Leveraging IFC-Based Locational Information for Work Package Generation to Support Analysis of Robot-Assisted Construction.
Saved in:
| Title: | Leveraging IFC-Based Locational Information for Work Package Generation to Support Analysis of Robot-Assisted Construction. |
|---|---|
| Authors: | Wong Chong, Oscar1 (AUTHOR) oscar.wong@utp.ac.pa, Zhang, Jiansong2 (AUTHOR) zhan3062@purdue.edu |
| Source: | Journal of Computing in Civil Engineering. May2026, Vol. 40 Issue 3, p1-20. 20p. |
| Subjects: | Building information modeling, Logic programming, Geographical positions, Construction planning, Industrialized building |
| Abstract: | The current labor shortage is adversely impacting the construction industry. Automation and artificial intelligence (AI) have the potential to transform the architecture, engineering, and construction (AEC) industry into a smarter, safer, and more efficient industry while relieving the effects of the labor shortage problem. Using AI as backbone, the purpose of this research is to introduce and leverage logic-based AI in supporting building information modeling (BIM) data extraction and information inference more efficiently. To achieve this, a logic-based method was developed for automated processing of locational information and inference of construction-ready information (namely, submodule packages) from industry foundation classes (IFC) to support the simulation and analysis of construction processes using robotics. State-of-the-art invariant signatures were applied to provide uniformity and facilitate the reuse of data from building information models. Tests were conducted on four BIM instance models (a home module, a duplex apartment, a warehouse, and an office building). Results showed great potential in the extraction of IFC data and the generation of the construction level information by using logic representation and reasoning. In the extraction of locational information, the proposed method achieved 84.6% recall and 66.3% precision overall on the four models in the first run, for a F1 measure of 74.3%. Following a systematic error analysis and iterative refinement of the logic ruleset over seven iterations, the proposed method achieved 100% precision and recall in the extraction of all target objects from a completely unseen model (residential building) during training. Furthermore, the proposed method successfully generated the corresponding wall submodule packages for the home module model, with a time efficiency of about 67 times faster compared to a procedural programming approach in processing locational information. [ABSTRACT FROM AUTHOR] |
| Copyright of Journal of Computing in Civil Engineering is the property of American Society of Civil Engineers and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.) | |
| Database: | Engineering Source |
Be the first to leave a comment!
