Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

ARCEAK: An Automated Rule Checking Framework Enhanced with Architectural Knowledge

Junyong Chen, Ling-I Wu, Minyu Chen, Xiaoying Qian, Haoze Zhu, Qiongfang Zhang, Guoqiang Li

TL;DR

This paper tackles automated rule checking (ARC) in the Architecture, Engineering, and Construction domain by addressing the bottleneck of translating regulatory text into machine-executable checks. It introduces ARCEAK, a two-stage framework that first extracts structured rule information (ED/EE) using knowledge-augmented prompts, then generates executable verification code via a knowledge-informed code framework and completion stage. Empirical results show substantial gains in information extraction and competitive performance in code generation, with knowledge augmentation markedly improving alignment to domain rules and model APIs. The approach promises near-full automation of translating building rules into verification code, enabling more reliable and scalable construction compliance while outlining practical avenues and limitations for real-world deployment.

Abstract

Automated Rule Checking (ARC) plays a crucial role in advancing the construction industry by addressing the laborious, inconsistent, and error-prone nature of traditional model review conducted by industry professionals. Manual assessment against intricate sets of rules often leads to significant project delays and expenses. In response to these challenges, ARC offers a promising solution to improve efficiency and compliance in design within the construction sector. However, the main challenge of ARC lies in translating regulatory text into a format suitable for computer processing. Current methods for rule interpretation require extensive manual labor, thereby limiting their practicality. To address this issue, our study introduces a novel approach that decomposes ARC into two distinct tasks: rule information extraction and verification code generation. Leveraging generative pre-trained transformers, our method aims to streamline the interpretation of regulatory texts and simplify the process of generating model compliance checking code. Through empirical evaluation and case studies, we showcase the effectiveness and potential of our approach in automating code compliance checking, enhancing the efficiency and reliability of construction projects.

ARCEAK: An Automated Rule Checking Framework Enhanced with Architectural Knowledge

TL;DR

This paper tackles automated rule checking (ARC) in the Architecture, Engineering, and Construction domain by addressing the bottleneck of translating regulatory text into machine-executable checks. It introduces ARCEAK, a two-stage framework that first extracts structured rule information (ED/EE) using knowledge-augmented prompts, then generates executable verification code via a knowledge-informed code framework and completion stage. Empirical results show substantial gains in information extraction and competitive performance in code generation, with knowledge augmentation markedly improving alignment to domain rules and model APIs. The approach promises near-full automation of translating building rules into verification code, enabling more reliable and scalable construction compliance while outlining practical avenues and limitations for real-world deployment.

Abstract

Automated Rule Checking (ARC) plays a crucial role in advancing the construction industry by addressing the laborious, inconsistent, and error-prone nature of traditional model review conducted by industry professionals. Manual assessment against intricate sets of rules often leads to significant project delays and expenses. In response to these challenges, ARC offers a promising solution to improve efficiency and compliance in design within the construction sector. However, the main challenge of ARC lies in translating regulatory text into a format suitable for computer processing. Current methods for rule interpretation require extensive manual labor, thereby limiting their practicality. To address this issue, our study introduces a novel approach that decomposes ARC into two distinct tasks: rule information extraction and verification code generation. Leveraging generative pre-trained transformers, our method aims to streamline the interpretation of regulatory texts and simplify the process of generating model compliance checking code. Through empirical evaluation and case studies, we showcase the effectiveness and potential of our approach in automating code compliance checking, enhancing the efficiency and reliability of construction projects.
Paper Structure (20 sections, 5 figures, 8 tables)

This paper contains 20 sections, 5 figures, 8 tables.

Table of Contents

  1. Introduction
  2. Background
  3. Large Language Model
  4. Information Extraction
  5. Code Generation
  6. Methodology
  7. Overview
  8. Preprocess
  9. Rule Information Extraction
  10. Verification Code Generation
  11. Evaluation
  12. Discussion
  13. Case Study: Correct and Incorrect Results
  14. Generating API or parameter
  15. Generating verification code for complex rule entry
  16. ...and 5 more sections

Figures (5)

  • Figure 1: The overall architecture of proposed ARCEAK
  • Figure 2: An example of preprocessing result
  • Figure 3: An example of code generation result
  • Figure 4: An example on compatibility between the LLMs and selected model. The code above is generated without external information and the code below is generated with entity information extracted in Entity Discovery.
  • Figure 5: An example of granularity in function generation. The code above is generated without knowledge augmentation and the code below is generated with knowledge augmentation