Answers are grounded in this paper and cite their sources.
The concepts and methodologies behind Model-based Systems Engineering (MBSE) hold great promises concerning the development of complex systems. Various projects have been carried out successfully during the last years and demonstrated the power behind the overall concept - and the practical problems to reach the ambitious overall goals. Whereas the first steps of MBSE like the iterative modeling of requirements, designs, behaviors, and tests became standard procedures in Systems Engineering (SE), the transition to simulation often is still restricted to quite simple scenarios. Although elaborated system models deliver all the information needed, the simulation of the overall system in prospective working environments interacting with other systems is rather an exception. The problem is that there is still quite a gap between the first SE steps and the various algorithms simulation technology can offer today. Major reasons for this seem to be the resulting complexity of the system model when modeling complex interactions, the complexity of using state-of-the-art simulation technology and the absence of simulation frameworks for simulations across multiple domains and disciplines. "Experimentable Digital Twins", a concept originally developed for the eRobotics methodology, seem to have the potential to close the gap between SE and simulation by introducing a new structuring element to configure simulations. A new simulation system architecture integrating well-known simulation algorithms provides Virtual Testbeds for the simultaneous simulation of a network of different Digital Twins interacting with each other in various ways (i.e. a network of different systems, their components and their working environment). This approach has been successfully used for a variety of different applications in multiple research areas. As one application, it allows for the simulation-based optimization of parameters, system structure etc.
No takes yet. Share an insight, caveat, or question.
Schluse et al. (2017) studied this question.