Abstract:
Agile project management prioritizes adoptability and collaboration, continues to face significant challenges in risk management, dynamic requirements, resource allocation, and project delays. A major concern is the inconsistency in meeting project deadlines, which results in inefficiencies and cost overruns. This research proposes an AI-driven theoretical framework for time-series forecasting, specifically tailored for JIRA, aiming to mitigate delays and enhance project predictability. The study employs a structured approach, commencing with data collection through surveys directed at project managers, team leaders, developers, QA engineers, business analysts, and UI/UX designers to identify critical risks inherent in agile environments. A comprehensive risk analysis is conducted to evaluate prevalent project challenges, thereby guiding the selection of an appropriate AI model for forecasting. The ARIMA model is developed and integrated with the JIRA API, which facilitates automated risk predictions and supports proactive decision-making. The results derived from this integration contribute to the refinement of the proposed theoretical framework, which is further validated by experts in the field. By utilizing predictive analytics, particularly through the ARIMA model, the framework analyses historical project data, forecasts delays in sprints, identifies resource bottlenecks, and anticipates trends in task completion. These AI-powered insights enable project managers to proactively manage risks, enhancing both sprint planning and resource distribution. Additionally, the study evaluates the framework’s applicability beyond JIRA, enhancing its relevance to a wider array of agile tools. Findings indicate that AI-driven forecasting enhances the predictability of sprints, reduces bottlenecks, and improves decision-making processes within agile methodologies. This research contributes to the evolution of agile practices by offering a scalable, AI-driven approach to risk mitigation, adaptable to various project management contexts. The proposed framework seamlessly integrates traditional agile methodologies with AI-driven automation, offering practical recommendations for incorporating AI-based forecasting into modern software development.
