Recent advancements in network science showed that the topological credentials (i.e., rank of relative importance) of network components (such as nodes and links), carry significant implications as it is critical to know which components contribute the most to the overall network performance. For transportation networks, critical components (roads, bridges) may become inaccessible for adjacent traffic due to day-to-day congestion or external disruptions (i.e., man-made or natural hazards) that significantly reduce the level of service. Hence, topological credentials of critical network components based on their connectivity need to be assessed to enhance the serviceability, i.e., improved travel time experience as well as the ability to recover from sudden disruptions. Although the literature on network science and transportation systems’ resilience has recently advanced, the empirical literature does not provide enough guidance on systematic applications of topological credentials to infer novel, more efficient strategies for transportation network resilience and serviceability.
The goal of the dissertation is to enhance the serviceability and resiliency of transportation networks based on the topological credentials of network components as well as systematic design interventions made on critical components. To achieve this goal, this dissertation emphasized coordinated and extensive network experiments conducted at different geographic scales (i.e., city, county, and state) by using real road network data from several locations, including Miami-Dade County, Sioux Falls, Boise, among others. Results indicate that network credentials change significantly when different attributes (i.e., vehicular traffic) are introduced to the network topology. Such credentials also contribute towards generating recovery schemes in the aftermath of any network disruptions. In addition, microscopic traffic simulations indicate that design interventions, such as increased number of travel lanes on critical links, help achieve better serviceability rather than intervening on less critical congested links. The methodologies and findings of the dissertation can help traffic managers and practitioners decide on recovery strategies and design interventions efficiently to ensure more serviceable and resilient transportation networks.