Ensemble machine learning models for predicting strength of concrete with foundry sand and coal bottom ash as fine aggregate replacements

Natural sand scarcity and environmental concerns have encouraged the adoption of industrial waste materials in sustainable concrete. Foundry sand (FS) and coal bottom ash (CBA), both industrial waste materials, offer potential as partial replacements for natural sand. However, predicting the compressive strength (CS) of such mixes is complex due to nonlinear interactions among their components. This study proposes a novel, machine learning-based framework for predicting the CS of concrete incorporating FS and CBA. A dataset of 172 mix designs was compiled from published literature. Nine machine learning models were evaluated, including traditional regressors and ensemble methods. The Extreme Gradient Boosting (XGBoost) model achieved the highest accuracy, with R² = 0.983, RMSE = 1.54 MPa, and MAPE = 3.47%. The key innovation of this work is the application of ensemble machine learning models for strength prediction in dual-waste concrete, which has been minimally explored in prior research. Feature importance analysis identified curing duration, superplasticizer dosage, cement content, and water-to-cement ratio as dominant predictors. This research demonstrates the effectiveness of artificial intelligence driven approaches in sustainable concrete mix design. By minimizing the need for trial and error experiments, the proposed method accelerates decision-making, reduces costs, and supports the circular economy by encouraging the use of industrial byproducts in construction.