Experimental and machine learning assessment of sustainable natural fiber reinforced concrete incorporating waste brick powder as an alternative binder materials

Md. Hamidul Islam, Shuvo Dip Datta, Sanjoy Mondol, Md. Habibur Rahman Sobuz, Jawad Ashraf, Rahat Aayaz, Abu Sayed Mohammad Akid, Md. Kawsarul Islam Kabbo,
Experimental and machine learning assessment of sustainable natural fiber reinforced concrete incorporating waste brick powder as an alternative binder materials,
Ain Shams Engineering Journal,
Volume 16, Issue 12,
2025,
103755,
ISSN 2090-4479,
https://doi.org/10.1016/j.asej.2025.103755.
(https://www.sciencedirect.com/science/article/pii/S2090447925004964)
Abstract: The rising demand for sustainable construction has prompted the use of natural fibers and waste brick powder (WBP) as eco-efficient alternatives in concrete production. This work aims to determine the feasibility of adding jute fiber (JF) as a reinforcing agent and waste brick powder (WBP) as a partial replacement for cement in concrete. Fresh, non-destructive, and microstructural tests for varying percentages of JF and WBP content, whereas compressive strength (CS) and splitting tensile strength were assessed for mechanical performance. Results show that concrete workability decreased as the replacement percentage of WBP and the addition of JF increased. Test results showed strength improvements of about 9.5% and 31.36% in splitting and compressive strength over the control specimen. Afterward, three machine learning models (RF, XGB, and SVR) were applied to predict the CS. RF achieved best performance with R2 of 0.957, while XGB yielded an R2 of 0.954. Furthermore, microstructural assessment showed WBP densified matrix by filling pores which enhanced strength.
Keywords: Waste brick powder; Jute fiber; Mechanical properties; Machine learning; Microstructure; Economic analysis