Predictive analysis of brick powder-enhanced self-compacting mortar using artificial neural networks

Introduction/purpose: Self-compacting mortars possess very good flowability which enables them to consolidate only through the force of their self-weight, not demanding any mechanical vibration. This type of mortars has a significant value in complicated construction and repair works. In this paper, research will be conducted on using brick powder as replacement for cement in SCM with a detailed study of effects on workability, compressive strength, and all other performance. Methods: The proposed research models the relationships of different parameters, brick powder content and fineness, with the resultant properties of mortar using artificial neural networks (ANN) models. Results: The results showed that the addition of brick powder up to 10% of cement as replacement improves mortar workability, giving the slump flow ranging from 306 to 309 mm and the funnel flow time between 4.8 and 5.4 s, while its compressive strength was ranging from 45 to 60 MPa at 28 days. However, at replacement levels higher than 20%, the slump flow was reduced to 285 mm, the time to funnel flow increased to 9 s, and the compressive strength decreased to 35 MPa. Conclusion: The study illustrates brick powder as a promising recycled material for SCM applications not only to reduce environmental impacts but also to improve performance, although optimization of its replacement levels should be taken carefully to balance workability and strength.