Estimation Trend in Agile Software Development

Estimating software development in Agile often creates more anguish than meaning. The disparity between actuality and estimation has compelled the development team to re-design various strategies, models, and methodologies for estimating software projects. Previous research have discovered techniques for estimation in Agile, including Planning Poker and Expert Judgment. The estimation addressed is confined solely to the realms of effort estimation and cost estimation. Indeed, Agile encompasses multiple cycles that can be estimated. Prior studies merely identified procedures and data suitable for estimation, without evaluating those that demonstrate significant effectiveness when utilized collaboratively. The application of appropriate approaches and data will dictate the efficacy of estimation within Agile Software Development. This article seeks to explore probable estimating trends in Agile Software Development in recent years. A review will be conducted on the selection of appropriate procedures and data for each estimation trend. This study employed a systematic literature review for analysis and reporting. The evaluation results indicate that the predominant trend in Agile estimating is user story estimation, accounting for $\mathbf{5 6 \%}$. This method is more effective when employing deep learning and machine learning with user story data, as opposed to use planning poker or case studies. This paper summarizes further estimates alongside suitable methodology and data.