Semi-Automated Generation of Logic Rules for Tabular Information in Building Codes to Support Automated Code Compliance Checking

Oct 4, 2026

Upload by SANS

Fully automated building code compliance checking requires extracting and transforming regulatory requirements from building codes into machine-processable forms. Existing work emphasizes requirements stated in text, while effective processing of requirements embedded in tables or charts remains insufficient. The proposed semi-automated approach extracts table-based requirements, transforms them into executable logic rules, and supports user refinement. Experiments using the International Building Code 2015 show iterative extraction accuracy improvement and successful rule representation for a Texas convenience store.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages35

Reading Time1 h 28 min

AccessFree to download

Abstract

Introduction

1 Semi-Automated Generation of Logic Rules for Tabular Information in Building Codes to 2 Support Automated Code Compliance Checking 3 Xiaorui Xue, S.M.ASCE 1 ; Jin Wu, S.M.ASCE 2 ; Jiansong Zhang, Ph.D., A.M.ASCE 3 * 4 5 Abstract 6 To fully automate building code compliance checking, regulatory requirements need to be 7 automatically extracted and transformed from building codes. Existing regulatory 8 requirement processing efforts mainly focus on building code requirements in text. A more 9 efficient approach for processing regulatory requirements in other parts of building codes, 10 such as tables or charts, remains to be addressed. The ability to process building code 11 requirements in all parts and formats is necessary for an automated code compliance 12 checking system to achieve full coverage of checkable building code requirements. To 13 address this gap, the authors proposed a semi-automated information extraction and 14 transformation method. The proposed method can extract building code requirements in 15 tables and convert extracted information to logic rules. Automated code compliance 16 checking systems can utilize the logic rules. The proposed method includes two main steps: 17 (1) tabular information extraction, and (2) rule generation. The tabular information 1 Automation and Intelligent Construction (AutoIC) Lab, School of Construction Management Technology, Purdue University, West Lafayette, IN, 47907, PH (765) [430-2009. email:](430-2009. email:) 18 extraction semi-automatically detects the layout of tables, extracts building code 19 requirements from tables, and transforms extracted information to databases. The rule 20 generation step automatically generates logic rules that can be directly executed by logic 21 reasoners. The rule generation step also provides options for users to further refine the 22 generated rules. The development of the tabular information extraction algorithm takes an 23 iterative approach. An experiment was conducted to develop a tabular information 24 extraction algorithm from Chapter 5 of the International Building Code 2015. The primary 25 version of the algorithm correctly processed 91.67% of tables in Chapter 10 of the 26 International Building Code 2015. After iterative refinements, the updated tabular 27 information extraction algorithm correctly processed all tables in this chapter. The rule 28 generation algorithm correctly generated logic rules that successfully represented the 29 applicable building code requirements for a testing building, a convenience store in Texas 30 based on the tabular information. 31 Introduction 32 Building Information Modeling (BIM) developed steadily in the past decade (Noor and Yi 33 2018) . Versatile usages of BIM prompted researches of BIM in the Architecture, 34 Engineering, and Construction (AEC) industry, for various purposes such as: (1) modular 35 construction of residential buildings (Alwisy et al. 2012),(2) construction cost estimation 36 (Lee et al. 2014), (3) collaboration of multi-disciplinary teams (Fernando et al. 2013), and 37 (4) energy simulation in a building’s early design stage (Kim et al. 2015) . Researchers also 38 combined BIM with other technologies, such as (1) laser scanning (Kim et al. 2015),(2) 39 Radio-frequency Identification (RFID) (Fang et al. 2016),(3) Virtual Reality (VR) (Du et 40 al. 2018),(4) computer vision (Asadi et al. 2019, Ham et al. 2016, Han et al. 2015), and (5) 41 natural language processing (Mutis et al. 2019, Xu and Cai 2020, Xu and Cai 2019, Xu et 42 al. 2020, Zhang and El-Gohary 2012a,b) to explore its full potential in the AEC industry. 43 Overall, benefits of BIM include: (1) facilitating cooperation among stakeholders (Volk et 44 al. 2014), (2) reducing construction cost and time (Bryde et al. 2013), (3) improving 45 construction safety (Zhang et al. 2013), and (4) decreasing carbon emission and energy 46 consu

FAQs

Why is automated building code compliance checking difficult to fully automate?

Regulatory requirements must be extracted and transformed from building codes into formats that can be executed automatically. Prior efforts focus mainly on text, not on requirements inside tables or charts.

What method is proposed to address missing table/chart processing?

A semi-automated information extraction and transformation method extracts requirements from tables and converts them into logic rules usable by automated compliance checking systems.

What are the two main steps of the proposed method?

(1) tabular information extraction that detects table layouts and extracts requirements into databases, and (2) rule generation that produces executable logic rules with options for user refinement.

Oct 4, 2026

Upload by SANS