In the context of Software Product Line (SPL) reuse paradigm, one of the most widely used models is the feature model. It describes the set of products in an SPL in terms of their features and the relationships among them. A feature is an externally desired service by the system. While features concern functional aspects, non-functional features must also be considered which have impact on the quality of the SPL derived systems. For this purpose, Extended feature models are proposed in the literature. In order to verify the consistency of such models, several techniques are proposed. We adopt a formal approach that represents Extended Feature Models and reason about them for proving consistency and other operations. In this paper, we show the impact of non-functional attributes on the analysis operations of feature models. In this work, we have resumed analysis operations of feature models listed in the literature. Moreover, we studied the effect of adding the non-functional attributes on these operations by giving examples. So this has enabled us to emphasize the presence of three types of constraints namely: constraint value, constraint attribute-attribute and constraint feature-attribute.
[1]
Douglas C. Schmidt,et al.
Selecting highly optimal architectural feature sets with Filtered Cartesian Flattening
,
2009,
J. Syst. Softw..
[2]
K. Czarnecki,et al.
Cardinality-Based Feature Modeling and Constraints : A Progress Report
,
2005
.
[3]
Sergio Segura,et al.
Automated analysis of feature models 20 years later: A literature review
,
2010,
Inf. Syst..
[4]
Donald D. Cowan,et al.
Efficient compilation techniques for large scale feature models
,
2008,
GPCE '08.
[5]
Thomas Thüm,et al.
Reasoning about edits to feature models
,
2009,
2009 IEEE 31st International Conference on Software Engineering.
[6]
Somnuk Phon-Amnuaisuk,et al.
Knowledge Based Method to Validate Feature Models
,
2008,
SPLC.