Quantifying Consequences of Externally Induced Failures Propagated through Systems during Functional System Design

ABSTRACT (maximum 200 words)Assessment of failure propagation and potential within complex systems is a field open for continued exploration in the arena of systems engineering. Risk assessment and failure modeling processes such as PRA,FTA, and FMEA/FMECA are more widely understood and utilized in industry, yet are not designed to fully address and objectively quantify the impact on systems when exposed to intentionally malicious attacks, particularly in early design stages where changes to system architectures are best effected. Further, current methods do not identify and standardize attack modes that are likely to affect systems during their life cycle. This work first defines attacks and discusses their difference from failures. The work then develops and discusses a hierarchical taxonomy of attack classes and mechanisms likely to affect a wide array of systems. Finally, it presents the Failure Path Length Method (FPLM) to quantify consequence on systems due to attacks on system functions by applying characteristics of those classified attacks to the functional architecture of a system. The author then implements the FPLM on a common EPS to verify applicability to realistic systems and objectively determine the consequence of an attack. The differences in consequence drive mitigating changes to the architecture of the EPS and validate the significant decision-making power provided to system designers by the proposed method during functional analysis and design.