Optimizing Sprint Planning in Agile Methodology Using Greedy Algorithm

Abstract

Sprint planning is a pivotal process in Agile-based software development, where project success heavily depends on the team's ability to select and deliver the most valuable tasks within limited time and resources. A core challenge in this process is determining the optimal set of tasks that can be completed in a sprint, considering the constraints imposed by story point capacity. This decision-making problem closely resembles the classic Knapsack Problem in combinatorial optimization. This paper investigates the implementation of the Greedy algorithm as a heuristic approach to solve this problem by selecting tasks based on their value-to-story-point ratio. The Greedy strategy simplifies task selection by making locally optimal decisions at each step, thereby enabling efficient prioritization of high-value tasks without exceeding the sprint limit. A comparative experiment using real-world data was conducted to evaluate the effectiveness of the Greedy method against manual selection. The results demonstrate that the Greedy algorithm not only utilizes story point capacity more efficiently but also maximizes the total value of tasks included within the sprint. In some scenarios, it even achieved higher priority scores while consuming fewer story points. These findings affirm the practicality of Greedy-based optimization in Agile environments, particularly for rapid and scalable sprint planning. Future work may explore hybrid models or more advanced algorithms such as Dynamic Programming for enhanced optimization outcomes.