Smart compliance checking frameworks for BIM standards

Zhu, Xiaofeng 2024. Smart compliance checking frameworks for BIM standards. PhD Thesis, Cardiff University. Item availability restricted.

Preview	PDF - Accepted Post-Print Version Download (15MB) | Preview
	PDF (Cardiff University Electronic Publications Form) - Supplemental Material Restricted to Repository staff only Download (1MB)

Abstract

As a process of generating and managing the digital representation of a built asset, the concept of Building information modelling (BIM) has been gradually accepted and standards for implementing BIM have also been released. Compliance checking for BIM standards plays a significant role in the AEC (Architectural Engineering and Construction) industry regarding ensuring quality, safety, and efficiency during the delivery of projects. As the concept and technology of BIM are accepted and popularised, BIM standards have undergone rapid development in recent years and gradually shaped comprehensive standard systems, which makes fast and accurate BIM compliance checking increasingly challenging. Although considerable effort has been made to address this issue over the past decades, existing checking methods are flawed in terms of granularity, comprehensiveness, and automation. Moreover, these methods suffer from systemic deficiencies that make them incapable of fulfilling the demand of checking compliance against complex BIM standard systems under a rapidly evolving environment. To address the above gaps, a novel framework for BIM compliance checking is developed in this thesis, which is comprised of an ontology-driven subjective checking framework and an evidence-driven objective checking framework. The ontological checking framework adopts a domain ontology as the knowledge model and bridges users and domain knowledge through an interactive web-based service, which enables comprehensive and flexible BIM compliance checking. The evidence-driven checking framework leverages advanced NLP (Natural Language Processing) techniques, large language models, and graph learning to automictically extract information from regulations and project documents and convert it into knowledge graphs, then assess compliance via graph alignment. The main outcomes of the research lie in the development of the ontological checking framework and the automatic checking framework. The ontological checking framework divides domain knowledge and compliance assessment into two separate components enabling flexible checking for multiple scenarios. The adoption of ontology allows the accumulation and integration of domain knowledge. The evidence-driven checking framework has successfully implemented fully automated BIM compliance checking. This research significantly contributes to information extraction, knowledge engineering and BIM compliance checking in the AEC domain. The proposed ontological framework outperforms existing methods in terms of granularity and comprehensiveness and enables flexible compliance checking on various scenarios. As the first attempt, the evidence-driven checking framework fills the gap in fully automated BIM compliance checking. Due to the generic development principles adopted in this research, the proposed method and developed framework can be further extended for other research areas related to information extraction, knowledge extraction and compliance checking.

Item Type:	Thesis (PhD)
Date Type:	Completion
Status:	Unpublished
Schools:	Engineering
Uncontrolled Keywords:	1) BIM 2) Compliance_checking 3) Graph_learning 4) Natural_language_processing 5) Ontology 6) Delphi_Method
Date of First Compliant Deposit:	22 May 2024
Last Modified:	22 May 2024 12:14
URI:	https://orca.cardiff.ac.uk/id/eprint/169050

Actions (repository staff only)

Edit Item