ITcon Vol. 31, pg. 201-224, http://www.itcon.org/2026/9

Enhancing regulatory compliance in AEC industry via LLM-powered decision frameworks

DOI:10.36680/j.itcon.2026.009
submitted:September 2025
published:February 2026
editor(s):Kumar B
authors:Entesar Al Nama, PhD
University of Bahrain
entesarjasim@gmail.com

Maqsood Mahmud, Assistant Professor
School of Computing, Faculty of Computing, Engineering and the Built Environment (CEBE),
Ulster University, York Street, BT151AP, Belfast, United Kingdom
m.mahmud@ulster.ac.uk

Huda Al Madhoob, Dr.
College of Engineering, University of Bahrain
halmadhoob@uob.edu.bh
summary:The AEC industry is a highly intricate ecosystem involving architects, structural engineers, civil consultants, and contractors working under intense time pressures, where the success of projects hinges on sound decision-making. One of the most persistent challenges is regulatory compliance—not merely understanding the rules, but accurately interpreting and applying them within real-world constraints. This research paper aims to enhance regulatory decision-making through AI, specifically a Q&A model, and stands out in five key ways: it tackles the widespread issue of building code violations that cause costly delays; it empowers professionals to verify compliance during the design phase, minimizing errors and saving resources; it proves that effective AI doesn’t require massive datasets, instead leveraging domain expertise and smart data strategies; and it introduces a scalable framework that can extend to broader regulatory domains and integrate with BIM tools for automated checks, offering a transformative approach to compliance in the AEC sector.
keywords:Bahrain, AEC, LLM, ChatGPT, Gemini, compliance to building regulations, decision-making, stakeholders, AI, data expansion
full text: (PDF file, 0.948 MB)
citation:Al Nama, E., Mahmud, M., & Al Madhoob, H. (2026). Enhancing regulatory compliance in AEC industry via LLM-powered decision frameworks. Journal of Information Technology in Construction (ITcon), 31, 201-224. https://doi.org/10.36680/j.itcon.2026.009
statistics: