Reducing Cascading Failure Risk by Increasing Infrastructure Network Interdependence.

Area of Science:

• Infrastructure resilience and network science.
• Complex systems analysis.
• Cyber-physical systems security.

Background:

• Critical infrastructure networks are increasingly interconnected, raising concerns about cascading failures.
• Existing models often oversimplify cascading mechanisms and neglect benefits of intelligent control.
• Topological models may not accurately capture real-world infrastructure dynamics.

Purpose of the Study:

• To compare the robustness of simple topological network models with more dynamic, realistic models of coupled infrastructures.
• To investigate the impact of interconnecting power and communication systems on overall vulnerability.
• To evaluate the role of intelligent control in mitigating cascading failures.

Main Methods:

• Comparison of a topological contagion model with a power grid model.
• Comparison of a percolation model with three interdependent power-communication system models.
• Development and analysis of a "smart" power network model coupled to a communication system.

Main Results:

• Detailed models yield different conclusions than simple topological models regarding network vulnerability.
• Increased coupling between power and communication systems generally decreases vulnerability, contrary to percolation model predictions.
• A "smart" power network coupled with a communication system shows reduced vulnerability, except in extreme scenarios.

Conclusions:

• Interconnecting networks with complementary capabilities can enhance robustness.
• Constraining internetwork failure propagation is key to leveraging coupling benefits.
• Realistic dynamic models are crucial for understanding and improving critical infrastructure resilience.