Optimizing Effort Estimation in Agile Software Development: Traditional vs. Advanced ML Methods

Effort estimation is a fundamental procedure that involves forecasting the required time, resources, and effort needed to successfully accomplish a project, assignment, or activity. A major challenge for the software industry is accurate effort estimation. In recent years, the software industry has moved towards Agile Methodology since its development in 2001. The Agile methodology emphasizes iterative and incremental delivery, which allows the software to be delivered in smaller increments and provides greater flexibility in response to changes in requirements. Within an agile framework, effort estimation encompasses both traditional and advanced algorithmic methods. This study conducts a comparative analysis, evaluating traditional and advanced machine learning techniques for effort estimation. By analyzing several previous studies, the research identifies the most suitable techniques to achieve accurate estimations. Traditional effort estimation techniques often fall short in accurately predicting the efforts needed for a software project due to data limitations, lack of detail, neglect of complexity, unaccounted risks, and an inability to adapt to changes. The limitations of conventional approaches can be overcome by machine learning algorithms that can be trained on past data to identify hidden patterns and relationships and produce more accurate and efficient effort predictions.