Comparative Evaluation of Traditional and Deep Learning Based Machine Learning Models for Concrete Compressive Strength Prediction

Authors

  • Atif Khan

    Department of Civil and Environmental Engineering, University of Louisiana at Lafayette, Lafayette, LA 70504, USA
  • Afsar Ali

    Department of Civil and Environmental Engineering, University of Louisiana at Lafayette, Lafayette, LA 70504, USA
  • Syed Saqib Mehboob

    Department of Civil Engineering, University of Engineering and Technology, Taxila 47050, Pakistan

DOI:

https://doi.org/10.30564/jaeser.v9i1.13199
Received: 6 November 2025 | Revised: 25 December 2025 | Accepted: 2 January 2026| Published Online: 8 January 2026

Abstract

Machine learning is widely used to predict concrete compressive strength because it captures nonlinear interactions among binders, aggregates, water content, and chemical admixtures. Tree-based ensemble models such as Random Forest and XGBoost often achieve high numerical accuracy; however, their discrete decision splitting mechanisms inherently produce stepwise response trends that may not reflect the smooth and continuous nature of material behavior. To overcome this limitation, this study proposes a hybrid CNN1D–BiLSTM model that integrates one dimensional convolutional neural networks for localized feature extraction with bidirectional long short term memory units to learn long range compositional dependencies and gradual transitions in mixture design. Linear Regression, Random Forest, XGBoost, and the proposed model were evaluated using 1,030 concrete mixtures from the Yeh dataset. Model performance was assessed using the coefficient of determination (R2), Root Mean Square Error (RMSE), Mean Absolute Error (MAE), Mean Absolute Percentage Error (MAPE), Nash Sutcliffe Efficiency (NSE), Mean Bias Error (MBE), and Percent Deviation Degree (PDD) to provide a comprehensive evaluation of predictive accuracy and systematic bias. The CNN1D–BiLSTM achieved the best testing performance with R2 and NSE of 0.9324 and RMSE of 4.17 MPa, while exhibiting minimal systematic bias as indicated by low MBE and PDD values. In addition to strong numerical performance, the proposed model generated smooth partial dependence trends consistent with hydration kinetics and continuous strength development. Shapley Additive Explanations (SHAP) analysis confirmed stable and physically meaningful contributions of key mixture components, demonstrating that the hybrid framework maintains high predictive capability while preserving realistic response behavior across varying mixture compositions.

Keywords:

Machine Learning; Concrete Compressive Strength; XGBoost; Sensitivity Analysis; Sustainable Construction; Data-Driven Prediction

References

[1] Moein, M.M., Saradar, A., Rahmati, K., et al., 2023. Predictive Models for Concrete Properties Using Machine Learning and Deep Learning Approaches: A Review. Journal of Building Engineering. 63, 105444.

[2] Ali, A., Khan, Q.U.Z., Mehboob, S.S., et al., 2024. Enhancing Multi-Objective Mix Design for GGBS-Based Geopolymer Concrete with Natural Mineral Blends under Ambient Curing: A Taguchi-Grey Relational Optimization. Ain Shams Engineering Journal. 15(5), 102708.

[3] Ruth Bola Oliveira, D.; Leite, G.; Possan, E.; et al., 2023. Concrete powder waste as a substitution for Portland cement for environment-friendly cement production. Construction and Building Materials. 397, 132382. DOI: https://doi.org/10.1016/j.conbuildmat.2023.132382

[4] Aziz, A., Mehboob, S.S., Tayyab, A., et al., 2024. Enhancing Sustainability in Self-Compacting Concrete by Optimizing Blended Supplementary Cementitious Materials. Scientific Reports. 14, 12326.

[5] Ali, A., Khan, Q.U.Z., Mehboob, S.S., 2024. Experimental Design Optimization Using Taguchi-Analysis for Compositional Parameters of Eco-Friendly GGBS Based Geopolymer Concrete. In Proceedings of the 2nd International Conference on Recent Advances in Civil Engineering and Disaster Management, Peshawar, Pakistan, 15–16 December 2022.

[6] Jain, S., Jain, R., Rao, K.R., et al., 2024. Leveraging Machine Learning to Minimize Experimental Trials and Predict Hot Deformation Behaviour in Dual Phase High Entropy Alloys. Materials Today Communications. 41, 110813.

[7] Khan, A., Khattak, M.J., Pesacreta, T.C., 2025. Development of Geopolymer-Based Green Construction Materials for Highway Infrastructure Using High-Content Sugarcane Bagasse Ash. Transportation Research Record. 2680(1), 121–137.

[8] Khan, A., Khattak, M.J., Pesacreta, T.C., et al., 2025. Microstructural, Mechanical, and Durability Assessment of Sustainable Geopolymers Synthesized Using Rice Husk Ash, a Byproduct of the Rice Industry. Journal of Sustainable Cement-Based Materials. 14, 2519–2536.

[9] Khan, A., 2025. Development of Green Construction Materials Using High-Content Sugarcane Bagasse Ash [Master’s Thesis]. University of Louisiana at Lafayette: Lafayette, LA, USA.

[10] Haq, I.U., Elahi, A., Khan, A., et al., 2025. The Addition of Natural Clay and Industrial and Agricultural Waste on the Performance of Green and Sustainable Concrete. Scientific Reports. 15, 17034.

[11] Sun, B., Huang, Y., Liu, G., et al., 2024. Prediction of Compressive Strength of Concrete under Various Curing Conditions: A Comparison of Machine Learning Models and Empirical Mathematical Models. Innovative Infrastructure Solutions. 9, 262.

[12] Yeh, I.-C., 1998. Concrete Compressive Strength. Available from: https://archive.ics.uci.edu/dataset/165/concrete+compressive+strength (cited 3 May 2025).

[13] Ilyas, I., Zafar, A., Afzal, M.T., et al., 2022. Advanced Machine Learning Modeling Approach for Prediction of Compressive Strength of FRP Confined Concrete Using Multiphysics Genetic Expression Programming. Polymers. 14, 1789.

[14] Sun, J., Dang, W., Wang, F., et al., 2023. Prediction of TOC Content in Organic-Rich Shale Using Machine Learning Algorithms: Comparative Study of Random Forest, Support Vector Machine, and XGBoost. Energies. 16, 4159.

[15] Ozcan, G., Gulbandilar, E., Kocak, Y., 2025. Prediction of Compressive Strengths of Portland Cement with Random Forest, Support Vector Machine and Gradient Boosting Models. Neural Computing and Applications. 37, 23495–23511.

[16] Ali, T., Onyelowe, K.C., Mahmood, M.S., et al., 2025. Advanced and Hybrid Machine Learning Techniques for Predicting Compressive Strength in Palm Oil Fuel Ash-Modified Concrete with SHAP Analysis. Scientific Reports. 15, 4997.

[17] Alsharari, F., 2025. Predicting the Compressive Strength of Fiber-Reinforced Recycled Aggregate Concrete: A Machine-Learning Modeling with SHAP Analysis. Asian Journal of Civil Engineering. 26, 179–205.

[18] Ghrici, A.A., Benzaamia, A., Mezzoudj, F., et al., 2025. SHAP-Enhanced Tree-Based Regression for Predicting the Compressive Strength of High Performance Concrete. Construction and Building Materials. 495, 143602.

[19] Berodier, E.M.J., 2015. Impact of the Supplementary Cementitious Materials on the Kinetics and Microstructural Development of Cement Hydration [PhD Thesis]. École Polytechnique Fédérale de Lausanne: Lausanne, Switzerland.

[20] Moradi, N., Tavana, M.H., Habibi, M.R., et al., 2022. Predicting the Compressive Strength of Concrete Containing Binary Supplementary Cementitious Material Using Machine Learning Approach. Materials. 15, 5336.

[21] Alzubaidi, L., Zhang, J., Humaidi, A.J., et al., 2021. Review of Deep Learning: Concepts, CNN Architectures, Challenges, Applications, Future Directions. Journal of Big Data. 8, 53.

[22] Liu, Y., Yu, H., Guan, T., et al., 2025. Intelligent Prediction of Compressive Strength of Concrete Based on CNN-BiLSTM-MA. Case Studies in Construction Materials. 22, e04486.

[23] Ziolkowski, P., 2025. Influence of Optimization Algorithms and Computational Complexity on Concrete Compressive Strength Prediction Machine Learning Models for Concrete Mix Design. Materials. 18, 1386.

[24] Bernal, S.A., de Gutiérrez, R.M., Pedraza, A.L., et al., 2011. Effect of Binder Content on the Performance of Alkali-Activated Slag Concretes. Cement and Concrete Research. 41, 1–8.

[25] Powers, T.C., 1947. A Discussion of Cement Hydration in Relation to the Curing of Concrete. Portland Cement Association: Chicago, IL, USA.

[26] Li, L.G., Kwan, A.K., 2015. Effects of Superplasticizer Type on Packing Density, Water Film Thickness and Flowability of Cementitious Paste. Construction and Building Materials. 86, 113–119.

[27] Yusslee, E., Beskhyroun, S., 2023. The Effect of Water-to-Binder Ratio (W/B) on Pore Structure of One-Part Alkali Activated Mortar. Heliyon. 9(1), e12983.

[28] Salman, H.A., Kalakech, A., Steiti, A., 2024. Random Forest Algorithm Overview. Babylonian Journal of Machine Learning. 2024, 69–79.

[29] Rezzoug, A., Khan, A., Murtaza, N., et al., 2025. Predicting Flood Energy Reduction in Vegetated Open Channel: Comparative Assessment of Hybrid Artificial Intelligence Techniques. Engineering Applications of Artificial Intelligence. 159, 111756.

[30] Özçift, A., 2011. Random Forests Ensemble Classifier Trained with Data Resampling Strategy to Improve Cardiac Arrhythmia Diagnosis. Computers in Biology and Medicine. 41(5), 265–271.

[31] Jaiswal, J.K., Samikannu, R., 2017. Application of Random Forest Algorithm on Feature Subset Selection and Classification and Regression. In Proceedings of the 2017 World Congress on Computing and Communication Technologies, Tiruchirappalli, India, 2–4 February 2017; pp. 65–68.

[32] Xia, S., Wang, G., Chen, Z., et al., 2018. Complete Random Forest Based Class Noise Filtering Learning for Improving the Generalizability of Classifiers. IEEE Transactions on Knowledge and Data Engineering. 31, 2063–2078.

[33] Al-Shamasneh, A.R., Kewalramani, M., Mahmoodzadeh, A., et al., 2025. Forecasting Compressive Strength of Concrete Containing Rice Husk Ash Using Various Machine Learning Algorithms. Scientific Reports. 15, 39162.

[34] Sahin, E.K., 2020. Assessing the Predictive Capability of Ensemble Tree Methods for Landslide Susceptibility Mapping Using XGBoost, Gradient Boosting Machine, and Random Forest. SN Applied Sciences. 2, 1308.

[35] Rouhani, M.M., Dolatshahi, A., 2025. Advanced Hybrid Models for Predicting Dynamic Compressive Strength of Rocks: Exploring Various Input Scenarios and Introducing Novel Empirical Approaches. Results in Engineering. 28, 107947.

[36] Nielsen, D., 2016. Tree Boosting with Xgboost—Why Does Xgboost Win Every Machine Learning Competition? Norwegian University of Science and Technology: Trondheim, Norway.

[37] Demir, S., Şahin, E.K., 2022. Liquefaction Prediction with Robust Machine Learning Algorithms (SVM, RF, and XGBoost) Supported by Genetic Algorithm-Based Feature Selection and Parameter Optimization from the Perspective of Data Processing. Environmental Earth Sciences. 81, 459.

[38] Rouhani, M.M., Dolatshahi, A., 2025. Predicting Mode I Fracture Toughness of CCNBD Quasi-Brittle Specimens Using Hybrid Gradient Boosting Algorithms. Theoretical and Applied Fracture Mechanics. 138, 104966.

[39] Khan, A., Sohail, A., Zahoora, U., et al., 2020. A Survey of the Recent Architectures of Deep Convolutional Neural Networks. Artificial Intelligence Review. 53, 5455–5516.

[40] Ding, B., Qian, H., Zhou, J., 2018. Activation Functions and Their Characteristics in Deep Neural Networks. In Proceedings of 2018 Chinese Control And Decision Conference (CCDC), Shenyang, China, 9–11 June 2018; pp. 1836–1841.

[41] Gholamalinezhad, H., Khosravi, H., 2020. Pooling Methods in Deep Neural Networks, a Review. arXiv preprint. arXiv:2009.07485.

[42] Li, C., Zhang, M., Zhang, X., 2023. Enhancing Concrete Creep Prediction with Deep Learning: A Soft-Sorted One-Dimensional CNN Approach. IEEE Access. 11, 139314–139325.

[43] Atef, S., Eltawil, A.B., 2020. Assessment of Stacked Unidirectional and Bidirectional Long Short-Term Memory Networks for Electricity Load Forecasting. Electric Power Systems Research. 187, 106489.

[44] Yu, Y., Si, X., Hu, C., et al., 2019. A Review of Recurrent Neural Networks: LSTM Cells and Network Architectures. Neural Computation. 31, 1235–1270.

[45] Chandra, N., Ahuja, L., Khatri, S.K., et al., 2021. Utilizing Gated Recurrent Units to Retain Long Term Dependencies with Recurrent Neural Network in Text Classification. Journal of Information Systems and Telecommunication (JIST). 2, 89.

[46] Liu, G., Guo, J., 2019. Bidirectional LSTM with Attention Mechanism and Convolutional Layer for Text Classification. Neurocomputing. 337, 325–338.

[47] Zhang, S., Liu, C., Jiang, H., et al., 2015. Feedforward Sequential Memory Networks: A New Structure to Learn Long-Term Dependency. arXiv preprint. arXiv:1512.08301.

[48] Lei, F., Liu, X., Li, Z., et al., 2021. Multihop Neighbor Information Fusion Graph Convolutional Network for Text Classification. Mathematical Problems in Engineering. 2021(1), 6665588.

[49] Piasta, W., Zarzycki, B., 2017. The Effect of Cement Paste Volume and w/c Ratio on Shrinkage Strain, Water Absorption and Compressive Strength of High Performance Concrete. Construction and Building Materials. 140, 395–402.

Downloads

How to Cite

Khan, A., Ali, A., & Saqib Mehboob , S. (2026). Comparative Evaluation of Traditional and Deep Learning Based Machine Learning Models for Concrete Compressive Strength Prediction. Journal of Architectural Environment & Structural Engineering Research, 9(1), 1–28. https://doi.org/10.30564/jaeser.v9i1.13199

Issue

Vol. 9 , Iss. 1 (January 2026): In Progress

Article Type

Article

License

Copyright © 2026 Atif Khan, Afsar Ali, Syed Saqib Mehboob

Creative Commons License

This is an open access article under the Creative Commons Attribution-NonCommercial 4.0 International (CC BY-NC 4.0) License.

Crossref
Scopus
Google Scholar
Europe PMC