Predicting the Compressive Strength of Ash‐Based Concrete Using Machine Learning Approach: Paving the Way to Sustainable Concrete

Industrial wastes, agricultural byproducts, and rock dust are considered suitable replacements for cement in eco‐friendly concrete production due to their comparable chemical properties. This study evaluates 13 ash sources as potential cement alternatives, explores machine learning (ML) algorithms to predict compressive strength () of ash‐blended concrete, and determines the optimal amount of ash to replace cement for sustainable concrete production. A dataset of 1148 concrete samples incorporating various ash amounts was collected, and five ML models, including artificial neural network (ANN), support vector regression (SVR), random forest (RF), decision tree (DT), and extreme gradient boosting (XGBoost), were developed to predict the of ash‐based concrete. The statistical analysis showed that concrete mix incorporating calcined red clay–rice husk (CRC–RH) ash consistently achieved the highest strength across all curing periods. This was followed by mixes containing water hyacinth ash and marble dust. This superior performance is likely attributed to the higher content of tricalcium aluminate and silicate, which enhance workability and early strength development. Among the ML models, XGBoost showed superior predictive accuracy, with the overall performance ranking as: XGBoost >RF >SVR >ANN >DT. Furthermore, this study determined that replacing 7.5%–12.4% of cement with ash maintained the optimal . These findings highlight the potential of waste‐driven alternative construction materials in sustainable concrete production and demonstrate ML as a powerful tool for efficiently predicting ash‐blended concrete strength, paving the way for optimized concrete mix design for greener construction.