Automated code compliance checking in the construction domain using semantic natural language processing and logic-based reasoning

Zhang, Jiansong

Permalink

https://hdl.handle.net/2142/89207 Copy

Description

Title

Automated code compliance checking in the construction domain using semantic natural language processing and logic-based reasoning

Author(s)

Zhang, Jiansong

Issue Date

2015-12-02

Director of Research (if dissertation) or Advisor (if thesis)

El-Gohary, Nora

Doctoral Committee Chair(s)

El-Gohary, Nora

Committee Member(s)

• El-Rayes, Khaled
• Liu, Liang
• Girju, Corina Roxana
• Golparvar-Fard, Mani

Department of Study

Civil & Environmental Engineering

Discipline

Civil Engineering

Degree Granting Institution

University of Illinois at Urbana-Champaign

Degree Name

Ph.D.

Degree Level

Dissertation

Keyword(s)

• automated compliance checking
• natural language processing
• building information modeling
• automated information extraction
• automated information transformation
• semantic systems
• automated construction management systems
• automated reasoning
• first order logic
• logic programming
• knowledge modeling
• industry foundation classes (IFC) extension

Abstract

Construction projects must comply with various regulations. The manual process of checking the compliance with regulations is costly, time consuming, and error prone. With the advancement in computing technology, there have been many research efforts in automating the compliance checking process, and many software development efforts led by industry bodies/associations, software companies, and/or government organizations to develop automated compliance checking (ACC) systems. However, two main gaps in the existing ACC efforts are: (1) manual effort is needed for extracting requirements from regulatory documents and encoding these requirements in a computer-processable rule format; and (2) there is a lack of a semantic representation for supporting automated compliance reasoning that is non-proprietary, non-hidden, and user-understandable and testable. To address these gaps, this thesis proposes a new ACC method that: (1) utilizes semantic natural language processing (NLP) techniques to automatically extract regulatory information from building codes and design information from building information models (BIMs); and (2) utilizes a semantic logic-based representation to represent and reason about the extracted regulatory information and design information for compliance checking. The proposed method is composed of four main methods/algorithms that are combined in one computational framework: (1) a semantic, rule-based method and algorithm that leverage NLP techniques to automatically extract regulatory information from building codes and represent the extracted information into semantic tuples, (2) a semantic, rule-based method and algorithm that leverage NLP techniques to automatically transform the extracted regulatory information into logic rules to prepare for automated reasoning, (3) a semantic, rule-based information extraction and information transformation method and algorithm to automatically extract design information from BIMs and transform the extracted information into logic facts to prepare for automated reasoning, and (4) a logic-based information representation and compliance reasoning schema to represent regulatory and design information for enabling the automated compliance reasoning process. To test the proposed method, a building information model test case was developed based on the Duplex Apartment Project from buildingSMARTalliance of the National Institute of Building Sciences. The test case was checked for compliance with a randomly selected chapter, Chapter 19, of the International Building Code 2009. Comparing to a manually developed gold standard, 87.6% precision and 98.7% recall in noncompliance detection were achieved, on the testing data.

Graduation Semester

2015-12

Type of Resource

text

Permalink

http://hdl.handle.net/2142/89207

Copyright and License Information

Copyright 2015 Jiansong Zhang

Owning Collections