Advancing mixed criticality scheduling techniques to support industrial applications

Safety critical software development is an extremely costly endeavour; software developers must forever target efficient processes that reduce software cost, while allowing significant increases in system size. The key challenge being how to reduce software cost, without compromising safety or quality.

The focus of this thesis is to research the development and temporal proof of a mixed criticality system. The thesis, which attempts to define an end to end process, begins by studying appropriate and efficient methods for assessing the timing performance of system components. The key being an approach that can be applied automatically at an early point in the design lifecycle.

The thesis then progresses to study how existing mixed criticality research needs to be advanced and matured in order to support an industrial safety critical application. This includes the definition of a scheduling model designed to provide the necessary protections advised by international aviation guidelines. In the final part of this thesis the timing process and mixed criticality system model are brought together to explore how a real system using these techniques could be validated.