Artificial Neural Network-Based Prediction of the Compressive Strength of Lightweight Concrete Reinforced with Industrial and Recycled Steel Fibers
Abstract
The incorporation of industrial waste into concrete has emerged as an effective strategy for improving sustainability while reducing the environmental impacts associated with construction materials. Among these recycled materials, waste steel fibers have shown considerable potential as concrete reinforcement. To ensure their efficient utilization, however, reliable prediction of concrete performance is essential. Since direct measurement of compressive strength is often expensive, time-consuming, or impractical for existing structures, the development of accurate predictive models offers an attractive alternative. This research proposes an artificial intelligence-based approach for estimating the compressive strength of concrete reinforced with recycled steel fibers. An Artificial Neural Network (ANN) trained using the Levenberg–Marquardt algorithm was developed from an experimental database consisting of 45 different concrete mix designs. The input variables included steel fiber content, cement content, water content, water-to-cement ratio, and superplasticizer dosage, while the output parameter was the compressive strength measured at curing ages of 7, 28, and 60 days. For model development, the dataset was divided into training (75%), testing (15%), and validation (15%) subsets. Experimental observations demonstrated that incorporating recycled steel fibers enhanced the compressive strength of concrete, which can be attributed to the effective dispersion of fibers within the cementitious matrix and the use of an appropriate fiber dosage. The developed ANN successfully captured the relationship between mixture proportions and compressive strength, producing highly accurate predictions. The validation results yielded a correlation coefficient exceeding 99%, confirming the robustness, reliability, and predictive capability of the proposed neural network model for estimating the compressive strength of steel fiber-reinforced concrete.
Keywords:
Compressive strength, Synthetic neural network, Steel fibers, ConcreteReferences
- [1] Alshihri, M. M., Azmy, A. M., & El-Bisy, M. S. (2009). Neural networks for predicting compressive strength of structural light weight concrete. Construction and building materials, 23(6), 2214–2219. https://doi.org/10.1016/j.conbuildmat.2008.12.003
- [2] Bui, D.-K., Nguyen, T. N., Ngo, T. D., & Nguyen-Xuan, H. (2020). An artificial neural network (ANN) expert system enhanced with the electromagnetism-based firefly algorithm (EFA) for predicting the energy consumption in buildings. Energy, 190, 116370. https://doi.org/10.1016/j.energy.2019.116370
- [3] Li, Y., Zhang, Q., Kamiński, P., Deifalla, A. F., Sufian, M., Dyczko, A., … ., & Atig, M. (2022). Compressive strength of steel fiber-reinforced concrete employing supervised machine learning techniques. Materials, 15(12), 1–18. https://doi.org/10.3390/ma15124209
- [4] Ramkumar, K. B., Kannan Rajkumar, P. R., Noor Ahmmad, S., & Jegan, M. (2020). A review on performance of self-compacting concrete – use of mineral admixtures and steel fibres with artificial neural network application. Construction and building materials, 261, 120215. https://doi.org/10.1016/j.conbuildmat.2020.120215
- [5] Altun, F., Kişi, Ö., & Aydin, K. (2008). Predicting the compressive strength of steel fiber added lightweight concrete using neural network. Computational materials science, 42(2), 259–265. https://doi.org/10.1016/j.commatsci.2007.07.011
- [6] Uysal, M., & Tanyildizi, H. (2012). Estimation of compressive strength of self compacting concrete containing polypropylene fiber and mineral additives exposed to high temperature using artificial neural network. Construction and building materials, 27(1), 404–414. https://doi.org/10.1016/j.conbuildmat.2011.07.028
- [7] Hameed, M. M., & AlOmar, M. K. (2020). Prediction of compressive strength of high-performance concrete: hybrid artificial intelligence technique. Applied computing to support industry: Innovation and technology (pp. 323–335). Cham: Springer International Publishing. https://doi.org/10.1007/978-3-030-38752-5_26%0A%0A
- [8] Erdal, H. I. (2013). Two-level and hybrid ensembles of decision trees for high performance concrete compressive strength prediction. Engineering applications of artificial intelligence, 26(7), 1689–1697. https://doi.org/10.1016/j.engappai.2013.03.014
- [9] Nagarajan, D., Rajagopal, T., & Meyappan, N. (2020). A comparative study on prediction models for strength properties of LWA concrete using artificial neural network. Revista de la construcción, 19(1), 103–111. http://dx.doi.org/10.7764/rdlc.19.1.103-111
- [10] Asteris, P. G., & Mokos, V. G. (2020). Concrete compressive strength using artificial neural networks. Neural computing and applications, 32(15), 11807–11826. https://doi.org/10.1007/s00521-019-04663-2
- [11] Nejati, F., & Mansourkia, A. (2023). Prediction of the compressive strength of lightweight concrete containing industrial and waste steel fibers using a multilayer synthetic neural network. Advances in bridge engineering, 4(1), 20. https://doi.org/10.1186/s43251-023-00100-6
- [12] N.S, N. P., P, K., & P, S. (2026). AI-driven sustainable strength prediction and experimental evaluation of high-performance fiber-reinforced concrete incorporating metakaolin. Scientific reports, 16(1), 13614. https://doi.org/10.1038/s41598-026-41115-z
- [13] Shi, S., & Xing, Z. (2025). Modeling of compressive strength of recycled concrete under different loading conditions by steel fibers based on back propagation neural network. International journal for housing science & its applications, 46(3), 6609. https://doi.org/10.70517/ijhsa463568
- [14] Stel’makh, S. A., Shcherban’, E. M., Beskopylny, A. N., Mailyan, L. R., Meskhi, B., Razveeva, I., … ., & Beskopylny, N. (2022). Prediction of mechanical properties of highly functional lightweight fiber-reinforced concrete based on deep neural network and ensemble regression trees methods. Materials, 15(19), 1–18. https://doi.org/10.3390/ma15196740
Issue 1