Enhancing effort and estimation in scrum-based agile projects with a proposed federated agile framework

• Published in: International journal of information technology (Singapore. Online) Vol. 17; no. 5; pp. 2983 - 2995
• Main Authors: Chakravorty, Geetanjali, Reddy, B. Ramachandra, khan, Danish Ali
• Format: Journal Article
• Language: English
• Published: Singapore Springer Nature Singapore 01.06.2025; Springer Nature B.V
• Subjects: Accuracy; Algorithms; Artificial Intelligence; Collaboration; Composition; Computer Imaging; Computer Science; Estimates; Feature selection; Fuzzy logic; Hybrid modes; Image Processing and Computer Vision; Literature reviews; Machine Learning; Mathematical analysis; Mathematical models; Neural networks; Optimization; Original Research; Pattern Recognition and Graphics; Software; Software development; Software Engineering; Teams; Vision; Work at home; Working conditions; Federated framework; Agile effort estimation; Business continuity planning; Remote work flexibility; Hybrid working; Skill composition
• ISSN: 2511-2104; 2511-2112
• DOI: 10.1007/s41870-025-02451-x

The successful completion of any software project cruces on meeting committed timelines but ensuring complete scope coverage along with accurate cost budget estimation. Several projects just fail due to cost overruns, often because of inaccuracies associated with estimated effort. Software effort estimation (SEE) is, therefore, critical throughout the entire software development lifecycle, hence accurate effort estimation (EE) has become an essential parameter for Agile projects. There is numerous traditional estimation models are present, but they are inadequate for Agile projects keeping nature of rapid changes in view. In other side Planning Poker and T shirt sizing are Agile estimation techniques which are widely used but stands with inconsistency due to their approach of subjective judgments. This study represents a federated framework incorporating Agile-specific critical success factors, as remote work flexibility or hybrid mode of operation and team skill composition including Senior and junior mix and plan for business continuity to enhance estimation accuracy on top of the best judgement. The main objective of this specific study is to identify key factors influencing Agile effort estimation and develop a mathematical model suitable for Agile projects irrespective of agile techniques used like scrum, kanban or lean etc., and lift-up the level of accuracy across diverse Agile projects. This article has considered four diversified projects of different sizes and complexities and divided them into user stories to start with the effort estimation associated with. Inputs included user story breakdown, screen count and size, API count for back-end integration, complexity, code reusability, hybrid work environment, and team composition. A mathematical model integrated these factors, adjusting for variables like remote work and Business Continuity Planning (BCP) to provide refined effort estimates. The findings concludes that effort estimation is more accurate when adjusting for remote work, skill mix, and BCP. In projects with minimal remote work (< 20%), effort estimates increased significantly when BCP and team skill mix adjustments were included, while hybrid models showed more stable estimates towards accuracy. Hence the federated framework enable the critical factors which are used to calibrate the overall effort estimation accuracy in agile projects.