Security impact assessment of industrial automation systems using genetic algorithm and simulation

Much of the research on security of industrial automation systems has focused on countermeasures such as intrusion detection, certificate management or public key infrastructures. Due to limited resources, countermeasures should be focused to prevent the attacks with highest potential for damage. The impact of an attack can only be determined through a detailed analysis of the interactions of the automation system and the physical system under control. Attacks against single components are similar to ordinary component failures, so our focus is on deliberate damage to several components, since such scenarios are not considered in reliability engineering methods used for industrial automation systems. A simulation based security impact assessment method is proposed, using genetic algorithms to explore the range of possible attacks.