Research Highlights
Probabilistic selection and design of concrete using machine learning
We develop a computer program that looks for patterns in physical properties of concrete. We apply the program to propose two concrete mix designs. The proposed concretes were experimentally tested to be robust and environmentally friendly.
Development of robust concrete mixes with a lower environmental impact is challenging due to natural variability in constituent materials and a multitude of possible combinations of mix proportions. Making reliable property predictions with machine learning can facilitate performance-based specification of concrete, reducing material inefficiencies and improving the sustainability of concrete construction. In this work, we develop a machine learning algorithm that can utilize intermediate target variables and their associated noise to predict the final target variable. We apply the methodology to specify a concrete mix that has high resistance to carbonation, and another concrete mix that has low environmental impact. Both mixes also fulfill targets on the strength, density, and cost. The specified mixes are experimentally validated against their predictions.
Our machine learning methodology has application to the broader field of accelerated materials design, allowing bespoke materials to be designed rapidly for each particular application. Furthermore, there are many examples of other verticals where information is embedded in noise, including autonomous vehicles, additive manufacturing, and information engineering, where machine learning offers the opportunity to accelerate development, understanding, and impact.
Probabilistic selection and design of concrete using machine learning
Jessica C. Forsdyke, Bahdan Zviazhynski, Janet M. Lees and Gareth J. Conduit, Data-Centric Engineering 4 (2023)