Unraveling the impact of machine learning on the AEC sector: academia and industry perspectives

The implementation of artificial intelligence (AI) and machine learning (ML) in the architecture, engineering, and construction (AEC) domain has gained significant attention within academic research and construction technology companies. The scientific field often relies on information collected from the industry for theoretical development and sometimes test cases. However, to the best of the authors’ knowledge, there has been no systematic effort to compare academic and industry trends and developments. In this work, we examine the development and application of ML in the AEC domain from both academic and industry perspectives. To investigate both perspectives, we implement a mixed-methods approach including an academic literature review, a web-based mapping of construction technology companies worldwide, and an online questionnaire targeting practitioners and technology companies. Overall, our mixed methods analysis reveals a strong alignment between academic research and industry practice in targeting early design and post construction phases, with both communities prioritizing the application of ML for energy efficiency, facility management, and site safety, particularly within the context of Building Information Modeling (BIM). However, industry devotes more effort to sustainability, cost and scheduling solutions, whereas academia lags in these areas and often relies on limited or synthetic datasets. Finally, both sectors identify data availability and quality, particularly the scarcity of large, labeled, domain-specific repositories, as the primary barrier to wider ML adoption in the AEC industry.