Cascading Edge Failures: A Dynamic Network Process

TL;DR

This paper introduces the Dynamic Bond Percolation process to model dependent edge failures in networks, analyzing how local rules influence global failure patterns and the emergence of systemic risks.

Contribution

It presents a novel stochastic model for dependent edge dynamics in networks, extending beyond independence assumptions and analyzing global behaviors.

Findings

The process converges to a stationary distribution.

Different substructures have varying failure susceptibilities.

Global failure or recovery patterns depend on local rules.

Abstract

This paper considers the dynamics of edges in a network. The Dynamic Bond Percolation (DBP) process models, through stochastic local rules, the dependence of an edge $(a,b)$ in a network on the states of its neighboring edges. Unlike previous models, DBP does not assume statistical independence between different edges. In applications, this means for example that failures of transmission lines in a power grid are not statistically independent, or alternatively, relationships between individuals (dyads) can lead to changes in other dyads in a social network. We consider the time evolution of the probability distribution of the network state, the collective states of all the edges (bonds), and show that it converges to a stationary distribution. We use this distribution to study the emergence of global behaviors like consensus (i.e., catastrophic failure or full recovery of the entire grid) or coexistence (i.e., some failed and some operating substructures in the grid). In particular, we show that, depending on the local dynamical rule, different network substructures, such as hub or triangle subgraphs, are more prone to failure.