Hardware Reference
In-Depth Information
Applying a Functional Size Measurement Procedure for
Defect Detection in MDD Environments
Beatriz Marín, Giovanni Giachetti, and Oscar Pastor
Centro de Investigación en Métodos de Producción de Software,
Universidad Politécnica de Valencia,
Camino de Vera s/n, 46022 Valencia, Spain
{bmarin,ggiachetti,opastor}@pros.upv.es
Abstract.
Nowadays, is widely accepted that functional size measurement is
essential to manage and control software projects. In order to obtain early indi-
cators for software projects, many functional size measurement procedures have
been developed to measure the functional size of conceptual models. To do this,
the measurement procedures assume that models do not present defects. How-
ever, this is an unreal assumption because, in practice, the conceptual models
can have defects that may affect the implementation of final applications. This
is especially important for software production processes based on MDD tech-
nology, where the conceptual models are key artifacts used as inputs in the
process of code generation. Therefore, this paper presents how a functional size
measurement procedure (which has been developed for the measurement of
conceptual models of a specific MDD environment) can help in the detection of
defects in conceptual models.
Keywords:
Conceptual Model, Functional Size, Measurement Procedure,
COSMIC, Model-Driven Development, Defect Detection.
1 Introduction
During the last few years, software production processes have evolved from the solu-
tion space (software product) to the problem space (conceptual models). Models are
abstractions of the reality that help to understand complex problems and their poten-
tial solutions [34]. Thus, Model-Driven Development (MDD) methods have been
emerged to take advantage of the benefits of the use of models, such as a simplified
view of the problem (using concepts that are much less bound to the underlying im-
plementation technology and are much closer to the problem domain); and an easy
way to specify, understand, and maintain the systems.
In a software production process based on MDD technology, the conceptual mod-
els are key artifacts that are used as input in the process of code generation. These
conceptual models must provide a holistic view of all the components of the final
application (including the structure of the system, the behavior, the interaction be-
tween the users and the system, etc.) in order to be able to automatically generate the
final application. To do this, the models (conceptual models) must have enough se-
mantic formalization to specify all the functionality of the final application and also to
