This study investigates the combined use of waste glass aggregate (WGA, 0–60% volumetric replacement of natural coarse aggregate) and silica fume (SF, 0–15% replacement of cement) in steel fiber-reinforced self-compacting concrete. The experimental program covered fresh, mechanical, and durability properties, along with production cost, embodied CO₂, and scanning electron microscopy (SEM) observations. The results showed that 15% SF improved strength and durability, whereas WGA contents above 45% reduced fresh performance. The optimum mixture, containing 15% SF and 30% WGA, increased the 90-day compressive, splitting-tensile, and flexural strengths by 23.72%, 28.90%, and 16.50%, respectively, compared with the control. This mixture also maintained surface water absorption below 7.81% and ultrasonic pulse velocity (UPV) values of 3532–3974 m/s. Even at 60% WGA and 15% SF, production cost and embodied CO₂ were reduced by 4.59% and 5.03%, respectively. SEM observations showed that high WGA contents increased voids, microcracks, and ettringite formation, which contributed to strength loss. Gene Expression Programming (GEP) models were developed using WGA, SF, and durability-related indicators as input variables. The models showed reliable predictive performance for strength, absorption, and water penetration depth. Overall, the findings support the practical use of WGA and SF for producing more sustainable and cost-effective self-compacting concrete.
N. Safaeian Hamzehkolaei, Iman Afshoon, Amirhossein Davarpanah Tanha Ghochan· Multiscale and Multidiscipli...· 1 citation
Large rupture strain fiber-reinforced polymer (LRS-FRP)-confined concretes are increasingly used in safety–critical infrastructure due to their high ductility and load-carrying capacity; however, accurate prediction of compressive strength (CS) in non-circular sections remains challenging due to non-uniform confinement induced by geometric irregularities, which limits the reliability of existing empirical models and design codes developed mainly for circular sections. To address this limitation, this study develops a reliability-oriented, data-driven framework that combines Bayesian-optimized ensemble machine learning, model interpretability, and uncertainty quantification. Six algorithms including random forest (RF), extremely randomized trees (ERT), extreme gradient boosting (XGBoost), histogram based gradient boosting (HistGBM), light gradient boosting machine (LightGBM) and categorical boosting (CatBoost) were trained using an experimental database of 174 non-circular LRS-FRP-confined concrete specimens. Model interpretability was achieved using Shapley additive explanations (SHAP), while predictive reliability was systematically evaluated through uncertainty-aware performance assessment. All models demonstrated strong generalization, with testing coefficients of determination (R
2
) ranging from approximately 0.96–0.99, and boosting-based methods consistently outperforming bagging approaches. CatBoost (testing R
2
≈ 0.985) exhibited the best overall performance, the lowest prediction errors, and the most reliable uncertainty estimates. Accordingly, the overall performance ranking was identified as CatBoost > HistGBM > XGBoost > ERT > LightGBM > RF. The results clearly indicate that high predictive accuracy alone is insufficient for reliable modeling of non-circular LRS-FRP-confined concrete and that uncertainty-aware evaluation is essential. SHAP-based analysis yielded physically consistent insights, identifying LRS-FRP thickness, unconfined concrete strength, and section corner radius as the dominant contributors to CS, while highlighting the critical role of post-transition LRS-FRP stiffness in sustaining effective confinement. Overall, the proposed framework offers an interpretable and reliability-aware alternative to conventional models and provides a robust predictive tool for engineering design and assessment of non-circular LRS-FRP-confined concrete.