Technical Name |
Generative Artificial Intelligence in Construction Quality Control |
Project Operator |
National Taiwan University |
Project Host |
陳俊杉 |
Summary |
This technology utilizes Retrieval-augmented Generation (RAG), Agentic Reasoning, and Knowledge Graphs to implement a generative AI computation service (AI as a Service, AIaaS) that facilitates the code compliance and generation of construction inspection forms based on construction specifications and drawings. It offers an innovative solution to longstanding challenges faced by engineering consulting firms and significantly enhances productivity. |
Scientific Breakthrough |
This technology utilizes Retrieval-augmented Generation (RAG), Agentic Reasoning, and Knowledge Graphs to achieve the code compliance and generation of construction inspection forms based on construction specifications and drawings. Retrieval-augmented Generation traces key specifications and drawings, Agentic Reasoning enhances the quality of verification outcomes, and Knowledge Graphs drive form design, making this a leading generative AI technology in the field of construction inspection. |
Industrial Applicability |
Sinotech company (our partner) evaluated the industrial benefits of this technology in terms of operational improvements, quality enhancement, knowledge management and talent cultivation, with specific performance gains exceeding NT$1.2 million per year. The AI as a Service (AIaaS) developed through this technology is a pivotal generative AI application in civil engineering management. It has strong spillover effects and can be subsequently applied to labor-intensive engineering tasks. |
Keyword |
Generative Artificial Intelligence Large Language Models Retrieval-augmented Generation Knowledge Graphs Agentic Reasoning Construction Specifications Construction Drawings Inspection Forms Quality Control Code Compliance |