Artificial Intelligence-Driven Construction Defect Identification System in Construction: Adoption Opportunities and Implementation Barriers

The integration of Artificial Intelligence (AI) within the Architecture, Engineering, and Construction (AEC) sector has transformed traditional project delivery by improving efficiency, safety, and decision-making. However, the use of AI in defect detection and management a critical yet underexplored domain remains limited in methodological depth. This study conducts a systematic and bibliometric review using the Scopus database, analyzed through Microsoft Excel to evaluate AI’s efficacy in identifying, classifying, and managing construction defects. Publications between 2020 and 2024 were screened based on inclusion and exclusion criteria, focusing on AI applications in Construction Defect Identification Systems (CDIS). Quantitative synthesis indicates that AI-driven methods such as pre-trained Convolutional Neural Networks (~99%), laser vision (97.10%), and Operational Modal Analysis (97%) consistently outperform conventional methods like grey correlation analysis (86.81%) and visual inspection (80%) by 10–20% in accuracy. Despite these advances, research in predictive and real-time defect assessment remains fragmented. The study emphasizes the need for integrated, data-driven frameworks that leverage AI’s predictive capabilities to establish proactive defect management systems, thereby enhancing quality standards and sustainable practices in the AEC industry.