Projekta realizācijas laiks:
Formal methods and industry standards have become the primary approach for ensuring quality in software development processes, where balancing development speed and quality remains a challenge. A successful project is based on a well-defined project management plan and scope, outlining objectives, timelines, budget, risk management, and quality assurance. Existing automation tools can plan and track tasks, identify risks, and support project documentation; however, challenges still persist, such as lack of contextual understanding, limited adaptability, and the need for human involvement, especially during the project initiation phase.
The aim of this project is to develop an approach for IT project management plan and scope generation using model-driven engineering concepts and artificial intelligence algorithms. The resulting solution defines the core sections of project documentation, which are populated with structured and context-aware textual content based on best practices in IT project management and software engineering standards. The approach emphasizes precise input data about the problem domain and transformation rules to ensure quality without compromising development speed.
The developed approach is supported by its tool prototype MAPS-AI (Model-Driven AI-Assisted IT Project Planning), which combines model-driven transformations with the capabilities of generative artificial intelligence. MAPS-AI transforms high-level conceptual models, such as business process models and project parameters, into structured project artefacts, including project management plans, scope descriptions, user stories, and product backlogs. The system ensures traceability between initial inputs and generated outputs, while leveraging advanced (meta-) prompt engineering techniques to produce contextually accurate and human-readable documentation.
MAPS-AI can reduce manual effort, improve documentation quality, and accelerate the project initiation process, while ensuring alignment with industry standards and best practices. The tool enables project teams to make more informed decisions, improve communication with stakeholders, and structure early project phases more effectively.
The solution has been developed with the support of the European Union Recovery and Resilience Facility under the research and development grant No. PA-2024/1-0015 “Model-Based Methodology for IT Project Management Plan and Scope Development Using Artificial Intelligence Algorithms,” within project No. 5.2.1.1.i.0/2/24/I/CFLA/003 “Implementation of consolidation and management changes at Riga Technical University, Liepaja University, Rezekne Academy of Technology, Latvian Maritime Academy, and Liepaja Maritime College towards excellence in higher education, science, and innovation.”
Project results are published:
Model Transformations Used in IT Project Initial Phases: Systematic Literature Review
Key Artefacts in the Initial Phases of IT Project Management: Systematic Mapping Study
Generation of IT Project Documentation Elements from a Model Transformation Chain
AI-Assisted Transformation of the Two-Hemisphere Model into Structured IT Project Backlog
MAPS-AI - A Tool for AI-Assisted Model-Driven Generation of IT Project Plan and Scope
Project follow-up publications:
Machine Learning-Based Story Point Estimation Using User Story Text within a BPMN-Driven Planning Context (in press)
MAPS-AI a Model-Driven AI-Assisted Tool for IT Project Plan and Scope: Case Study and Evaluation (in press)
MAPS-AI: Model-Based Approach for IT Project Scope Using Artificial Intelligence (in press)
