A Systematic Literature Review on the Applications of Deep Learning and Artificial Intelligence in Construction Supply Chain Management

Abstract

The construction industry increasingly confronts challenges related to inefficiency, cost overruns, and supply chain vulnerabilities, while the rapid evolution of deep learning and artificial intelligence offers novel pathways for transformation. This systematic literature review aims to comprehensively map and synthesize the existing body of research on how these technologies are applied across construction supply chain management. We followed a rigorous and reproducible systematic review methodology, involving a structured search, a multi-stage screening process, and a thematic analysis of the selected studies. The review results reveal a fragmented but rapidly growing research landscape. We identified seven principal dimensions of application: AI-driven optimization for demand forecasting and logistics, risk management and disruption prediction, integration within broader project management contexts, sustainability and carbon footprint tracking, the convergence with emerging technologies like digital twins and blockchain, sector-specific adaptations, and strategic frameworks addressing adoption barriers. A key finding is that while predictive and optimization models are prevalent, research on generative AI and holistic digital twin integration remains nascent. Furthermore, our synthesis shows that studies often focus on isolated technical performance metrics rather than on system-level impact or real-world deployment challenges. Consequently, we conclude that a significant gap exists between algorithm development and practical, scalable implementation within the fragmented and risk-averse construction sector. This review provides a structured taxonomy of current knowledge and critically identifies the most pressing directions for future research, particularly the need for robust validation in dynamic project environments and the development of integrated frameworks that address both technical and organizational facets of AI adoption.