Rio McDade

Project: Machine learning for material performance predictions across climates

Supervisors: Vladimir Gusev, Keiller Nogueira

Industry Partner: Beckers Group

What inspired you to pursue this project and join the DAMC CDT?

My inspiration to pursue a PhD arises from a personal motivation to drive impact in real world settings.

My academic background combines a BSc in Computer Science and an MSc in Advanced Computer Science from the University of Liverpool. Pursuing a PhD represents a rewarding challenge that allows me to extend interests and skills developed within the University’s Computer Science community, particularly in applied machine learning.

More specifically, my research interests include computational sustainability, efficient algorithms, and machine learning. My project is supported by Beckers Group, a leading global supplier of sustainable coatings and paints. This partnership allows these research interests to be applied at scale, focusing on the development of sustainable and efficient methods that can support material deployment worldwide over long time horizons.

Finally, the DAMC CDT fosters a collaborative environment among students through its academic and industrial partnerships, encouraging a strong research community. Given the applied nature of my research, the CDT provides an opportunity to engage with this interdisciplinary environment while further developing skills relevant to real world research and long term Research and Development.

What is your research project about, and what impact do you hope it will have?

My research project involves handling multiple forms of data, including material performance, climate, and chemistry, to develop machine learning models capable of quantifying and predicting the performance of materials deployed globally. Material performance is typically assessed through measurable degradation responses, such as colour and gloss. These responses can be influenced by climatic cycles, material composition, and colour itself. The project therefore utilises Gaussian Processes to provide regression-based predictions of material degradation. Using Gaussian Processes, the project aims to estimate performance across different materials using data grounded in chemistry and climate, supporting decisions around material deployment, warranties, and industrial Research and Development. Because real world material testing requires long evaluation periods, costs accumulate significantly. A robust predictive approach could reduce reliance on long deployment cycles by enabling performance evaluation directly from development data rather than extended real world testing.

What has been the most exciting or rewarding part of your PhD journey so far and how does your project benefit from being part of an interdisciplinary CDT?

My project benefits from being part of an interdisciplinary CDT such as DAMC, where collaboration across disciplines is actively encouraged. My research draws on machine learning and chemistry to understand material degradation, and the CDT provides opportunities to engage with expertise beyond computer science. In addition, working with the industrial sponsor, Beckers Group, provides valuable exposure to industrial expectations and constraints, allowing research directions and design choices to remain aligned with real world applications and long term Research and Development.