Vulnerability of nodes under controlled network topology and flow autocorrelation conditions

- The paper presents a simulation study of node vulnerability in networks.
- Two types of networks are investigated: Poisson and Power Law.
- Analysis is based on the topology of the network and also on the network autocorrelation of flows.
- The results indicate that network autocorrelation is an important attribute when assessing the impact of disruption to nodes.

Infrastructure networks, including transportation, telecommunications, and energy grids, are essential for the correct functioning of many economic and social processes. Considering the import of networked critical infrastructure, it is clear that understanding the characteristics of networks that make them vulnerable to disruption is a valuable endeavor. As recent research demonstrates, network vulnerability is a very relevant issue on the design and operation of any networked process in general, and transportation in particular, where the main interest is on identifying network elements that are singularly sensible to disruptions to their integrity. While there have been some suggestions from the literature that the configuration of the network may influence the vulnerability of individual elements, a systematic investigation of network topology and vulnerability has not been conducted. Furthermore, neither has there been an account of the relevance of coherence in the contents of the network, or in other words, the autocorrelation of flows. Accordingly, the objective of this paper is to investigate, from the perspective of the integrity of nodes, the vulnerability to disruption of individual network elements, while controlling for network topology and autocorrelation. Analysis is based on extensive numerical simulations for different network configurations and spatial autocorrelation levels, and the results obtained demonstrate that the proper evaluation of network vulnerability requires not only an understanding of its topology, but also of the distribution of flows.