A Proposed Estimation Method Framework in Scrum Development Methodology for Tech Startups رسالة ماجستير

Abstract

Tech startups often face challenges in accurately estimating the number of backlog items that can be completed during a sprint, using the scrum agile development methodology. This can result in a negative impact on the relationship with customers, including costs and trust. One major factor that affects the delivery of features is the presence of defects or bugs, which can prevent backlog items from being marked as complete and shipped to the production environment. To address this problem, we propose a new estimation methodology that combines testing and development estimates to account for the impact of bugs on the completion of features. Our approach is divided into two parts: a quantitative analysis to evaluate the model's effectiveness, and an applied implementation in a real startup development cycle. In the quantitative analysis, we assess the model's quality using various metrics. In the applied implementation, we test the model in the actual development process of a startup, using the results of the quantitative analysis to guide the approach. The results of our study show that our method significantly improved the done percentage of sprints, going from 84.2% to 95.6%. This demonstrates a significant improvement compared to the previous results without using our approach. Our method offers a potential solution for tech startups facing difficulties with backlog item estimation and aims to improve the accuracy and efficiency of the development process. However, it is important to note that if the approach is not used correctly, it can result in a gap in development capacity and a deviation from agile best practice.