Study-Disjoint Machine Learning for Concrete Compressive Strength

Author

Muhammad Safi Ullah, Laiba Qamar

Abstract

 This study developed and evaluated machine-learning models for predicting 28-day cylinder compressive strength of concrete using a literature-derived dataset of 1,456 mixtures collected from 136 source studies. Eight recorded mixture quantities were used as the original inputs. Source-study identity and cross-study duplicate relationships were incorporated into the validation design to prevent linked observations from appearing on both sides of a train-test split. Seven prediction approaches were assessed using repeated nested grouped cross-validation, with model selection and hyperparameter tuning performed only within the training partitions. ExtraTrees gave the lowest mean outer-fold error among the evaluated model families, with an MAE of 9.758 MPa, RMSE of 12.884 MPa, and study-macro-MAE of 10.392 MPa. Conventional random validation produced substantially more optimistic performance than study-disjoint validation; for ExtraTrees, R² decreased from 0.844 to 0.471 and MAE increased from 4.783 to 9.807 MPa. Applicability-domain assessment showed higher errors for observations classified outside the represented training domain. Permutation importance identified the water-to-binder ratio as the strongest model-specific predictor. Group-aware prediction intervals achieved approximately 91% empirical coverage for a nominal 90% level. The results show that study-disjoint validation provides a more rigorous estimate of model transferability across literature-derived concrete datasets than conventional random splitting.

Keywords

Concrete compressive strength; Machine learning; Supplementary cementitious materials; ExtraTrees; Grouped cross-validation; Applicability domain; Permutation importance; Prediction uncertainty

Full Text:

Download Paper PDF

References

[1] W. Ben Chaabene, M. Flah, and M. L. Nehdi, “Machine learning prediction of mechanical properties of concrete: Critical review,” Construction and Building Materials, vol. 260, Art. no. 119889, 2020.

[2] I. Núñez, A. Marani, M. Flah, and M. L. Nehdi, “Estimating compressive strength of modern concrete mixtures using computational intelligence: A systematic review,” Construction and Building Materials, vol. 310, Art. no. 125279, 2021.

[3] Z. Li, J. Yoon, R. Zhang, F. Rajabipour, W. V. Srubar III, I. Dabo, and A. Radlińska, “Machine learning in concrete science: Applications, challenges, and best practices,” npj Computational Materials, vol. 8, Art. no. 127, 2022.

[4] W. Ahmad, A. Ahmad, K. A. Ostrowski, F. Aslam, P. Joyklad, and P. Zajdel, “Application of advanced machine learning approaches to predict the compressive strength of concrete containing supplementary cementitious materials,” Materials, vol. 14, no. 19, Art. no. 5762, 2021.

[5] F. Aslam and M. Z. Shahab, “Supplementary cementitious materials in blended cement concrete: Advancements in predicting compressive strength through machine learning,” Materials Today Communications, vol. 38, Art. no. 107725, 2024.

[6] A. A. B. Loureiro and R. Stefani, “Comparing the performance of machine learning models for predicting the compressive strength of concrete,” Discover Civil Engineering, vol. 1, Art. no. 19, 2024.

[7] M. K. Elshaarawy, M. M. Alsaadawi, and A. K. Hamed, “Machine learning and interactive GUI for concrete compressive strength prediction,” Scientific Reports, vol. 14, Art. no. 16694, 2024.

[8] X. Liu, S. Mei, X. Wang, and X. Li, “Estimation of compressive strength of concrete with manufactured sand and natural sand using interpretable artificial intelligence,” Case Studies in Construction Materials, vol. 21, Art. no. e03840, 2024.

[9] H. Shimakawa, A. Kumada, and M. Sato, “Prevention of leakage in machine learning prediction for polymer composite properties,” Journal of Chemical Information and Modeling, vol. 64, no. 9, pp. 3621–3629, 2024.

[10] S. Durdy, M. W. Gaultois, V. V. Gusev, D. Bollegala, and M. J. Rosseinsky, “Random projections and kernelised leave one cluster out cross validation: Universal baselines and evaluation tools for supervised machine learning of material properties,” Digital Discovery, vol. 1, no. 6, pp. 763–778, 2022.

[11] W. M. Mesfin and H. K. Kim, “Parametric study on global estimation models for compressive strength adopting various machine learning algorithms in concrete,” Engineering Applications of Artificial Intelligence, vol. 136, Art. no. 108888, 2024.

[12] H. Nguyen, T. Vu, T. P. Vo, and H.-T. Thai, “Efficient machine learning models for prediction of concrete strengths,” Construction and Building Materials, vol. 266, Art. no. 120950, 2021. doi: 10.1016/j.conbuildmat.2020.120950.

[13] T. Nguyen-Sy, J. Wakim, Q.-D. To, M.-N. Vu, T.-D. Nguyen, and T.-T. Nguyen, “Predicting the compressive strength of concrete from its compositions and age using the extreme gradient boosting method,” Construction and Building Materials, vol. 260, Art. no. 119757, 2020. doi: 10.1016/j.conbuildmat.2020.119757.

[14] M. R. Kaloop et al., “Compressive strength prediction of high-performance concrete using gradient tree boosting machine,” Construction and Building Materials, vol. 264, Art. no. 120198, 2020. doi: 10.1016/j.conbuildmat.2020.120198.

[15] P. G. Asteris, A. D. Skentou, A. Bardhan, P. Samui, and K. Pilakoutas, “Predicting concrete compressive strength using hybrid ensembling of surrogate machine learning models,” Cement and Concrete Research, vol. 145, Art. no. 106449, 2021. doi: 10.1016/j.cemconres.2021.106449.

[16] Y. Xu, W. Ahmad, A. Ahmad, K. A. Ostrowski, M. Dudek, F. Aslam, and P. Joyklad, “Computation of high-performance concrete compressive strength using standalone and ensembled machine learning techniques,” Materials, vol. 14, no. 22, Art. no. 7034, 2021. doi: 10.3390/ma14227034.

[17] H. S. Ullah, R. A. Khushnood, F. Farooq, J. Ahmad, N. I. Vatin, and D. Y. Z. Ewais, “Prediction of compressive strength of sustainable foam concrete using individual and ensemble machine learning approaches,” Materials, vol. 15, no. 9, Art. no. 3166, 2022. doi: 10.3390/ma15093166.

[18] P. Huang, K. Dai, and X. Yu, “Machine learning approach for investigating compressive strength of self-compacting concrete containing supplementary cementitious materials and recycled aggregate,” Journal of Building Engineering, vol. 79, Art. no. 107904, 2023. doi: 10.1016/j.jobe.2023.107904.

[19] B. Sun, W. Cui, G. Liu, B. Zhou, and W. Zhao, “A hybrid strategy of AutoML and SHAP for automated and explainable concrete strength prediction,” Case Studies in Construction Materials, vol. 19, Art. no. e02405, 2023. doi: 10.1016/j.cscm.2023.e02405.

[20] G. Liu and B. Sun, “Concrete compressive strength prediction using an explainable boosting machine model,” Case Studies in Construction Materials, vol. 18, Art. no. e01845, 2023. doi: 10.1016/j.cscm.2023.e01845.

[21] W. Wang, Y. Zhong, G. Liao, Q. Ding, T. Zhang, and X. Li, “Prediction of compressive strength of concrete specimens based on interpretable machine learning,” Materials, vol. 17, no. 15, Art. no. 3661, 2024. doi: 10.3390/ma17153661.

[22] Y. Zhang, W. Ren, Y. Chen, Y. Mi, J. Lei, and L. Sun, “Predicting the compressive strength of high-performance concrete using an interpretable machine learning model,” Scientific Reports, vol. 14, Art. no. 28346, 2024. doi: 10.1038/s41598-024-79502-z.

[23] W. Zhang, J. Guo, C. Ning, R. Cheng, and Z. Liu, “Prediction of concrete compressive strength using a Deepforest-based model,” Scientific Reports, vol. 14, Art. no. 18918, 2024. doi: 10.1038/s41598-024-69616-9.

[24] S. Kumar, R. Kumar, B. Rai, and P. Samui, “Prediction of compressive strength of high-volume fly ash self-compacting concrete with silica fume using machine learning techniques,” Construction and Building Materials, vol. 438, Art. no. 136933, 2024. doi: 10.1016/j.conbuildmat.2024.136933.

[25] M. A. Hariri-Ardebili, P. Mahdavi, and F. Pourkamali-Anaraki, “Benchmarking AutoML solutions for concrete strength prediction: Reliability, uncertainty, and dilemma,” Construction and Building Materials, vol. 423, Art. no. 135782, 2024. doi: 10.1016/j.conbuildmat.2024.135782.

[26] P. Tamuly and V. Nava, “Machine learning based conformal predictors for uncertainty-aware compressive strength estimation of concrete,” Construction and Building Materials, vol. 487, Art. no. 141844, 2025. doi: 10.1016/j.conbuildmat.2025.141844.

[27] A. Thoriya and H. Rangwala, “Calibrated probabilistic prediction of compressive strength in palm oil fuel ash-modified concrete using conformal learning,” Scientific Reports, 2026. doi: 10.1038/s41598-026-60994-w.

[28] A. Dunn, Q. Wang, A. Ganose, D. Dopp, and A. Jain, “Benchmarking materials property prediction methods: The Matbench test set and Automatminer reference algorithm,” npj Computational Materials, vol. 6, Art. no. 138, 2020. doi: 10.1038/s41524-020-00406-3.

[29] H. Kaneko, “Evaluation and optimization methods for applicability domain methods and their hyperparameters, considering the prediction performance of machine learning models,” ACS Omega, vol. 9, no. 10, pp. 11453–11458, 2024. doi: 10.1021/acsomega.3c08036.

[30] K. Boonpalit, J. Kinchagawat, and S. Namuangruk, “Expanding the applicability domain of machine learning model for advancements in electrochemical material discovery,” ChemElectroChem, vol. 11, no. 10, Art. no. e202300681, 2024. doi: 10.1002/celc.202300681.

[31] L. E. Schultz, Y. Wang, R. Jacobs, and D. Morgan, “A general approach for determining applicability domain of machine learning models,” npj Computational Materials, vol. 11, Art. no. 95, 2025. doi: 10.1038/s41524-025-01573-x.

Share your valuable work from Social Media Buttons