Abstract
Transportation networks are critical to the social and economic function of nations. Given the continuing increase in the populations of cities throughout the world, the criticality of transportation infrastructure can only be expected to increase. Many transportation network vulnerability assessment methods are based on the static traffic assignment problem, which determines the distribution of traffic demand over a network for a static snapshot in time. However, transportation networks are dynamic because network demand is time varying. Methods that consider the dynamic nature of networks are needed to identify where and when vulnerabilities arise within networks as well as effective mitigation strategies. This dissertation describes the tools built to enable dynamic transportation network vulnerability assessment algorithms, namely, a software application to extract static maps and a smartphone app to collect dynamic network demand data. Additionally, this dissertation makes several algorithmic contributions including (i) an exhaustive quantitative dynamic vulnerability assessment technique to identify vulnerable edges in a dynamic network, (ii) a method to quantify the fuel consumed and pollution generated due to network disruptions, (iii) a game theoretic method to identify network vulnerability for all edges and time intervals in parallel, (iv) a method to efficiently measure network vulnerability by only simulating a small area around the disruption, thus reducing computation time, and (v) an optimization problem solved with a genetic algorithm to place rerouting devices in the network such that vehicles avoid the disruption in order to reduce vulnerability cost effectively.