A Knowledge Graph–Augmented Large Language Model for Automated Quality Compliance Checking in Construction
TL;DR
This research proposes a large language model–based approach combined with structured algorithms that process domain-specific competency questions and unstructured construction documents to build ontologies and extract structured knowledge graphs to support the automated generation of quality inspection checklists and compliance checking in construction.
Abstract
Our research aims to investigate how large language models (LLMs) can be aligned with diverse domain knowledge to support the automated generation of quality inspection checklists and compliance checking in construction. To achieve this, we propose a large language model–based approach combined with structured algorithms that process domain-specific competency questions and unstructured construction documents to build ontologies and extract structured knowledge graphs. An experiment was conducted to demonstrate the feasibility and effectiveness of our approach, showing that our method effectively balances resource efficiency with the complexity of construction quality data, accurately capturing and representing key domain information. Our approach benefits practitioners, researchers, and stakeholders by supporting timely planning, execution, and verification of construction quality on-site. Furthermore, our findings lay a foundation for future research and practical development of smarter, automated construction quality management systems.
