Preserving Consistency of Interrelated Models during View-Based Evolution of Variable Systems

Coping with different and changing requirements leads to concurrent products (variability in space) and subsequent revisions (variability in time). Moreover, products consist of interrelated models that represent different kinds of artifacts. Dependencies and redundancies between interrelated models within a product and across products can quickly lead to inconsistencies during evolution. Thus, dealing with both variability dimensions uniformly while preserving consistency of interrelated models is a major challenge when developing large and long-living variable systems. Recent research addresses uniform management of variability in space and time by unifying concepts and operations from software product line engineering and software configuration management. However, consistency preservation for interrelated models, which is a major research topic in model-driven software development, has hardly been considered in variability management. We propose an approach that builds on recent efforts for unifying variability in space and time and leverages view-based consistency preservation for systems comprised of different kinds of interrelated models. We evaluate our approach by applying it to two real-world case studies: the well-known ArgoUML-SPL, that is based on the UML modeling tool ArgoUML, and MobileMedia, a mobile application for media management. Our results show that, by manually evolving only the Java models of products, other interrelated models (i.e.,~UML class diagrams) and the remaining affected products can be kept consistent fully automatically.