METHOD OF COMPARATIVE ANALYSIS OF PREVIOUS ESTIMATES AND ACTUAL EFFORTS OF THE TEAM DURING THE DEVELOPMENT OF SOFTWARE PROJECTS IN THE AGILE ENVIRONMENT

Abstract

The object of the study is the issue of increasing the productivity of production processes in outsourcing companies. The subject of the research is strategies for organizing production processes in an Agile environment. The purpose of the article is to improve the efficiency of work stage planning based on a comprehensive assessment of story points. The research task is to develop a method for comparative analysis of software project development histories to support strategies for the productive use of agile technologies. The following results were obtained. A method for comparative analysis of initial estimates and actual efforts of teams during software development in Agile environments was proposed and developed. The method makes it possible, when planning project stages (sprints), to evaluate the completion time and the probability of completing a stage within that time based on task complexity. Additionally, it allows estimating the required time for a planned task based on given complexity and forecast reliability. The proposed method also enables the assessment of risks of failing to complete a stage within the scheduled time. The method was tested on real-world data. The applicability of a heuristic rule for assessing task complexity based on the Fibonacci sequence was confirmed. The probability of completing a task within the planned time, based on the applied time equivalent of a story point, is approximately the same for different complexity levels and significantly less than one. This indicates existing planning risks and suggests the likely presence of a systemic factor reducing development process efficiency. A potential direction for further development of the proposed method is the use of cumulative computation techniques