Model-based Variant Management in Automotive Systems Engineering
Stefan Otten, Thomas Glock, Carl Philipp Hohl, Eric Sax · 2019
The upcoming trends of autonomous driving and electrification in the automotive sector lead to an increasing amount of electronic and software. Highly interconnected and cross-domain functionalities with sophisticated algorithms pose a significant challenge on the automotive systems engineering process, which needs to cope with the added complexity. Managing and handling the variants of different Electric/Electronic (E/E) systems throughout the entire development is one of the key factors to achieve traceability from requirements phase until production. In this paper, we propose an approach for variant management of E/E-systems within a model-based engineering process. The presented hybrid characteristics model combines different variant management concepts such as the feature- oriented domain analysis (FODA) and orthogonal variability modeling. It is divided into three submodels and allows the model-based expression of variants for all perspectives within concept and development of E/E systems, reaching from functional aspects of the system to generic product characteristics. Therefore, it enhances a consistent and formalized description of variants throughout the entire development process and supports the configuration of product variants in an early systems engineering phase. Additionally, configuration of derivatives and reuse of early system and conceptual design decision and models is enhanced. The approach was integrated within an E/E architecture development tool and is evaluated in productive use. A showcase for an exemplary system is demonstrated.