Improving Request for Information (RFI) Processing in Construction Projects using Natural Language Processing (NLP) Techniques

Afzal, Muneeb

Improving Request for Information (RFI) Processing in Construction Projects using Natural Language Processing (NLP) Techniques

Afzal, Muneeb

Permalink

Publication Type:: Thesis
Issue Date:: 2024

Open Access

Copyright Clearance Process

Recently Added
In Progress
Open Access

This item is open access.

Adobe PDF

Download thesisAdobe PDF (4.18 MB)

View statistics

Full metadata record

Field	Value	Language
dc.contributor.author	Afzal, Muneeb
dc.date.accessioned	2025-09-23T01:50:35Z
dc.date.available	2025-09-23T01:50:35Z
dc.date.issued	2024
dc.identifier.uri	http://hdl.handle.net/10453/190035
dc.description	University of Technology Sydney. Faculty of Design, Architecture and Building.	en_US.UTF-8
dc.description.abstract	Request for Information (RFI) documents are essential for clarifying issues throughout the construction project lifecycle, yet their management remains resource intensive. Delays or unresolved RFIs often result in cost overruns and schedule disruptions. Existing research practices rely heavily on manual review, which is time-consuming and error-prone, while industry solutions lack the capacity to generate actionable insights. To address these challenges, this research proposes an automated natural language processing (NLP) pipeline for efficient RFI analysis. The research developed three models: (1) a deep learning-based recurrent neural network model that categorises RFIs by project phase, supplemented with topic modelling for issue visualisation; (2) a multiclass text classification model to identify predominant RFI issue types; and (3) a named entity recognition model to extract key components, locations, and drawing references. These models support early detection of recurring construction issues and generate insights from unstructured RFI statements. Evaluated on real-world construction datasets, the models achieved high precision, recall, F1 scores, and overall accuracy. This research concludes by providing a practical roadmap for implementing NLP-driven RFI management systems, enhancing communication workflows, improving documentation quality, and laying the foundation for data-driven, efficient construction practices.	en_US.UTF-8
dc.format	Thesis (PhD)
dc.language.iso	en_US	en_US.UTF-8
dc.relation	https://opus.lib.uts.edu.au/bitstream/10453/190035/1/thesis.pdf
dc.rights	info:eu-repo/semantics/openAccess
dc.rights	The author owns the copyright in this thesis including all reproduction and reuse rights for the work. The work may not be altered without the permission of the copyright owner. Attribution is essential when quoting or paraphrasing from this thesis.
dc.rights	© 2024 Muneeb Afzal
dc.rights	au.edu.uts.lib/cph
dc.title	Improving Request for Information (RFI) Processing in Construction Projects using Natural Language Processing (NLP) Techniques	en_US.UTF-8
dc.type	Thesis
utslib.copyright.status	open_access	*

Abstract:

Request for Information (RFI) documents are essential for clarifying issues throughout the construction project lifecycle, yet their management remains resource intensive. Delays or unresolved RFIs often result in cost overruns and schedule disruptions. Existing research practices rely heavily on manual review, which is time-consuming and error-prone, while industry solutions lack the capacity to generate actionable insights. To address these challenges, this research proposes an automated natural language processing (NLP) pipeline for efficient RFI analysis. The research developed three models: (1) a deep learning-based recurrent neural network model that categorises RFIs by project phase, supplemented with topic modelling for issue visualisation; (2) a multiclass text classification model to identify predominant RFI issue types; and (3) a named entity recognition model to extract key components, locations, and drawing references. These models support early detection of recurring construction issues and generate insights from unstructured RFI statements. Evaluated on real-world construction datasets, the models achieved high precision, recall, F1 scores, and overall accuracy. This research concludes by providing a practical roadmap for implementing NLP-driven RFI management systems, enhancing communication workflows, improving documentation quality, and laying the foundation for data-driven, efficient construction practices.

Please use this identifier to cite or link to this item:

http://hdl.handle.net/10453/190035