The integration of Artificial Intelligence (AI) has revolutionized the way AAC is produced and applied, addressing key challenges related to cost, efficiency, and environmental sustainability. AI-driven optimization algorithms have significantly enhanced AAC manufacturing processes by reducing material waste, improving resource allocation, and minimizing energy usage. This not only ensures cost-effective production but also supports eco-friendly practices.

AI has also introduced advanced predictive maintenance systems that monitor the health and performance of AAC structures in real-time. These systems can detect potential vulnerabilities early and recommend timely interventions, which extend the lifespan of infrastructure and help avoid costly repairs. Furthermore, AI-based decision support tools have improved project planning and execution by optimizing timelines, resource distribution, and scheduling. This ensures efficient completion of construction projects, minimizing delays and preventing budget overruns, thus fostering more streamlined and effective operations in the construction industry.

1. Machine learning in AAC development

Machine learning models have become indispensable tools in the development and optimization of alkali-activated concrete. These models provide highly accurate predictions of key properties such as compressive strength, cost, and carbon footprint, enabling engineers and architects to make informed decisions in the design and application of AAC. By leveraging the analytical power of machine learning, researchers can gain deeper insights into the material’s performance, ultimately advancing sustainable construction practices.

2. Advanced machine learning techniques

Several advanced machine learning techniques have proven effective in the study of AAC. Among these, XGBoost stands out for its exceptional prediction accuracy, consistently achieving R² values exceeding 91%. This makes it a preferred choice for tasks requiring high precision. Meanwhile, Random forest (RF) is renowned for its reliability in managing non-linear relationships and handling complex datasets, ensuring robust and dependable predictions. On the other hand, Artificial neural networks (ANN) excel in identifying intricate patterns within the data, achieving correlation coefficients as high as 96%, showcasing their capability to model complex material behaviors.

3. Key parameters identified by machine learning

Machine learning techniques have also been instrumental in identifying critical parameters that influence AAC properties. Curing age and temperature are two such factors, as they play a vital role in determining the compressive strength of the material. Additionally, the material composition, particularly the concentrations of slag and sodium hydroxide, significantly impacts AAC’s performance and durability. Another crucial consideration is the liquid-to-solid ratio; lower ratios are associated with improved compressive strength, emphasizing their importance in mix design optimization.

4. Bridging the research gap with AI

Despite the progress achieved through machine learning, there remains a need for more comprehensive exploration of AI-driven methodologies to fully optimize AAC. Researchers are focused on integrating AI and engineering principles to address this gap. Their goals include developing predictive models that accurately estimate compressive strength, cost, and environmental impact. Additionally, they aim to elucidate the intricate relationships between mix design parameters and AAC performance, there by enhancing material efficiency. To facilitate practical applications, efforts are underway to create a user-friendly Graphical user interface (GUI) that allows users to make preliminary estimates without incurring the costs of extensive experimentation. This innovative approach promises to bridge the gap between research and application, paving the way for a new era of sustainable and efficient construction practices.