Beyond Story Points in Agile software development: A systematic review of alternative effort estimation techniques
Journal of Systems and Software, cilt.244, 2027 (SCI-Expanded, Scopus)
- Yayın Türü: Makale / Derleme
- Cilt numarası: 244
- Basım Tarihi: 2027
- Doi Numarası: 10.1016/j.jss.2026.113075
- Dergi Adı: Journal of Systems and Software
- Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus, ABI/INFORM, Aerospace Database, Applied Science & Technology Source, Compendex, INSPEC, Business Source Ultimate (EBSCO), Engineering Source (EBSCO), Technology Collection (ProQuest)
- Anahtar Kelimeler: Agile effort estimation, Function point, Fuzzy logic, Machine learning, No estimation, Software effort estimation, Story point estimation, Survey, T-Shirt Sizing, Use case point, XAI
- Marmara Üniversitesi Adresli: Evet
Özet
Accurate software effort estimation remains a critical challenge in Agile Software Development (ASD), as traditional Story Point-based approaches often suffer from subjective biases, anchoring effects, and limited inter-team comparability. This study presents a comprehensive systematic literature review (SLR), conducted in accordance with the PRISMA protocol, to identify and evaluate alternative effort estimation techniques proposed between 2020 and 2025. The review categorizes a diverse range of methodologies, including traditional metric-based approaches (Function Point, Use Case Point), expert-based techniques (T-Shirt Sizing), data-driven machine learning models, and emerging natural language processing (NLP) and large language model-based estimation methods. Through a comparative and multi-dimensional taxonomy, these techniques are examined across various agile contexts-such as enterprise systems, distributed teams, and CI/CD environments—-while assessing their performance in terms of accuracy, efficiency, and explainability. The findings indicate a clear trend toward hybrid and explainable (XAI) approaches and context-aware estimation models that integrate automated data-driven insights with human expertise. In addition, the study highlights key methodological gaps, including the scarcity of standardized datasets and the limited availability of robust validation practices in industrial settings. Overall, this research provides a structured reference framework that supports informed decision-making in agile effort estimation and outlines directions for future research in agile effort forecasting.