A Systematic Review on Intelligent Prediction of Inorganic Building Materials Performance

Abstract The construction industry is crucial for economic and social development. Inorganic materials, which rely on natural minerals and are affected by uncertainties, hold a large share in the construction market. As building materials are from process—intensive industries, complex and continuous processing magnifies deviations, directly affecting product quality. Computational intelligent methods are effective for accurately predicting product quality. This paper focuses on inorganic building materials and systematically reviews computational intelligent techniques in this field. It comprehensively explores 234 related studies in 6 key areas (concrete, ceramics, glass, clay bricks, cement, and steel), compiles prediction models, evaluates them, analyzes model configurations and properties to gain insights into the field and identify optimal approaches. It points out model limitations, such as high computational costs, data-hungry, and suggests future research directions like practicality and promoting green initiatives through material circulation.