Predicting compressive strength of recycled aggregate concrete using AI with experimental validation

Abstract

Growing concerns about environmental impact and rising construction costs have encouraged the use of recycled aggregates in concrete. This study offers an experimental and AI-driven assessment of the compressive strength of recycled aggregate concrete. Concrete was prepared by varying recycled fine and coarse aggregates, water-cement ratios, plasticizer content, and parent concrete strength. The measured 28-day compressive strength ranged from 31.8 to 45.2 MPa, and a steady reduction was observed as recycled aggregate levels and water absorption increased. The results highlight how aggregate quality, abrasion resistance, and replacement percentage influence mechanical performance. To extend the laboratory findings, nine machine learning models were developed to estimate compressive strength based on all mix parameters. Support Vector Regression delivered the strongest performance with an R² of 0.998, followed by Random Forest (0.996). K-Nearest Neighbors and XGBoost achieved R² values of 0.947 and 0.942, while Gradient Boosting, Linear Regression, Lasso, Elastic Net, and Ridge obtained R² scores of 0.904, 0.893, 0.879, 0.846, and 0.842. Nonlinear algorithms clearly performed better than linear models. The results show that AI tools can reliably predict compressive strength, reducing the need for extensive testing and helping streamline mix design for recycled aggregate concrete.

Keywords

Recycled aggregate concrete / Compressive strength / Machine learning / Support Vector Regression / Random Forest / XGBoost / KNN / Gradient Boosting / Elastic Net / Durability / Sustainable construction

References

[1] Ajdukiewicz A, Kliszczewicz A. Influence of recycled aggregates on mechanical properties of high-strength concrete. Cement and Concrete Composites, 2002, 242269279.

[2] Akbar A, Liew JYR. Environmental impact of cement production and concrete sustainability. Journal of Cleaner Production, 2020, 258: 120543.

[3] Akbar A, Liew KM. Assessing recycling potential of carbon fiber reinforced plastic waste in eco-efficient cement materials. Journal of Cleaner Production, 2020, 274: 123001.

[4] Altman NS. An introduction to kernel and nearest-neighbor nonparametric regression. The American Statistician, 1992, 463175185.

[5] Bai X. Experimental study on mechanical properties and compressive constitutive model of recycled concrete under sulfate attack considering the effects of multiple factors. Buildings, 2023, 1492761.

[6] Bhanja S, Sengupta B. Prediction of compressive strength of concrete using regression analysis. Cement and Concrete Research, 2005, 35918481855.

[7] Bilim C, Atis CD, Tanyildizi H, Karahan O. Predicting compressive strength of GGBS concrete using ANN. Advances in Engineering Software, 2009, 405334340.

[8] Bilim C, Koksal F, Tosun T. Prediction of concrete strength using ANN. Expert Systems with Applications, 2009, 36237363742.

[9] Bilim C, Ozbakkaloglu T, Tanyu BF. Artificial neural networks for predicting compressive strength of concrete. ACI Materials Journal, 2009, 1063203211

[10] Breiman L. Random forests. Machine Learning, 2001, 451532.

[11] Cakiroglu M, Pal S, Sharma R. Environmental challenges of construction waste: A review. Journal of Environmental Management, 2023, 335: 117142.

[12] Chakradhara Rao M, Bhattacharyya SK, Barai SV. Influence of field recycled coarse aggregate on properties of concrete. Materials and Structures, 2011, 441205220.

[13] Chen, T., & Guestrin, C. (2016). XGBoost: A scalable tree boosting system. KDD Proceedings, 785–794. https://doi.org/10.1145/2939672.2939785

[14] Chicco D, Warrens MJ. The coefficient of determination R-squared is more informative than SMAPE, MAE, MSE and RMSE in regression analysis evaluation. PeerJ Computer Science, 2021, 7: e623.

[15] Dantas AT, Leite M, Nagahama K. ANN-based strength prediction of concrete with waste. Construction and Building Materials, 2013, 38: 717-722.

[16] Deng A, Naderpour H, Ghanizadeh A. Predicting strength of recycled concrete using ML. Materials Today: Proceedings, 2018, 5112409124099.

[17] Ebid AM. Paper introducing controlled random perturbation/data augmentation for limited engineering datasets. Journal of Engineering Sciences, Assiut University, 2025.

[18] Emmert-Streib F, Dehmer M. Evaluation of regression models: Model assessment, model selection and generalization error. Machine Learning and Knowledge Extraction, 2019, 11521551.

[19] Farhangi V, Jahangir H, Ghafari E. ML-based prediction of recycled aggregate concrete. Materials Today Communications, 2021, 28: 102633.

[20] Field A. Discovering statistics using IBM SPSS, 20134Sage

[21] Friedman J. Greedy function approximation: Gradient boosting. Annals of Statistics, 2001, 29511891232.

[22] Geurts P, Ernst D, Wehenkel L. Extremely randomized trees. Machine Learning, 2006, 63: 3-42.

[23] Ghanizadeh A, Naderpour H, Deng Q. AI-based strength prediction for recycled concrete. Journal of Cleaner Production, 2019, 215: 376-385

[24] Goodfellow I, Bengio Y, Courville A. Deep learning, 2016MIT Press

[25] Hansen TC, Narud H. Strength of recycled concrete with crushed aggregates. Concrete International, 1983, 517983

[26] Hastie T, Tibshirani R, Friedman J. Elements of statistical learning, 2009Springer

[27] Hoerl AE, Kennard RW. Ridge regression: Biased estimation for nonorthogonal problems. Technometrics, 1970, 1215567.

[28] Huang B, Yan Q. Evaluating the mechanical properties of recycled aggregate concrete with variable coarse and fine aggregate replacements. Discover Civil Engineering, 2025.

[29] International, A. (2024). ASTM C150/C150M-24: Standard specification for Portland cement

[30] Jahangir H, Eidgahee D. Hybrid AI modeling for FRP–concrete bond strength. Composite Structures, 2020, 257: 113160.

[31] Jahangir H, Farhangi H, Ghodrati B. Ensemble prediction of RAC strength. Construction and Building Materials, 2021, 271: 121654

[32] Katz A. Concrete made from partially hydrated recycled aggregate. Cement and Concrete Research, 2003, 335703711.

[33] Khalil E, AbouZeid M. Framework for cement plants assessment through cement production improvement measures for reduction of CO₂ emissions towards net zero emissions. Construction Materials, 2025, 5220.

[34] Khan K, Ahmad W, Amin MN, Aslam F, Ahmad A, Al-Faiad MA. Comparison of machine learning methods for recycled aggregate concrete strength prediction. Materials, 2022, 15103430.

[35] Khatib JM. Concrete with fine recycled aggregates. Cement and Concrete Research, 2005, 354763769.

[36] Lundberg SM, Lee SI. A unified approach to interpreting model predictions. Advances in Neural Information Processing Systems, 2017, 30: 4765-4774

[37] Miller C. A review of model evaluation metrics for machine learning in regression, classification, and clustering. PeerJ Computer Science, 2024, 10: eXXX

[38] Molnar C. Interpretable machine learning: A guide for making black box models explainable, 2020Lulu.com

[39] Müller AC, Guido S. Schanafelt D. Introduction to machine learning with python (or the corresponding chapter in the cited Springer volume). Machine learning, 2024Springer

[40] Naderpour H, Ghanizadeh M, Behnood A. RAC compressive strength prediction via ANN. Journal of Cleaner Production, 2018, 184: 640-650

[41] Onyelowe KC, Gnananandarao T, Ebid AM, Mahdi HA, Razzaghian Ghadikolaee M, Al-Ajamee M. Evaluating the compressive strength of recycled aggregate concrete using novel artificial neural network. Civil Engineering Journal, 2022, 8816791693.

[42] Onyelowe KC, Gnananandarao T, Jagan J, Ahmad J, Ebid AM. Innovative predictive model for flexural strength of recycled aggregate concrete from multiple datasets. Asian Journal of Civil Engineering, 2023, 24: 1-15.

[43] Onyelowe KC, Kamchoom V, Hanandeh S, Anandha Kumar S, ZabalaVizuete RF, Santillán Murillo RO, Zurita Polo SM, Torres Castillo RM, Ebid AM, Awoyera P, Arunachalam KP. Physics-informed modeling of splitting tensile strength of recycled aggregate concrete using advanced machine learning. Scientific Reports, 2025, 15. 7135

[44] Padmini AK, Ramamurthy K, Mathew M. Influence of parent concrete on RAC properties. Cement and Concrete Research, 2002, 321218931901

[45] Pal R, Singh S. Role of impact and compression-based crushing on the physical, chemical, and morphological characteristics of recycled concrete aggregates. Journal of Materials in Civil Engineering, 2024.

[46] Plevris, V., Solórzano, G., Bakas, N. P., & Ben Seghier, M. E. (2022). Investigation of performance metrics in regression analysis and machine learning based prediction models. Proceedings of the 8th European Congress on Computational Methods in Applied Sciences and Engineering (ECCOMAS 2022). https://doi.org/10.23967/eccomas.2022.155

[47] Ryu GS. Strength and durability of recycled aggregate concrete. Construction and Building Materials, 2002, 16195101

[48] Shin YS, Kim GH. Predicting compressive strength of recycled aggregate concrete by multiple regression analysis. Applied Mechanics and Materials, 2013, 253: 546-549.

[49] Smola A, Scholkopf B. Tutorial on support vector regression. Statistics and Computing, 2004, 143199222.

[50] Steurer M, Schulz W, Werwatz A. Metrics for evaluating the performance of machine learning prediction models. International Journal of Forecasting, 2021, 366812828.

[51] Tu C, Belarbi R, Pigeon M, Tam V. Properties of concrete with recycled aggregate. Cement and Concrete Research, 2006, 361153159

[52] Wolpert D. Stacked generalization. Neural Networks, 1992, 52241259.

[53] Yeh I-C. ML prediction of concrete properties. Cement and Concrete Research, 2007, 375676685

[54] Zain MFM, Safiuddin M, Mahmud HB. Development of high-performance concrete using silica fume and superplasticizer. Cement and Concrete Research, 2002, 321115122.

[55] ZárateVillacres ANJ, Ebid AM, Ulloa N, Onyelowe KC, Zuñiga Rodriguez MG, Andrade Valle AI. Multiple AI predictive models for compressive strength of recycled aggregate concrete. Cogent Engineering, 2024, 1112385621

[56] Zhang J, Li X, Chen Y. Stacking ensemble for concrete strength prediction. Engineering Applications of Artificial Intelligence, 2022, 109: 104617

[57] Zhang P, Sun X, Wang F, Wang J. Mechanical properties and durability of geopolymer recycled aggregate concrete: A review. Polymers, 2023, 153615.

[58] Zhang X, Sun X, Wang F, Wang J. Predicting compressive strength of recycled aggregate concrete using machine learning and hyperparameter optimization techniques. Frontiers in Earth Science, 2023, 11: 1112105.

[59] Zou H, Hastie T. Regularization and variable selection via the elastic net. Journal of the Royal Statistical Society. Series b, Methodological, 2005, 672301320.

Rights & permissions

The Author(s)