Validating a size measure for effort estimation in model-driven Web development


Abstract:

Web development is moving towards model-driven processes whose goal is the development of Web applications at a higher level of abstraction based on models and model transformations. This brings new opportunities to the Web project manager to make early estimates of the size and the effort required to produce Web applications based on their conceptual models. In the last few years, several studies for size and effort estimation have been performed. However, there are no studies regarding effort estimation in model-driven Web development. In this paper, we present the validation of a model-based size measure (OO-HFP) for Web effort estimation in the context of a model-driven Web development method. The validation is performed by comparing the pbkp_rediction accuracy that OO-HFP provides with the accuracy provided by the standard function point analysis (FPA) method. The results of the study (using industrial data gathered from 31 Web projects) show that the effort estimates obtained for projects that are sized using OO-HFP are more accurate than the effort estimates obtained using the standard FPA method. This suggests that by following a model-driven development approach, the size measure obtained at the conceptual model of a Web application can be considered a suitable pbkp_redictor of effort. © 2010 Elsevier Inc. All rights reserved.

Año de publicación:

2010

Keywords:

  • Model-driven development
  • Web Engineering
  • Effort estimation
  • Functional Size Measurement
  • Measure validation

Fuente:

scopusscopus

Tipo de documento:

Article

Estado:

Acceso restringido

Áreas de conocimiento:

  • Ingeniería de software
  • Software

Áreas temáticas:

  • Programación informática, programas, datos, seguridad
  • Ciencias de la computación
  • Dirección general