Interpretable Machine Learning and PSO-Based Optimization for Predicting the Mechanical Performance of Steel Fiber-Reinforced Recycled Aggregate Concrete: A Dual Focus on Compressive and Splitting Tensile Strengths
DOI:
https://doi.org/10.54536/ajcec.v1i1.5407Keywords:
Compressive Strength, Machine Learning, Particle Swarm Optimization (PSO), Splitting Tensile Strength, Steel Fiber-Reinforced Recycled Aggregate ConcreteAbstract
The study selected Multilayer Perceptron (MLP), Gaussian Process Regression (GPR), and Extreme Gradient Boosting (XGBoost) as the ML algorithms to develop models that estimate the compressive strength (fcu) and splitting tensile strength (fsp) of steel fiber-reinforced recycled aggregate concrete (SFR-RAC). The study had 465 compressive strength and 339 splitting tensile strength samples from concrete mixes with contrasting proportions. The training and evaluation of models used an 80/20 split, and their hyperparameters were improved through Particle Swarm Optimization (PSO). Assessing how well the models work was done with four statistical measurements: coefficient of determination R², mean absolute error (MAE), root mean squared error (RMSE), and mean absolute percentage error (MAPE). XGBoost could predict outcomes more effectively than any model, and RF or MLP were close behind. To find out how the inputs affect the model results, feature importance analysis and SHapley Additive exPlanations (SHAP) were carried out. It was shown that the water content, the amount of cement, and the proportion of fibers in the concrete all affect its strength. These proposed models explain in detail how SFR-RAC mixtures work, which helps create environmentally friendly concretes with outstanding strength. Later research could use more data and predictor variables to see if these models apply well.
References
Carneiro, J. A., Lima, P. R. L., Leite, M. B., & Toledo Filho, R. D. (2014). Compressive stress–strain behavior of steel fiber reinforced-recycled aggregate concrete. Cement and Concrete Composites, 46, 65–72.
Chen, Y., He, Q., Liang, X., Jiang, R., & Li, H. (2022). Experimental investigation on mechanical properties of glass fiber reinforced recycled aggregate concrete under uniaxial cyclic compression. Cleaner Materials, 6, 100164.
Chen, Y., Lin, H., Cao, R., & Zhang, C. (2021). Slope stability analysis considering different contributions of shear strength parameters. International Journal of Geomechanics, 21(4), 04020265.
Ergen, F., & Katlav, M. (2024). Investigation of optimized machine learning models with PSO for forecasting the shear capacity of steel fiber-reinforced SCC beams with/out stirrups. Journal of Building Engineering.
Freling, R. (2024). Integrated vehicle and crew scheduling in public transport. CiteseerX.
Huang, X., Zhang, J., Sresakoolchai, J., & Kaewunruen, S. (2021). Machine learning aided design and prediction of environmentally friendly rubberised concrete. Sustainability, 13, 1691.
He, Y. (2014). Experimental study on the flexural behavior of steel fiber recycled concrete beams. Buildings, 13(12), 3046. https://doi.org/10.3390/buildings13123046
Hoang, H. G. T., Nguyen, T. A., & Ly, H. B. (2025). Interpretable machine learning models for concrete compressive strength prediction. Innovative Infrastructure Solutions.
Hu, W. (2013). Experimental study on tensile strength of steel fiber recycled concrete. Research and Application of Building Materials, 3–5.
Jagadesh, P., de Prado-Gil, J., Silva-Monteiro, N., & Martínez-García, R. (2023). Assessing the compressive strength of self-compacting concrete with recycled aggregates from mix ratio using machine learning approach. Journal of Materials Research and Technology, 24, 1483–1498. https://doi.org/10.1016/j.cscm.2022.e01046
Jalilifar, H., Sajedi, F., & Kazemi, S. (2016). Investigation on the mechanical properties of fiber reinforced recycled concrete. Civil Engineering Journal, 2, 13–22. https://doi.org/10.28991/cej-2016-00000009
Jagadesh, P., de Prado-Gil, J., Silva-Monteiro, N., & Martínez-García, R. (2023). Assessing the compressive strength of self-compacting concrete with recycled aggregates from mix ratio using machine learning approach. Journal of Materials Research and Technology, 24, 1483–1498.
Jin, S. (2014). Experimental study on flexural properties of layer type hybrid-steel fiber reinforced high strength recycled concrete beams. Yanbian University.
Jin, L., Yao, H., & Li, X. (2024). A hybrid machine learning model using PSO and SHAP for predicting concrete strength in diverse environmental conditions. Construction and Building Materials.
Kashem, A., Anzer, A., Jagirdar, R., & Sojib, M. S. (2025). Metaheuristic-based machine learning approaches of compressive strength forecasting of steel fiber reinforced concrete with SHapley Additive exPlanations.
Kazmi, S. M. S., Munir, M. J., Wu, Y.-F., Patnaikuni, I., Zhou, Y., Xing, F. (2019). Axial stress-strain behavior of macro-synthetic fiber reinforced recycled aggregate concrete. Cement and Concrete Composites, 97, 341–356.
Koopialipoor, M., Asteris, P. G., Mohammed, A. S., Alexakis, D. E., Mamou, A., & Armaghani, D. J. (2022). Introducing stacking machine learning approaches for the prediction of rock deformation. Transportation Geotechnics, 34, 100756.
Kotwal, S., Singh, H., & Kumar, R. (2022). Experimental investigation of steel fibre reinforced self-compacting concrete (SCC) using recycled aggregates as partial replacement of coarse aggregates. Materials Today: Proceedings, 48, 1032–1037.
Kou, S. C., Poon, C. S., & Chan, D. (2007). Influence of fly ash as cement replacement on the properties of recycled aggregate concrete. Journal of Materials in Civil Engineering, 19, 709–717.
Li, P., Liu, Z., Lu, Y., Lin, C., & Ma, W. (2023). Mechanical behavior of steel fiber reinforced recycled aggregate concrete under dynamic triaxial compression. Composite Structures, 320, 117161.
Li, J., Cheng, J., Shi, J., & Huang, F. (2012). Brief introduction of back propagation (BP) neural network algorithm and its improvement. Advances in Computer Science and Information Engineering, 553–558.
Li, Z., Zhang, Y., & Huang, Q. (2025). Advanced machine learning models for prediction of material properties with optimization. Mathematical Modelling of Engineering Problems.
Liu, L., & Zhang, C. (2024). Metaheuristic optimization of machine learning models for predicting concrete compressive strength. SpringerLink Journal of Computational Design and Engineering.
Liu, Z., Fang, L., Jiang, D., & Qu, R. (2022). A machine-learning based fault diagnosis method with adaptive secondary sampling for multiphase drive systems. IEEE Transactions on Power Electronics.
Miah, M. S., Rahman, M. A., & Tareq, M. (2024). Hybrid machine learning models for forecasting the compressive strength of concrete under varying mix designs. Materials Science and Engineering.
Moayedi, H., Mehrabi, M., Bui, D. T., Pradhan, B., & Foong, L. K. (2020). Fuzzy-metaheuristic ensembles for spatial assessment of forest fire susceptibility. Journal of Environmental Management, 260, 109867.
Peng, Y., & Unluer, C. (2023). Modeling the mechanical properties of recycled aggregate concrete using hybrid machine learning algorithms. Resources, Conservation and Recycling, 190, 106812.
Rafiei, M. H., Khushefati, W. H., Demirboga, R., & Adeli, H. (2016). Neural network, machine learning, and evolutionary approaches for concrete material characterization. ACI Materials Journal, 113, 781–789. https://doi.org/10.14359/51689340
Singh, P. K., & Rajhans, P. (2023). Experimental investigation and SVR model to predict the mechanical properties of RAC by enhancing the characteristic of RCA using surface treatment method along with modified mixing approach. Construction and Building Materials, 393, 132032. https://doi.org/10.1016/j.conbuildmat.2023.132032
Shin, D. W., & Hwang, Y. (2025). PSO-optimized XGBoost for concrete strength prediction. Advanced Structural Engineering.
Tan, Z., & Wong, Y. (2024). Hybrid models for concrete property prediction: A comparative study using machine learning and PSO. Journal of Engineering Science and Technology.
Urkude, N., Hora, M. S., & Utkarsh. (2024). Machine learning model for predicting structural performance of concrete using optimization algorithms. Materials Science Forum.
Wang, Y., Zou, R., Liu, F., Zhang, L., & Liu, Q. (2021). A review of wind speed and wind power forecasting with deep neural networks. Applied Energy, 304, 117766.
Xie, S.-J., Lin, H., Chen, Y.-F., & Wang, Y.-X. (2021). A new nonlinear empirical strength criterion for rocks under conventional triaxial compression. Journal of Central South University, 28, 1448–1458.
Xu, J., Zhao, X., Yu, Y., Xie, T., Yang, G., & Xue, J. (2019). Parametric sensitivity analysis and modelling of mechanical properties of normal-and high-strength recycled aggregate concrete using grey theory, multiple nonlinear regression and artificial neural networks. Construction and Building Materials, 211, 479–491.
Zhao, G., He, X., & Wu, F. (2025). Enhanced machine learning models for the prediction of concrete properties using SHAP and PSO optimization. Computational Materials Science.
Zhang, W., & Tang, Z. (2021). Numerical modeling of response of CFRP–Concrete interfaces subjected to fatigue loading. Journal of Composites for Construction, 25, 04021043.
Zhan, C., Dai, Z., Soltanian, M. R., & Zhang, X. (2022). Stage-wise stochastic deep learning inversion framework for subsurface sedimentary structure identification. Geophysical Research Letters, 49, e2021GL095823.
Zhang, Y., Si, Z., Huang, L., Yang, C., & Du, X. (2023). Experimental study on the properties of internal cured concrete reinforced with steel fibre. Construction and Building Materials, 393, 132046.
Zhou, M., Wang, W., & Liu, J. (2024). PSO-based machine learning algorithms for concrete strength prediction with SHAP optimization. Computers & Concrete.
Downloads
Published
Issue
Section
License
Copyright (c) 2025 Hamza Naciri, Ouahib Alaoui, Hamza Zaouri, Jinjun Xu
This work is licensed under a Creative Commons Attribution 4.0 International License.
