DOI Services Persistent Identifiers for Scholarly Research
You are viewing the preserved copy. Open the article on the journal's website →

Qualitative Research Journal For Social Studies • pp. 23-45

INTEGRATION OF ARTIFICIAL INTELLIGENCE, AGILE METHODOLOGY, AND SOFTWARE ARCHITECTURE FOR IMPROVING SOFTWARE PROJECT SUCCESS

  • Hashma Jawad Department of Software Project Management ,Superior University, Lahore,Punjab, Pakistan Corresponding Author Email: SU92-MSPMW-S26-001@SUPERIOR.EDU.PK
  • Maham Amin Department of Software Engineering,Superior University, Lahore,Punjab, Pakistan Corresponding Author Email: SU92-MSSEW-S26-001@SUPERIOR.EDU.PK
  • Muhammad Waseem Iqbal Department of Software Engineering ,Superior University, Lahore,Punjab, Pakistan Supervisor Email: waseem.iqbal@superior.edu.pk
  • Saleem Zubair Department of Computer Science and Information Technology,Superior University, Lahore Punjab, Pakistan Supervisor Email : saleem.zubair@superior.edu.pk
  • Ayesha Saddiqa Department of Computer Science and Information Technology,Superior University, Lahore Punjab, Pakistan Supervisor Email : ayeshasaddiqa@superior.edu.pk

Abstract

The landscape of software engineering is evolving dramatically as a result of the merging of Artificial Intelligence

(AI), Agile methodologies, and Software Architecture to meet the needs in today's world such as information

overload and "guesstimation". This is especially significant in places such as Pakistan, where the IT industry is a

worldwide freelancer leader and has a 40% failure rate of projects. The study highlights key findings, such as

having Generative AI (GenAI) help with sprints can boost productivity by up to 88% and the AI-Based

Architecture Decision Support System (AIDSS) gives architects 100% accuracy on the best tech options for the

job. Moreover, the AIDSS framework is known to increase the efficiency of documentation as much as 40% of

time and has been reported to achieve up to a 95% success rate for expert-validated decision making in prior

literature, with this study's own experiment achieving 100% compliance on a smaller test set. This study aims to

look into the possibility of introducing AI-based Decision Support Systems (DSS) and machine learning

algorithms (Random Forest Classifiers and Linear Regression) to take project estimation from subjective "best

guesses" to the successful "big data facts". The study suggests a strategy called "Agile Architecture" to reduce

"architectural smells" and the “hit the wall” effect by having AI agents regularly check on architectural "smells".

The study concludes that the "Human-AI Hybrid" approach should be adopted, with AI serving as a partner to

maintain ethical standards and human creativity, while also complying with international norms such as IEEE

1471.

Record

Identifier
https://doi-services.com/1105/1739
DOI
10.63878/bx38p858
Published
2026-03-02
Version
1
Original
https://studentsconsultancyhome.com/qrj-social/index.php/qrj/article/view/1739