Cascading failures in coupled networks with both inner-dependency and inter-dependency links

Area of Science:

• Network Science
• Statistical Physics

Background:

• Coupled networks exhibit complex behaviors influenced by dependencies between nodes.
• Understanding percolation transitions in these systems is crucial for network resilience.

Purpose of the Study:

• To investigate percolation in coupled networks with both inner- and inter-dependency links.
• To determine how the ratio of dependency link types affects system robustness and transition behavior.

Main Methods:

• Simulations of coupled network models.
• Analysis of percolation transitions under varying dependency link ratios.
• Development of a theoretical approach for model analysis.

Main Results:

• Fragile, discontinuous percolation transitions occur when dependency links are predominantly inner or inter.
• Robust, continuous transitions emerge when inner and inter-dependency links are balanced.
• A critical average network degree exists, below which discontinuous transitions are always observed, irrespective of link ratio.

Conclusions:

• The balance of dependency link types, not just density, dictates percolation transition type in coupled networks.
• System robustness can be optimized by tuning the ratio of inner to inter-dependency links.
• Network topology (average degree) plays a critical role in the nature of percolation transitions.