Resumo

CAMPOS WRIGHT, Wisleidys  e  TRUJILLO CASANOLA, Yaimí. Improvement of the effort estimation model in projects of the University of Informatics Sciences. Rev cuba cienc informat [online]. 2021, vol.15, n.4, suppl.1, pp. 314-332.  Epub 01-Dez-2021. ISSN 2227-1899.

Planning in the world of Software Industry has become one of the main challenges for project management, and a fundamental activity to develop high quality software. It requires estimates that guarantee the efficiency of the product. The application of good estimation methods is essential for planning because of the increasing influence they have on accurate and predictable control, production processes and software products. In June 2016, the production process of the University of Informatics Sciences was being evaluated by the Software Industry Excellence Center of Mexico in conjunction with specialists from the university. It was certified as Level 2 of the Capability Maturity Model Integration, detecting gaps in the Project Planning area related to effort estimation. In this context, the construction of an estimation model for the effort in software development is proposed, adapted to the different scenarios that the university has, which aims to improve accuracy using an Artificial Neural Network of Hybrid Radial Wavelet type, combining the concepts of a heuristic system of Local Search and the Colony of Bees algorithm. The Delphi method and expert criteria are used to validate the contribution of the proposal. It is concluded that the proposed methodology helps the estimation process of university software development projects.

Palavras-chave : software development; project planning; artificial neural networks; effort estimation; learning algorithm.

        · resumo em Espanhol     · texto em Espanhol     · Espanhol ( pdf )