AI-Assisted Transformation of the Two-Hemisphere Model into Structured IT Project Backlog

Proceedings of the 66th International Scientific Conference on Information Technology and Management Science of Riga Technical University (ITMS'2025) 2025
Oksana Ņikiforova, Rihards Bobkovs, Oscar Pastor Lopez

This paper presents an AI-assisted approach for transforming the two-hemisphere model of IT project initiation into structured product backlog with correspondence to Agile software development. The proposed solution enables the automatic generation of IT project scope documentation and task lists based on user stories extracted from the problem domain process model, as well as create the problem domain UML use case diagram from the two-hemisphere model. By analyzing the application of the Agile SCRUM methodology to task organization, the research defines transformation rules for converting process elements into backlog items. Large Language Model is integrated into the solution to automate task generation, prioritization, and decomposition. The resulting data can be seamlessly imported into project management tools such as Jira and Leantime. Emphasis is placed on structured data representation, automated project backlog generation and tool integration.

Atslēgas vārdi
model-driven engineering, business process modeling, IT project management, agile backlog generation, large language models, process automation, requirements engineering
DOI
10.1109/ITMS67030.2025.11236542
Hipersaite
https://ieeexplore.ieee.org/document/11236542

Ņikiforova, O., Bobkovs, R., Pastor Lopez, O. AI-Assisted Transformation of the Two-Hemisphere Model into Structured IT Project Backlog. No: Proceedings of the 66th International Scientific Conference on Information Technology and Management Science of Riga Technical University (ITMS'2025), Latvija, Rīga, 9.-10. oktobris, 2025. ieeexplore.ieee.org: ieee, 2025, 1.-7.lpp. Pieejams: doi:10.1109/ITMS67030.2025.11236542

Publikācijas valoda
English (en)