ABSTRACT
When developing a safety critical system, there are many aspects that need to be balanced against each other in order to reach an optimal design such as safety requirements, reliability goal, performance specifications and budget constraints. In an early design stage, it is vital to be able to screen the design space for a set of promising design alternatives for further studies. This paper proposes an approach capable of investigating the trade-offs described above, combining the techniques for system safety and reliability analysis with optimization methods. Markov analysis is employed for modeling the system safety and reliability characteristics and a Genetic Algorithm is used for optimization. The proposed method is applied to the design of an electric supply system for an aircraft, involving selection of components from different suppliers. First a model is built for each objective, i.e. cost, safety, and reliability. The models are validated and optimization is performed. The obtained result is the selection of suppliers for each component in the system in order to achieve a balance between system safety, reliability, and other design objectives.
