Eureka! Institutional Repository

The effort estimation in the implementation of a software project, is a difficult process because of the uncertainty concerning the variable functional requirements of the project. However, this process is very important and often determines the success or failure of a project. If the prediction is not accurate, then the cost of the software project development grows. Over the last 15 years, most of the large software companies have adopted agile methodologies for the implementation of a software project. Agile methods are a subset of iterative and evolutionary methodologies based on improvement and optimization of the development processes. Each iteration, is a small independent project, with activities spanning requirements analysis, design, implementation, testing, and customer acceptance. Upon completion of each iteration grows and evolves a subset of the final system. This work presents the agile methodologies and more specifically the Scrum methodology. Then are presented four classic widespread effort assessment methodologies in software projects, methods COCOMO II, Wideband Delphi, Bayesian Belief Networks and CoBRA. Also, a higher effort estimation methodology is presented, the Planning Poker process, which is used in the context of agile methodologies. Finally, a comparison of this methodology to the previous four classic effort assessment methodologies.