Complex Network Method of Evaluating Resilience in Surface Transportation Networks

Abstract

A complex network analysis methodology was adopted to evaluate structural resilience in surface transportation networks with the use of examples of the U.S. highway and Interstate networks in Connecticut and the Indiana interurban railroad network. Resilience in these networks was evaluated alongside that of a biological network, which through millions of years of evolution had developed an adaptive behavior in which cost, efficiency, and resilience were optimized in the feeding network that was constructed. Disruptions in the networks were simulated by using link-based targeted and random strategies. With simulation results, network performance under disruption was assessed by using two metrics: global efficiency and the relative size of the giant component for each disruption strategy. The biological network exhibits superior resistance to disruption regardless of strategy, a quality attributed to its redundant and cyclic weblike network structure and its innate ability to adapt to disruptions by developing network structures that have been honed through millions of years of evolution. In addition, linear correlations between network structural metrics such as the average degree, density, and average clustering coefficient were explored and analyzed.