With the rise of software-defined vehicles (SDV) the automotive industry is undergoing a paradigm shift. Not only do SDV provide new or enhanced user experiences, but they also require novel approaches in vehicle development. Vehicles can no longer be understood as isolated systems; they must be seen as integrated entities in a system of systems (SoS). In particular, functions that are realized by software have control flows that can be distributed across the entire SoS and can be implemented in alternative ways. In the development of vehicles this provides an increased degree of architectural freedom by using alternative services from the SoS. However, this is also accompanied by an increase in complexity.

The aim of this work is to present an approach for capturing and leveraging architectural alternatives for functionalities of vehicles. An SoS engineering meta model is being utilized to describe the composition of vehicle functions across the SoS. Data structured by this meta model is taken from a reference architecture of a vehicle and enhanced by the incorporation of SoS data. This database is being parsed into a graph structure to leverage graph science capabilities.

Consequently, path finding algorithms are employed to identify and assess alternative concepts for reference functions from a holistic perspective. These path finding algorithms identify services and their ports or so-called interfaces within functions to transfer information and control flows in different ways considering multiple variants and criteria. Finally, a prospective examination of additional applications of this approach, particularly with respect to recursive macro-level analysis is presented.