Authors:
Yenisei Delgado Verdecia
and
Thelma Elita Colanzi
Affiliation:
State University of Maringa (UEM), Brazil
Keyword(s):
Product Line Architecture, Correlated Metrics, Search-based Design, Evaluation Model.
Related
Ontology
Subjects/Areas/Topics:
Enterprise Information Systems
;
Information Systems Analysis and Specification
;
Software Engineering
;
Software Metrics and Measurement
Abstract:
The Product Line Architecture (PLA) plays a central role at the products development from a Software Product
Line (SPL). PLA design is a people-intensive and non-trivial task. So, PLA design can be considered a
hard problem which could be formulated as an optimization problem with many factors to be solved by search
algorithms. In this sense, the approach named MOA4PLA (Multi-Objective Approach for Product-Line Architecture
Design) was proposed to automatically identify the best alternatives for a PLA design. This approach
originally included metrics to evaluate basic design principles, feature modularization, design elegance and
SPL extensibility. However, there are other relevant properties for PLA design. For this reason, the evaluation
model of MOA4PLA was extended with metrics to measure the level of similarity and adaptability of the PLA.
The objective of this work is to investigate the possible correlation between the metrics related to similarity
and variability in order to d
ecrease the number of functions to be optimized. To do this, three experiments were
carried out. Empirical results allow to learn some lessons regarding to these metrics in the referred context.
(More)