Enhancing decision-making in software development :  an automated system for software requirements change impact analysis

Zamani, Kareshna

Enhancing decision-making in software development : an automated system for software requirements change impact analysis

Zamani, Kareshna

Permalink

Publication Type:: Thesis
Issue Date:: 2025

Open Access

Copyright Clearance Process

Recently Added
In Progress
Open Access

This item is open access.

Adobe PDF

Download thesisAdobe PDF (4.13 MB)

View statistics

Full metadata record

Field	Value	Language
dc.contributor.author	Zamani, Kareshna
dc.date.accessioned	2026-01-05T01:58:49Z
dc.date.available	2026-01-05T01:58:49Z
dc.date.issued	2025
dc.identifier.uri	http://hdl.handle.net/10453/191197
dc.description	University of Technology Sydney. Faculty of Engineering and Information Technology.	en_US.UTF-8
dc.description.abstract	Change Impact Analysis (CIA) is a critical task in software requirements engineering, aiming to predict the effects of requirement changes on related artifacts and systems. Traditional CIA methods often rely on manual inspection and heuristic-based reasoning, which are time-consuming and error-prone. This research addresses these limitations by proposing an automated framework for Software Requirements Change Impact Analysis (SRCIA), leveraging advances in Machine Learning (ML), Natural Language Processing (NLP), and Artificial Intelligence (AI). The framework integrates a range of approaches, including traditional ML models, NLP-based techniques, BEIR-based retrieval methods, and a Retrieval-Augmented Generation (RAG) system, to assess their effectiveness across multiple datasets of varying complexity. Evaluation metrics such as precision, recall, F1 score, BLEU, and ROUGE are used to benchmark performance. A central contribution is the development of a RAG-based solution that combines Large Language Models (LLMs) with modern information retrieval techniques. By incorporating vector database tools like LanceDB and FAISS, along with prompt engineering strategies, the framework achieves accurate and context-aware impact predictions. This enables robust adaptation to real-world, unstructured, and evolving requirements. The research provides a practical, scalable, and extensible solution to support automated CIA in complex software projects.	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/10453/191197/1/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	© 2025 Kareshna Zamani
dc.rights	au.edu.uts.lib/cph
dc.title	Enhancing decision-making in software development : an automated system for software requirements change impact analysis	en_US.UTF-8
dc.type	Thesis
utslib.copyright.status	open_access	*

Abstract:

Change Impact Analysis (CIA) is a critical task in software requirements engineering, aiming to predict the effects of requirement changes on related artifacts and systems. Traditional CIA methods often rely on manual inspection and heuristic-based reasoning, which are time-consuming and error-prone. This research addresses these limitations by proposing an automated framework for Software Requirements Change Impact Analysis (SRCIA), leveraging advances in Machine Learning (ML), Natural Language Processing (NLP), and Artificial Intelligence (AI). The framework integrates a range of approaches, including traditional ML models, NLP-based techniques, BEIR-based retrieval methods, and a Retrieval-Augmented Generation (RAG) system, to assess their effectiveness across multiple datasets of varying complexity. Evaluation metrics such as precision, recall, F1 score, BLEU, and ROUGE are used to benchmark performance. A central contribution is the development of a RAG-based solution that combines Large Language Models (LLMs) with modern information retrieval techniques. By incorporating vector database tools like LanceDB and FAISS, along with prompt engineering strategies, the framework achieves accurate and context-aware impact predictions. This enables robust adaptation to real-world, unstructured, and evolving requirements. The research provides a practical, scalable, and extensible solution to support automated CIA in complex software projects.

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

http://hdl.handle.net/10453/191197