Hardware Reference
In-Depth Information
because all the data movements that occur in the final application could be traceable
to the conceptual model. This measurement procedure has been automated, providing
the measurement results in a few minutes and using minimal resources [24]. How-
ever, the OOmCFP measurement procedure assumes that the conceptual model has
high quality, that is, the OOmCFP procedure assumes that the conceptual model is
correct, complete, and without defects. Obviously, this is an unreal assumption be-
cause several times the conceptual models present defects. In the following section we
discuss this issue in order to use the measurement procedure to improve the quality of
the conceptual models.
4 Improving the Quality of Conceptual Models Using a FSM
Procedure
In the literature, there is no consensus for the definition of quality of conceptual mod-
els. There are several proposals that use different terminology to refer to the same
concepts. There are also many proposals that do not even define what they mean by
quality of conceptual models. In order to achieve consensus about the definition of
quality of conceptual models and then improve the quality of these kind of models,
we have adopted the definition proposed by Moody [29]. This definition is based on
the definition of quality of a product or service in the ISO 9000 standard [10]. There-
fore, we understand the quality of a conceptual model to be “
The total of features and
characteristics of a conceptual model that bear on its ability to satisfy stated or im-
plied needs
”.
To evaluate the quality of software products, the ISO 9126 standard [13] has been
defined. This standard defines a set of characteristics and sub-characteristics that are
oriented to evaluate the quality of software products from three perspectives: the
internal quality of software products [15], the external quality of software products
[14], and the quality in use of software products [16]. However, since the ISO 9126
standard has been illustrated in the evaluation of the quality of final applications, it is
necessary to select the characteristics, sub-characteristics and metrics that can be
applied to the conceptual models in order to evaluate their quality.
In the last few years, several proposals have emerged to evaluate the quality of
conceptual models based on the ISO 9126 standard: for instance, Genero et al., Li and
Henry, Lorenz and Kidd, Bansiya and Davis, Chidamber and Kemerer, etc. A detailed
description of these proposals can be found in [21]. These proposals focus on the
evaluation of the maintainability of conceptual models [21]. In addition, there are also
proposals that attempt to evaluate the usability of software products in the conceptual
models, for instance, Panach et al. [30], and Abrahao et al. [1]. Despite the great
number of proposals that present metrics to evaluate the internal quality of conceptual
models, none of the proposals has performed an analysis of the defect types that can
be identified in conceptual models, and the conceptual constructs that must be meas-
ured in order to achieve quality characteristics in the conceptual model.
Defect detection refers to found anomalies in software products in order to correct
them and, therefore, obtain software products of better quality. The IEEE 1044 stan-
dard classification for software anomalies [9] define an anomaly as a
ny condition that
deviates from expectations based on requirements specifications, design documents,
