Extração, síntese e avaliação de modelos de features em linhas de produto de software

Abstract

Since the advent of software product lines, they have proved a successful strategy for building products with related features. Over the years, the way of representing characteristics common and variables of a family of products using a feature model has become increasingly popular. Feature may be defined as a logical unit of behavior specified by a set of requirements that is relevant to any of the stakeholders. Although a model of features may help manage a SPL, this production requires a lot of time and effort. To reduce this time and effort in creating feature models were developed some reverse engineering process that is used for synthesizing a feature model. The construction of models from the process of reverse engineering is performed using a set of configurations and dependencies containing a resource hierarchy. From this, the reverse engineering process defines a set of execution steps to extract one or more models. The objective of this work is to define an algorithm to be implemented in DyMMer tool to perform the construction of various features models with different structures, using a reverse engineering process from a set of product configurations is performed the synthesis of various features models. To accomplish this evaluation was used measures proposed in the literature to be applied directly in the structure of feature models. The methodology was divided into three parts: (i) research and study of the tools available to perform evaluation feature models, (ii) collect measures, and (iii) evaluation of synthesized models resulting from the elaboration which has as input a set of product configurations. As a result of this work, it was possible to define a set of steps for the feature models extraction from a set of product configurations using reverse engineering and a way to evaluate a model structure without the knowledge domain.