Integrated Safety-Security Risk Assessment for Production Systems: A Use Case Using Bayesian Belief Networks

Abstract

Industrial control systems (ICSs) are complex networked systems that enable automation of large-scale processes. Depending on the application domain, the risk of the failure of components can have catastrophic repercussions. Up to now, a safety risk assessment is carried out to identify and narrow down possible failures. However, with the recent increase of cybersecurity attacks, a need of an integrated safety and security risk assessment is rising. This encompasses a comprehensive approach to assess the risks associated with ICSs and develop strategies for mitigating those risks. This paper proposes Bayesian Belief Network (BBN) as a representative of a probabilistic method and show its suitability for an integrated safety and security risk assessment. The method is evaluated by means of a use case. It provides risk propagation of functional safety, human safety and shows a propagation path from security to functional safety. The assessment is based on practical vulnerability assessments, technical documentations, manual observation and expert opinions.