Date of Award
Doctor of Philosophy (PhD)
Computational Modeling & Simulation Engineering
Modeling and Simulation
The integration of Information Technology (IT) and Operational Technology (OT) in Cyber-Physical Systems (CPS) has resulted in increased efficiency and facilitated real-time information acquisition, processing, and decision making. However, the increase in automation technology and the use of the internet for connecting, remote controlling, and supervising systems and facilities has also increased the likelihood of cybersecurity threats that can impact safety of humans and property. There is a need to assess cybersecurity risks in the power grid, nuclear plants, chemical factories, etc. to gain insight into the likelihood of safety hazards. Quantitative cybersecurity risk assessment will lead to informed cyber defense remediation and will ensure the presence of a mitigation plan to prevent safety hazards. In this dissertation, using Energy Delivery Systems (EDS) as a use case to contextualize a CPS, we address key research challenges in managing cyber risk for cyber defense remediation.
First, we developed a platform for modeling and analyzing the effect of cyber threats and random system faults on EDS's safety that could lead to catastrophic damages. We developed a data-driven attack graph and fault graph-based model to characterize the exploitability and impact of threats in EDS. We created an operational impact assessment to quantify the damages. Finally, we developed a strategic response decision capability that presents optimal mitigation actions and policies that balance the tradeoff between operational resilience (tactical risk) and strategic risk.
Next, we addressed the challenge of management of tactical risk based on a prioritized cyber defense remediation plan. A prioritized cyber defense remediation plan is critical for effective risk management in EDS. Due to EDS's complexity in terms of the heterogeneous nature of blending IT and OT and Industrial Control System (ICS), scale, and critical processes tasks, prioritized remediation should be applied gradually to protect critical assets. We proposed a methodology for prioritizing cyber risk remediation plans by detecting and evaluating critical EDS nodes' paths. We conducted evaluation of critical nodes characteristics based on nodes' architectural positions, measure of centrality based on nodes' connectivity and frequency of network traffic, as well as the controlled amount of electrical power. The model also examines the relationship between cost models of budget allocation for removing vulnerabilities on critical nodes and their impact on gradual readiness. The proposed cost models were empirically validated in an existing network ICS test-bed computing nodes criticality. Two cost models were examined, and although varied, we concluded the lack of correlation between types of cost models to most damageable attack path and critical nodes readiness.
Finally, we proposed a time-varying dynamical model for the cyber defense remediation in EDS. We utilize the stochastic evolutionary game model to simulate the dynamic adversary of cyber-attack-defense. We leveraged the Logit Quantal Response Dynamics (LQRD) model to quantify real-world players' cognitive differences. We proposed the optimal decision making approach by calculating the stable evolutionary equilibrium and balancing defense costs and benefits. Case studies on EDS indicate that the proposed method can help the defender predict possible attack action, select the related optimal defense strategy over time, and gain the maximum defense payoffs. We also leveraged software-defined networking (SDN) in EDS for dynamical cyber defense remediation. We presented an approach to aid the selection security controls dynamically in an SDN-enabled EDS and achieve tradeoffs between providing security and Quality of Service (QoS). We modeled the security costs based on end-to-end packet delay and throughput. We proposed a non-dominated sorting based multi-objective optimization framework which can be implemented within an SDN controller to address the joint problem of optimizing between security and QoS parameters by alleviating time complexity at O(MN2). The M is the number of objective functions, and N is the population for each generation, respectively. We presented simulation results that illustrate how data availability and data integrity can be achieved while maintaining QoS constraints.
"Cyber Defense Remediation in Energy Delivery Systems"
(2020). Doctor of Philosophy (PhD), Dissertation, Computational Modeling & Simulation Engineering, Old Dominion University, DOI: 10.25777/f744-2a87