I. Introduction
In the development of embedded systems, MATLAB Simulink[1] is the most widely used industrial tool in the automotive industry. Modeling real world problems like braking systems or engine controllers result in Simulink models containing several ten thousands of blocks and more. These embedded systems are often deployed in safety critical areas which requires comprehensive quality assurance measures. In practice, quality assurance methods like testing are well established in the development of embedded systems with Simulink. But static methods are often limited to manual reviews and syntactical checks or are performed on the code generated from models. Due to complex structural hierarchies and model referencing, it is hard to recognize data and control flow when reviewing complex Simulink models manually.