Fuzzy-enabled failure behaviour analysis for dependability assessment of networked systems

Dependability assessment of networked component-based systems requires fine-grained modelling of the failure behaviour and propagation aspects of individual components. We have recently introduced a formalism called FI⁴FA, enabling the analysis of I⁴ (incompletion, inconsistency, interference and impermanence) failures as well as the analysis of the corresponding mitigations. FI⁴FA, like its predecessors, assumes that the failure behaviour of individual components is well defined in a deterministic way. However, in reality there exist multiple sources of uncertainty (e.g. in the estimates of individual component failure attributes, in the semantics of their composition, etc.) that require consideration. In this paper, we propose a new technique for the assessment of the failure behaviour of networked component-based systems. The proposed technique builds on FI⁴FA and incorporates the specificities of failure behaviours of networked systems in the I⁴. The specification of the failure behaviour of individual components is based on fuzzy sets, which have the potential to model and thus address the uncertainty aspect.