Due to the rising complexity in the automotive domain a suitable modeling framework to facilitate development as well as interdisciplinary communication is needed. In the domain of smart grids, the Smart Grid Architecture Model (SGAM) framework turned out to be a suitable tool for coping with complexity whilst providing cross-disciplinary understanding. A similar approach exists for the automotive domain: the Automotive Reference Architecture Model (ARAM) framework. As opposed to other automotive frameworks, this one offers interoperability with the smart grid framework, which simplifies cross-domain collaboration. During the development of ARAM, the Software Platform Embedded Systems (SPES) methodology gained momentum. This methodology can be considered an architectural model while the automotive framework can be considered an analysis model, according to the domain-specific systems engineering approach. In order to follow this approach, a transformation between the two models must be possible. This work-in-progress paper aims at presenting first approaches of an adapted version of the ARAM framework, proposing a reworked structure that is compatible with both, SGAM and SPES.

The fourth industrial revolution resulted in the emergence of new ways for engineering production lines with the goal to optimize manufacturing processes by reducing expenses at the same time. This trend is mainly supported by the application of interconnected components, which are managed in a distributed way, as impelled by the Industrial Internet of Things (IIoT) or Cyber-physical Systems (CPS). While the complexity of these Systems of Systems (SoSs) is constantly rising, requirements engineering before actual implementing the system becomes increasingly difficult. Thus, comprehensive frameworks providing a foundation for model-based systems engineering, like the Reference Architecture Model Industrie 4.0 (RAMI 4.0) or the Software Platform Embedded Systems (SPES), have been introduced. Although providing a standardized methodology, missing formalizations impede the modeling of systems according to detailed architectural specifications. To address these issues, this paper introduces an approach for eliciting, specifying and managing requirements according to a particular process and aligned to architectural standards. In the end, the result is evaluated by a real-world case study. The outcome of this work will either verify the future system to develop by ensuring traceability or enable interoperability by dealing as a consistent reference point across viewpoints, systems or even domains.

Engineering of vehicular embedded systems is a difficult task, as the ongoing integration of Cyber-physical Systems (CPS) or automation potentials during vehicle development leads to increasing complexity. In particular, to develop current or future Electric Vehicles (EV) is challenging due to different domains to consider while engineering the sub-components of the vehicle itself. Thus, in order to enable a mutual development of vehicular embedded systems on multiple abstraction levels, the Software Platform Embedded Systems (SPES) has been introduced. To cope with the complexity, this framework introduces viewpoints and hierarchy layers in shape of a matrix. However, while additionally multiple domains have to be considered when developing an EV, the SPES methodology is also missing specifications, impeding its application in actual industrial scenarios. To deal with both of the mentioned issues, this paper introduces an approach for Model-based Systems Engineering of electric vehicle systems based on SPES. By doing so, this framework is further refined by an architecture definition based on the ISO 42010 and a corresponding development process. By utilizing the EV use case, the outcome is thereby validated towards its industrial feasibility, which will enhance the applicability of SPES on the one hand and contribute to the Automotive area by dealing with the increasing complexity while engineering vehicular embedded systems on the other hand.

The transformation from original production line manufacturing towards complex value creation networks causes new challenges for manufacturing companies to stay in touch with or even delimit from the competition. This new trend, resulting from the emergence of the Industrial Internet of Things (IIoT), offers new automation possibilities by interconnecting all system components with each other. However, as those components, mainly known as Cyber-physical Systems (CPS) are usually systems themselves, the complexity of such a manufacturing system continously rises, which even has to be classified as a complex System of Systems (SoS). Having recognized the issue of the upcoming difficulties when engineering such a system, several German institutes introduced the Reference Architecture Industrie 4.0 (RAMI 4.0), a three-dimensional framework that offers standards or methods for developing system architectures on basis of different views. Nevertheless, at the current point of view, this approach though looks good on paper, but almost none practical applications are existing yet. Thus, this paper focuses on the model-based development of system architectures according to the specifications of RAMI 4.0 in order to evaluate the practical applicability of this reference architecture for usage in actual industrial projects. This is done by utilizing two actual industrial case studies, whose models are created with the help of a specifically designed software tool, the so-called RAMI Toolbox. Based on the outcome of this work, the usage of RAMI 4.0 for application in the industrial area could be sustainably consolidated.