Construction & Engineering
AI models predict chloride resistance in recycled aggregate concrete
Researchers combined neural networks and response-surface methods to predict chloride resistance in recycled aggregate concrete.
Researchers developed an AI-assisted approach for designing recycled aggregate concrete with improved resistance to chloride penetration. The method combines artificial neural networks with response surface methodology. Chloride penetration is a durability concern because it can contribute to reinforcement corrosion and reduce the service life of concrete structures. Recycled aggregate concrete can support more sustainable construction, but its performance must be carefully designed for the intended environment. The research uses AI to explore material relationships and identify concrete designs that balance recycled content with durability requirements. This application places machine learning within materials engineering rather than project administration or site automation. It could help engineers evaluate more combinations of mix variables than conventional trial-and-error approaches. The work is relevant to infrastructure and building projects seeking to reduce material impacts while maintaining technical performance. The source describes the approach as an AI-assisted design method, so practical deployment would still require laboratory validation, standards compliance, and field testing. Nevertheless, it demonstrates how AI is being applied to improve the engineering of lower-impact construction materials.
Original source: AZoBuild