Formalisation of an industrial approach to monitoring critical data

Abstract

A large class of safety-critical control systems contains monitoring subsystems that display certain system parameters to (human) operators. Ensuring that the displayed data are sufficiently fresh and non-corrupted constitutes an important part of safety requirements. However, the monitoring subsystems are typically not a part of a safety kernel and hence often built of SIL1-SIL2 components. In this paper, we formalise a recently implemented industrial approach to architecting dependable monitoring systems, which ensures data freshness and integrity despite unreliability of their components. Moreover, we derive an architectural pattern that allows us to formally reason about data freshness and integrity. The proposed approach is illustrated by an industrial case study. © 2013 Springer-Verlag.