ABSTRACT Basalt Fiber Reinforced Concrete (BFRC) is being recognized as an eco-friendly advanced material with reduced environmental footprint, higher mechanical performance and long-term durability. However, its compressive strength prediction still appears to be a difficult problem due to the nonlinearity caused by the interaction of the mix components. This paper demonstrates a production-quality hybrid ML model to predict 28-day compressive strength of BFRC, providing an economical alternative to laborious and expensive laboratory-based testing. A well-defined database of 450 samples is generated involving the essential parameters, such as cement (440 kg/m3), SCMs (110 kg/m3), basalt fiber (0–4.25 kg/m3), fine aggregate (740 kg/m3), and coarse aggregate (975 kg/m3), water (121 kg/m3), superplasticizer (3.52 kg/m3), and curing period (3–365 days). The pre-processed and normalized data were partitioned into the training set (80%) and the test set (20%). Five ML models Gradient Boosting (GB), Cat Boost (CB), Light GBM (LGB), and their hybrid ensembles: GB+CB, GB+LGB were trained and compared using different metrics such as R2, RMSE, MAE, MedAE, etc. Among all ML models, the GB+LGBM model showed the best performance with R2 = 0.9445, RMSE = 5.99 MPa, MAE = 3.62 MPa, and MedAE = 2.26 MPa on the test set. SHAP analysis revealed that coarse aggregates (SHAP ≈+8) and cement (SHAP ≈+7) were the most influential factors, while the remaining water content and the excessive dosage of fiber were disadvantageous. Estimated compressive strength varied from 20 to 140 MPa. This study shows a novel approach by demonstrating the ability of ensemble ML models to capture complex concrete behavior, providing a data-driven approach for sustainable manufacturing of concrete. However, full reliance on the literature-based dataset still has significant limitations, which will increase noise, and these limitations can be further overcome by experimental validation in future studies.
Accurate prediction of the mechanical strength of Basalt Fiber Reinforced Concrete (BFRC) is critical for structural design, safety assessment, and the advancement of sustainable infrastructure in civil engineering. Traditional prediction methods often fail to capture the nonlinear relationships between BFRC mix proportions and resulting strength characteristics, leading to unreliable estimations. To address this limitation, this study proposes the Optimized Moment Balanced Machine (OMBM), an advanced machine learning model developed to improve the predictive accuracy of BFRC strength parameters. The model was trained and evaluated using key input features, including cement content, silica fume, fly ash, superplasticizer, water, aggregate composition, and fiber property parameters. The performance of the OMBM was benchmarked against four established machine learning models, such as Least Squares Support Vector Machine (LSSVM), Backpropagation Neural Network (BPNN), K-Nearest Neighbors (KNN), and Linear Regression (LR). Results from ten-fold cross-validation show that OMBM consistently outperforms the comparison models across five evaluation metrics. It achieved the lowest RMSE (2.411), MAE (1.788), and MAPE (4.08%), along with the highest values for correlation coefficient (R = 0.978), and coefficient of determination (R2 = 0.956). Furthermore, the OMBM achieved a Reference Index (RI) score of 1.000, which confirms its position as the leading predictive model within this comparative framework. These results confirm the robustness and reliability of the proposed OMBM model, making it a highly effective tool for accurate strength prediction of BFRC. This approach offers significant potential for the advancement of sustainable infrastructure by enabling more accurate and efficient use of concrete materials.
R. R. Khasani, Ferry Hermawan, Yuliana Usman· IOP Conference Series: Earth...· 0 citations
The combination of these two approaches demonstrates that flexural strength of SFRC may be forecasted using data-driven models with high reliability and minimises the reliance on the large-scale laboratory testing which makes the assessment process more effective.
Alkali-activated recycled aggregate concrete (AARAC) offers a sustainable alternative to traditional concrete but suffers from complex, non-linear mechanical behavior that challenges conventional prediction methods. This study develops and compares five machine learning models, linear regression (LR), M5P, Random Forest (RF), K-Nearest Neighbors (KNN) and XGBoost, for predicting the compressive strength (Cs), flexural strength (Fs), splitting tensile strength (Ss), pull-out bond strength (PT), and water absorption (Wa%) of AARAC. A dataset of 360 experimental samples, incorporating natural aggregate, recycled concrete aggregate (RCA), cement block aggregate (CBA), water-to-cement ratio (W/C), alkaline treatment status, and slump, was used. Models were evaluated via train/test split (80/20) and 10-fold cross-validation using R2, MAE, RMSE, and MAPE. Random Forest achieved the highest test R2 (0.8736) and lowest test MAPE (1.418%) and XGBoost (R2 = 0.8605, MAPE = 1.557%). KNN and M5P performed moderately, while LR was the weakest (R2 = 0.6958, MAPE = 2.147%). All tree-based models exhibited overfitting, with training R2 up to 0.98. Scatter plot analysis revealed systematic underprediction by RF for Cs (constant offset of ~2 MPa) and increasing bias for PT, Ss, and Wa% at higher values. XGBoost gave perfect predictions for PT and Wa% but underpredicted Cs and Fs. K-fold cross-validation confirmed XGBoost as the most robust (mean R2 = 0.9844). Correlation analysis showed W/C strongly increases Wa% (r = 0.80) and decreases PT (r = −0.73); RCA negatively affects mechanical properties, while CBA and alkaline treatment improve them. The study concludes that ensemble tree models, particularly Random Forest, are superior for AARAC prediction, but systematic bias requires post hoc calibration.
Ahmed D. Almutairi, Abd Al-Kader A. Al Sayed· Buildings· 0 citations
The rapid rise in global population and industrial activity has intensified environmental challenges, particularly carbon dioxide (CO₂) emissions from the cement and concrete industry. Biochar, a carbon-rich byproduct of biomass pyrolysis, has emerged as a promising solution for sustainable construction by enhancing carbon sequestration and improving mechanical performance when partially substituting cement. This study integrates experimental evidence with advanced machine learning (ML) techniques to evaluate the compressive strength, cost-efficiency, and carbon footprint of biochar-incorporated concrete. A comprehensive dataset of nine input parameters, including cement, aggregates, silica fume, fly ash, biochar, water, superplasticizer, and curing age was modeled using multiple ML approaches. Among the models tested, the hybrid XGB-Histogram Gradient Boosting (XGB-HistGB) model consistently achieved the best overall performance, with a testing R2 of 0.958, the lowest mean absolute error (3.03), and minimal prediction bias. This model outperformed standalone algorithms and other hybrids, providing reliable accuracy across compressive strength, cost, and embodied CO₂ predictions. SHAP and partial dependence analyses confirmed fine aggregate, curing age, and superplasticizer as the most influential parameters, while biochar dosage required careful optimization to balance strength retention with sustainability benefits. A user-friendly graphical interface was also developed, enabling real-time prediction of compressive strength, material cost, and CO₂ emissions based on user-defined mix proportions. Overall, the findings demonstrate that biochar can be effectively integrated into sustainable concrete formulations, and the XGB-HistGB model offers a powerful AI-driven predictive framework to optimize both structural performance and environmental outcomes.
M. Uddin, Md. Samsuzzaman Sobuz, Mohamed Ghalla et al.· Scientific Reports· 0 citations
This study proposes an experiment-driven machine learning (ML) framework to address experimental limitations in estimating compressive strength within a controlled experimental domain of locally sourced fiber-reinforced concrete (FRC). Concrete mixes incorporating 1, 2, and 3% lathe scrap steel fibers were prepared using a standard 1:2:4 mix with brick chips as coarse aggregate. Due to practical constraints, intermediate fiber contents (e.g., 1.5% and 2.5%) were not experimentally cast. To overcome this limitation, supervised ML models were employed to interpolate compressive strength at fractional fiber dosages, thereby reducing experimental cost and time. A total of 48 experimental observations obtained from laboratory-tested cylindrical specimens were used to develop and evaluate four ML algorithms—Gaussian Process Regression (GPR), Artificial Neural Network (ANN), Support Vector Machine (SVM) and Random Forest (RF)—under a five-fold cross-validation framework. Among these, GPR exhibited the best performance among the evaluated models (R² = 0.884, RMSE = 1.647 MPa), effectively capturing the nonlinear relationship between curing age, fiber content, and compressive strength. SHAP-based interpretability analysis indicated that curing age is the primary governing factor in strength development, while fiber content contributes through secondary nonlinear effects. The results further indicate that moderate fiber inclusion improves compressive strength, whereas excessive fiber content may reduce performance due to workability loss, fiber clustering, and increased matrix heterogeneity. These findings are consistent with established concrete mechanics, where fiber reinforcement plays a more dominant role in tensile behavior than in compressive strength. The proposed framework demonstrates the capability of probabilistic ML models for accurate interpolation within controlled experimental conditions using high-quality laboratory data. This study provides a practical, data-driven approach for strength estimation and mix optimization of low-cost, locally sourced FRC, supporting sustainable construction practices and efficient utilization of industrial waste materials.
Global demand for sustainable construction materials and concerns about environmental pollution from by-products of manufacturing industries have intensified research for viable alternatives to aggregates in concrete production. This research investigated steel slag aggregate (ssa) as a partial replacement for coarse aggregates. Samples of 150 mm concrete cubes and 100 x 100 x 500 mm prisms were prepared at a 1:2:4 mix ratio and a water-cement ratio of 0.5, with ssa replacing coarse aggregates at 0%, 15%, 30%, 45%, and 60% by weight for 7, 14, and 28 days compressive and flexural strength tests. In addition, a validated random forest (rf) and multiple linear regression (mlr) algorithm were developed to predict the compressive strength of samples. The analysis of ssa for x-ray fluorescence (xrf) revealed a high 34.125% silicon dioxide (sio₂) composition and a minimum of 0.196% for strontium oxide (sro), a specific gravity of 3.07, 1680 kg/m³ bulk density, 21.28% and 8.37% aggregate crushing value and impact value were evaluated, respectively. The 15%, 30%, 45%, and 60% ssa replacement samples exhibited improved strengths of 15.63 n/mm², 17.07 n/mm², 18.30 n/mm², and 19.10 n/mm², while the flexural strengths increased up to 45% ssa (4.45 n/mm²) before declining at 60% (3.85 n/mm²). The mean absolute error (mae), mean square error (mse) and a coefficient of determination (r²) for rf were 1.20, 2.07 and 0.85, while mlr recorded 1.46, 3.77 and 0.72, respectively. The xrf suggests an improved aggregate bonding potential, and the physical characterisation revealed that ssa was within the acceptable limits for structural applications. The compressive and flexural strengths increased with ssa content up to 45%, after which strength properties declined, offering optimal mechanical performance. The mlr achieved a robust prediction accuracy with an r² value of 0.85. In conclusion, the research supports the potential of industrial by-products in promoting greener and more cost-effective construction practices
Akintayo Adeniji, W. Kupolati, Everardt A. Burger et al.· Kufa journal of Engineering· 0 citations