Future automotive on-board networks are expected to integrate various functions on only a few centralized processing platforms. Combined with attack surfaces that originate from the external connectivity of modern vehicles, this turns the design of secure on-board networks into a challenging endeavor. Therefore, we present a formal model to describe both confidentiality and integrity requirements of applications in such a network using security levels. The proposed model is then integrated into an existing design methodology for the automatic configuration of access protection units in MPSoCs. The resulting methodology ensures that the above-mentioned requirements are automatically enforced during runtime and is validated using a safety-critical example scenario from the automotive domain.