Accurately predicting the carbonation depth of recycled aggregate (RA) concrete is essential for durability assessment. Based on a dataset of 682 experimental samples, this study employed seven machine learning algorithms to develop prediction models for the carbonation depth of RA concrete. The Optuna framework was utilized to conduct 500 trials of hyperparameter optimization for these models, with the objective of minimizing the 5-fold cross-validated mean squared error. Results indicate that model performance improved significantly after optimization. Among them, the XGBoost model achieved the best performance, with a coefficient of determination (R²) of 0.9789, root mean squared error (RMSE) of 1.0811, mean absolute error (MAE) of 0.6972, mean absolute percentage error (MAPE) of 8.7932%, variance accounted for (VAF) of 97.8966%, and mean bias error (MBE) of 0.0641 on the test set. Explainability analysis using SHapley Additive exPlanations (SHAP) further revealed that exposure time is the most significant factor influencing the carbonation depth prediction. Additionally, considering that the database incorporates both natural and accelerated carbonation conditions, the samples were partitioned based on CO₂ concentration and conducts a stratified performance evaluation. The results demonstrate that the model maintains high predictive accuracy under natural carbonation as well as across different accelerated carbonation intervals, indicating that, within the scope covered by the current dataset, the proposed approach provides a highly accurate and interpretable tool for predicting the carbonation depth of recycled aggregate concrete.

With the growing demand for sustainable development in the mining industry, cemented paste backfill (CPB) materials, primarily composed of tailings, play a crucial role in mine backfilling and underground support systems. To enhance the mechanical properties of CPB materials, fiber reinforcement technology has gradually gained attention, though challenges remain in predicting its performance. This study develops a hybrid model based on the adaptive equilibrium optimizer (adap-EO)-enhanced XGBoost method for accurately predicting the uniaxial compressive strength of fiber-reinforced CPB. Through systematic comparison with various other machine learning methods, results demonstrate that the proposed hybrid model exhibits excellent predictive performance on the test set, achieving a coefficient of determination (R2) of 0.9675, root mean square error (RMSE) of 0.6084, and mean absolute error (MAE) of 0.4620. Input importance analysis reveals that cement-tailings ratio, curing time, and concentration are the three most critical factors affecting material strength, with cement-tailings ratio showing a positive correlation with strength, concentrations above 70% significantly improving material strength, and curing periods beyond 28 days being essential for strength development. Fiber parameters contribute secondarily but notably to material strength, with fiber length exhibiting an optimal range of approximately 12 mm. This study not only provides a high-precision strength prediction model but also reveals the inherent correlations between various parameters and material performance, offering scientific basis for mixture optimization and engineering applications of fiber-reinforced CPB materials.