Data Mapping from Building Information Model into Brick Ontology
Loading...
URL
Journal Title
Journal ISSN
Volume Title
Sähkötekniikan korkeakoulu |
Master's thesis
Unless otherwise stated, all rights belong to the author. You may download, display and print this publication for Your own personal use. Commercial use is prohibited.
Authors
Date
2024-05-29
Department
Major/Subject
Control, robotics and autonomous systems
Mcode
ELEC3025
Degree programme
AEE - Master’s Programme in Automation and Electrical Engineering (TS2013)
Language
en
Pages
62
Series
Abstract
This thesis addresses the imminent challenges and opportunities presented by the new Finnish Building Act of 2025, focusing on the possibility of integrating Building Information Modelling (BIM) data into Brick Smart Ontologies. This integration is crucial for advancing facility maintenance practices. Therefore, this thesis emphasises the importance of utilising BIM data more effectively, as well as meeting the requirements of the upcoming legislation. By exploring the potential of Natural Language Processing (NLP) for automating the data mapping processes, the study aims to bridge the gap between current BIM usage and facility management protocols. The research methodology adopted in this thesis involves a throughout literature review to establish the theoretical foundation, followed by case study analyses to understand the practical implications and challenges of BIM-Brick integration. An experimental design employing the Universal Sentence Encoder (USE) NLP model to automate the data mapping process between the YTV2020 BIM datasets and the Brick Class ontology dataset comprises the core of the thesis's practical application. This approach not only aims to enhance the accessibility of BIM data for various smart building applications but also to demonstrate the feasibility and efficiency of such integration in real-world scenarios. The study's results reveal challenges in the accuracy and practical utility of using NLP models for BIM data and Brick Smart Ontologies integration. Although the NLP model demonstrated capability in matching BIM data to Brick classes, the matches were not consistently accurate, missing more suitable classifications that were not detected by the model. This necessitates human intervention for verification and highlighting limitations in the current approach. Furthermore, the matched data proved less beneficial for facility maintenance purposes than anticipated. These findings underscore the need for further refinement in the application of NLP models within this context, suggesting that future research should focus on improving model accuracy and the utility of matched data for building maintenance. Despite these challenges, the research provides a foundation for an automated data matching system for the BIM-Brick integration process.Description
Supervisor
Ihasalo, HeikkiThesis advisor
Mikala, JuliusKeywords
building information modelling, facility management, natural language processing, universal sentence encoder