Early Insight in Systems Design through Modeling and Simulation

Abstract

In early design stages, system architects mostly rely on estimations to make design decisions. These are based on the available information at hand and their experience. Modeling and simulation is almost exclusively applied in more detailed stages of design. In this paper we present an approach aimed at making better informed design decisions, early in the design process.Our approach focuses on giving insights in early design through simulations and models that are usually only provided in more detailed design stages. To do so, we propose a framework and address three conflicts that arise when connecting techniques from early and detailed design stages. These are dealing with uncertainty, accommodating multidisciplinary views and accounting for more divergent design space exploration strategies.The approach has been applied to a medical imaging system, to analyze a possible latency reduction. The goal of this case study was to gain realistic insight in system latency using a highly abstracted system model and a generic simulation model. Insights gained with these models confirmed that a new design reduces system latency and deals better with large variations in latency. The underlying structure of the approach has proven itself to be feasible. Further research is necessary to determine whether the approach can cover a broader range of applications and to evaluate how the full approach can be implemented.