A MACHINE LEARNING-DRIVEN COMPUTATIONAL MODEL FOR PREDICTION OF COMPRESSIVE STRENGTH AND MECHANICAL PERFORMANCE OF SUSTAINABLE CONCRETE COMPOSITE CONCRETE
Abstract
The use of waste glass in concrete offers a sustainable solution for reducing environmental waste while conserving natural resources. This study investigates the mechanical performance of concrete in which crushed waste glass is used as a partial replacement for natural fine aggregate at replacement levels of 0%, 5%, 15%, and 20%. Experimental tests were conducted to determine compressive strength, flexural strength, splitting tensile strength, modulus of elasticity, density, ultrasonic pulse velocity, water absorption, and alkali–silica reaction behaviour at curing ages of 7, 14, and 28 days. The results showed that concrete containing 5% to 15% waste glass achieved improved long-term mechanical performance without harmful expansion, indicating this range as optimal for sustainable concrete production. In addition to experimental evaluation, machine learning models were developed to predict compressive strength using material composition parameters. Artificial Neural Network, Random Forest, and linear regression models were implemented in Python using the scikit-learn framework, while NumPy and Pandas were used for data processing and augmentation, and Matplotlib was employed for graphical visualization. Model robustness was enhanced using data augmentation and 5-fold cross-validation. Among the developed models, the Random Forest algorithm demonstrated the highest prediction accuracy, followed by the Artificial Neural Network, while linear regression showed comparatively lower performance. The study confirms that combining experimental investigation with AI- and ML-based modelling provides a reliable and practical approach for predicting the performance of sustainable concrete incorporating waste glass.
KEYWORDS: Waste glass concrete; Sustainable materials; Artificial neural network; Random forest; Machine learning; Mechanical properties
Full Text:
PDFReferences
Abdelkader, E. M., Elshafie, M., & Rashad, A. M. (2022). Machine learning prediction of mechanical properties of sustainable concrete incorporating waste materials. Journal of Building Engineering, 50, 104127.
Afshinnia, K., & Rangaraju, P. R. (2021). Impact of glass powder on mechanical and durability properties of concrete. Construction and Building Materials, 273, 121693.
Aliabdo, A. A., Abd Elmoaty, M., & Aboshama, A. Y. (2020). Utilization of waste glass powder in the production of cement and concrete. Construction and Building Materials, 242, 118142.
Almeshal, I., Tayeh, B. A., & Alyousef, R. (2022). Sustainable concrete incorporating recycled glass aggregates. Sustainability, 14(5), 2748.
Breiman, L. (2001). Random forests. Machine Learning, 45(1), 5–32. https://doi.org/10.1023/A:1010933404324
Chou, J. S., Pham, A. D., & Nguyen, T. K. (2021). Predicting compressive strength of concrete using machine learning. Engineering with Computers, 37(4), 2341–2357.
Du, H., & Tan, K. H. (2021). Properties of concrete incorporating recycled glass aggregate: A review. Journal of Cleaner Production, 318, 128517.
Flood, I., & Kartam, N. (1994). Neural networks in civil engineering. Journal of Computing in Civil Engineering, 8(2), 126–145.
Goodfellow, I., Bengio, Y., & Courville, A. (2016). Deep learning. MIT Press.
Haykin, S. (2009). Neural networks and learning machines (3rd ed.). Pearson.
Ismail, Z. Z., & Al-Hashmi, E. A. (2022). Reuse of waste glass in concrete: A critical review. Materials Today: Proceedings, 58, 2045–2052.
Kou, S. C., & Poon, C. S. (2020). Long-term performance of concrete with recycled glass. Cement and Concrete Composites, 113, 103724.
Li, X., Wang, Y., & Zhou, Z. (2025). Data-driven mix design of sustainable concrete using ensemble learning. Automation in Construction, 160, 105183.
Nguyen, H., Zhang, Y., & Li, H. (2022). ANN-based prediction of concrete strength using mixture parameters. Materials, 15(9), 3114.
Nguyen, H., Zhang, Y., & Li, H. (2024). Explainable machine learning models for sustainable concrete. Construction and Building Materials, 379, 131133.
Shao, Y., Lefort, T., & Moras, S. (2023). Alkali–silica reaction mitigation in glass-containing concrete. Cement and Concrete Research, 168, 107183.
Zhang, J., Zhao, Y., & Wang, Z. (2023). Random Forest models for eco-friendly concrete strength prediction. Sustainability, 15(3), 2217.
Refbacks
- There are currently no refbacks.
Copyright © 2022-2025. Department of Civil Engineering, Nnamdi Azikiwe University. All Rights Reserved..