The Use of Artificial Neural Network Model to Predict the Compressive Strength of Sustainable Geopolymer Concrete: A Systematic Review

Geopolymer concrete (GPC) is an alternative building material that has gained attention recently as a more environmentally friendly alternative to traditional Portland cement-based concrete. This type of concrete utilizes industrial wastes or agricultural byproduct ashes as the primary source of binder materials, in contrast to traditional Portland cement. GPC is gaining popularity due to the environmental issues related to cement manufacturing. The compressive strength of all concrete composites, including GPC, is a fundamental mechanical characteristic. A novel approach to forecast the compressive strength of GPC is the application of artificial neural networks (ANNs). This review paper aims to introduce artificial neural networks and emphasize their usefulness as a computational technique for modelling complex functional connections among different parameters affecting the compressive strength of GPC. The review paper is structured into three sections. The first section introduces neural networks and their applications as modelling tools. The second section focuses on using neural networks to model GPC made from different source materials. Finally, the third section summarizes the key findings and contributions of the paper, emphasizing the potential of ANNs to revolutionize the design and optimization of geopolymer concrete for more sustainable and eco-friendly construction practices. From the extensive literature review, it is clear that comparing the results derived from ANNs with experimental findings shows excellent agreement, indicating the reliability and robustness of ANNs in estimating the compressive strength of GPC.