A hybrid effort estimation technique for agile software development (HEETAD)

Abstract

Software development becomes a complex process when the software grows in size or complexity making it difficult to estimate usage of resources or development costs. Software effort estimation is that part of development which helps in assessing resource prior to development. An estimate is a quantified evaluation of the effort necessary to carry out a given development task and most often expressed in terms of durations. Effort estimation is done with an intent to aggregate individual estimates and obtain the overall duration, effort or cost of a software project. The workforce is measured as effort and the total time required is defined for a task in effort estimations which is usually expressed in units (Man-day, Man-month, and Man-year). Most other factors like cost or total time required to developed software are dependent on these estimations. Further, Algorithms used for estimating software developments efforts, may also be imprecise. Thus, Effort estimations plays an important part of software development in planning and monitoring projects. Agile methodology is relatively a new set of practices in software development. Agile estimations are based on many factors. Improperly recorded information from Agile methods can result in erratic estimations thus creating an impending need for precise effort estimations. It is difficult to find a single technique which can suit all conditions. Hence, this paper attempts to estimate agile development efforts by using a hybrid technique based on function points and user stories. Results of the proposed technique demonstrate that the arrived effort estimations based on user stories are efficient.