Probabilistic Modeling and Study of Cybersecurity Attacks in Industrial Control Systems of Plants

摘要

Cybersecurity attacks on plant industrial control systems (ICS) have been on the rise. In this paper, we review plant vulnerabilities, plant cyberattacks, and security controls. We also develop and evaluate a probabilistic model for cybersecurity attacks to plant ICS with five possible attacks, three cybersecurity defense strengths, and six cybersecurity defense scenarios. A model of the overhead costs of the six cyberattacks is also presented. We then evaluate the probability of success of these six modeled attacks, particularly under the six defense scenarios. This model helps in assessing the effectiveness of cybersecurity defenses facing a variety of plant cyberattacks and can be customized for different plant situations. Our results indicate that investing in securing the lower ICS network hierarchy levels is more effective than securing the higher levels including the enterprise network level.