On the Choice of Effective Artificial Intelligence Methods for Cyber Defense of Industrial Systems

Unlike Information Systems, in Industrial Automation and Control Systems (IACS) there is a second component of Cyber Security – the Functional Safety, which is particularly important due to the presence of the so-called Dangerous Failures (those that threaten the life and health of people and the state of the environment). This determines a specific approach in the case of applying Artificial Intelligence methods for achievement of acceptable level of IACS Cyber Security. This article describes an approach and a corresponding procedure for choosing an appropriate method from the point of view of Functional Safety. Claiming the universality of this approach, for the purpose of concretization and visualization, authors illustrate it with an application in one of the important sections of the Critical National Infrastructure – the Railroad. In case of decentralized approach to the security devices in the various “ contact points “ of the system, an appropriate selection procedure should be carried out for each of them depending on the prevailing Cyber Threats for the given point.