Improving quality of software product line by analysing inconsistencies in feature models using an ontological rule-based approach

Abstract

In software product line engineering, feature models (FMs) represent the variability and commonality of a family of software products. The development of FMs may introduce inaccurate feature relationships. These relationships may cause various types of defects such as inconsistencies, which deteriorate the quality of software products. Several researchers have worked on the identification of defects due to inconsistency in FMs, but only a few of them have explained their causes. In this paper, FM is transformed to predicate-based feature model ontology using Prolog. Further, first-order logic is employed for defining rules to identify defects due to inconsistency, the explanations for their causes, and suggestions for their corrections. The proposed approach is explained using an FM available in Software Product Line Online Tools repository. It is validated using 26 FMs of discrete sizes up to 5,543 features, generated using the FeatureIDE tool and real-world FMs. Results indicate that the proposed methodology is effective, accurate, and scalable and improves software product line. Copyright © 2017 John Wiley & Sons, Ltd