Relating risk and reliability predictions to design and development choices

Ideally, planning and managing the design and development of complex systems should: 1) consider the entire lifecycle (design, development, testing, integration, deployment, operation and decommissioning). 2) Take risk and reliability into account, as well as more traditional measures such as cost and performance, when making tradeoff decisions. A key enabler to all of the above is the ability to relate design and development choices to risk and reliability predictions. The challenge is that in many cases the number of design and development alternatives is large. In this context, the problem of relating reliability predictions to design and development choices is non-trivial. Over several years we have developed a risk-based model the hallmark of which is the explicit representation of risk mitigations as options. We describe how this model functions, and the major implications of making mitigation options first class objects within an (otherwise relatively simple) analysis model. We also describe elaborations to this model´s representation of risks, notably by the incorporation of fault tree notions. These improve the fidelity of the designs we are able to represent, and also offer the ability to represent design alternatives within the same framework. Finally, we describe the connections we are building between our risk analysis tool and other risk tools