article · Scientific Reports
Determining the mechanical properties of high-strength concrete, particularly compressive strength, is essential for safety purposes. Standard physical laboratory testing is both costly and time-consuming, but artificial intelligence methods offer an alternative to reduce these demands. A machine learning approach was established using the Python programming language to forecast concrete compressive strength based on original experimental test data. The models analysed essential mix design parameters, including cement content, silica fume, water, superplasticiser, sand, gravel, and curing age. Several regression models were investigated and optimised via hyperparameter tuning, with evaluation conducted using Mean Absolute Error, Mean Squared Error, and R-squared metrics. The XGBoost model achieved the best performance with an R-squared value of approximately 0.94, confirming that machine learning can deliver accurate and reliable predictions for high-strength concrete performance.
Ensuring concrete meets strict safety specifications is critical across building and infrastructure projects, but physical testing creates project delays and requires expensive laboratory resources. By predicting compressive strength directly from material proportions and curing age, computational methods offer a faster, lower-cost way to evaluate concrete properties without relying entirely on destructive testing.
This methodology could enable construction contractors, concrete suppliers, and materials testing laboratories to rapidly evaluate mix designs. Because the work is validated on experimental laboratory data rather than deployed in an industrial tool, it represents an applied and tested stage of development. Moving towards real-world adoption would require packaging the Python models into accessible quality-control software for engineering teams.
AI-generated from the published abstract. Always read the original work before citing.
Abstract Identifying the mechanical properties of High Strength Concrete (HSC), particularly compressive strength, is critical for safety purposes. Concrete compressive strength is determined by using laboratory experiments, which are costly and time-consuming. Artificial intelligence (AI) methods reduce time and money. This research proposes a machine learning (ML) model using the Python programming language to predict the compressive strength of HSC. The dataset used for the models was obtained from original experimental tests. Important parameters, namely cement content, silica fume, water, superplasticizer, sand, gravel, and curing age, were taken as input to predict the output, which was the compressive strength. Various regression models were investigated for the prediction of outcome compressive strength. To optimize the models, hyperparameters were tuned, and measures such as Mean Absolute Error (MAE), Mean Squared Error (MSE), and R-squared were used for evaluation. XGBoost ( R 2 ≈ 0.94) outperformed other models, demonstrating ML’s potential for HSC strength prediction and demonstrated that Python can be successfully applied to establish accurate and reliable prediction models.
This page summarises published work. The authoritative version sits with the publisher.
DOI: 10.1038/s41598-025-10342-1
Is something wrong with this record? Report it or request removal.
Discussion
Have you built on this work, tried to replicate it, or seen it applied in practice? Share what you know. Verified researchers and MARATTO™ domain experts can open a discussion, and any member can reply. Contributions are reviewed before they appear.
No discussion yet. Open the first thread.
New to MARATTO™? Create a free account.
