Approaching Pattern Catalogues for Variability Realization in Model-Based Systems and Software Engineering

Abstract

Managing variants, versions, and variability in system-level engineering is becoming increasingly important in industry and represents a challenge for model-based systems engineering. With this paper, we want to raise awareness that guidance and decision support when it comes to realizing variants and variability in SysML and UML models is needed. Although there are already a significant number of approaches to variability realization at source code level, there is a lack of practice-oriented guidance at model level. This creates great uncertainty and ad-hoc approaches when dealing with system changes or parallel modeling of system variants. Due to lack of modularization and standardized variability realization, interwoven, and overly complex system models result. In this paper, we raise awareness of the need for appropriate guides and decision support tools, present our pattern catalogue framework for commonly encountered scenarios in variability realization, discuss feedback from industry on this collection, and show the application of some patterns on an exemplary weather station.