified assumptions and fail to capture the complex, variable environmental conditions that structures face in the real world. This leads to premature failures, costly maintenance, and inefficient resource allocation. The problem is global in scope, affecting both developed and developing nations, and demands innovative, scalable solutions.

This PhD project aims to revolutionize corrosion prediction by integrating physics-informed machine learning (PIML) with domain-specific engineering knowledge. By embedding physical laws and corrosion mechanisms into data-driven models, the research will produce more accurate, generalizable, and interpretable predictions of corrosion progression under diverse environmental conditions. The ultimate goal is to develop a decision-support framework that enables engineers and asset managers to assess the remaining service life of infrastructure more reliably, prioritize interventions, and design maintenance strategies that are both cost-effective and environmentally responsible. By bridging the gap between AI and civil engineering, this project contributes to the global effort to build resilient infrastructure, safeguard public safety, and reduce the carbon footprint of the built environment.

Funding Notes

The funding available is for UK home student. However, if you are an international student and are willing to manage your own living costs, you are welcome to apply. Alternatively, I can support international students through external funding applications through avenues such as Common wealth fellowships, Schlumberger Foundation, etc.