FEASIBILITY OF ANN IN AGILE METHODOLOGIES FOR EFFORT ESTIMATION

Abstract

Software development effort estimation is the process of predicting the most realistic use of effort required to develop or maintain software based on incomplete, uncertain and/or noisy input. Poor estimation may be the cause of significant challenges in project management and in the software quality. Therefore, it is important to make the use of estimation models and appropriate techniques to avoid losses caused by poor estimation. Software effort estimation has been an important and difficult task since the evolution of the software. Many formal and informal methods have been proposed for software estimation. It is important for estimation methods to generate realistic software estimates to build the trust of customers as well as team members. Unrealistic estimates are major factors for either software project failure or decreasing the quality of the software. Agile software processes try to minimize the impact of insufficient estimation accuracy by ensuring that the most important functionality is developed first. This is achieved through a flexible development process with short iterations. However, there is still a need for accurate estimates, as these are the basis for staffing, planning, prioritization and contract negotiations. To address the effort estimation related issues in case of agile software development, I will be proposing a new way to estimate effort required to develop software which is going to be built by agile methodologies. This new way incorporates categorization of various important project characteristics into 10 metrics (Team, Complexity, performance etc) which can be helpful to increase the accuracy of effort estimation by the use of artificial neural networks.