Software Project Risk Identification Based on Scrum Artifacts

Abstract

Risk identification, as a foundational step in risk management, is essential for the success of software development projects, including those using the Scrum methodology. However, risk management in Scrum often lacks structured practices, relying instead on the iterative and collaborative nature of the methodology. This paper presents a novel risk identification framework tailored for Scrum projects, leveraging Scrum artifacts and their associated attributes to enhance risk detection and analysis. The framework encompasses automated data collection from project management environments, such as Jira, thus minimizing the effort required from the project team for risk-related data gathering. By monitoring deviations in artifact attributes and applying predefined thresholds, the framework facilitates the detection of risks at any point in the project lifecycle. Experimental evaluation across three distinct Scrum projects demonstrated the framework’s applicability in identifying a wide range of risks, including long-term ones. The proposed framework contributes to Agile project management by offering a structured, automated approach to risk identification, fostering a culture of proactive risk management within Scrum teams.