Analysis of Urban Traffic Network Vulnerability and Classification of Signalized Intersections

This work studies urban road traffic networks and discusses potential ways to assess the criticality of urban links and nodes. A series of indicators are defined that are useful to evaluate spatial and temporal features and facilitate the identification of congestion hot spots. One way to achieve this is by utilizing concepts from graph theory (e.g. connectivity, efficiency, betweenness). These indicators are based on the structure of the network (graph) and do not require any traffic data. On the other hand, one could utilize historical traffic data (e.g. flows, densities, average speeds) in order to evaluate the importance of an urban link (or node) to the overall performance of the network. Such an analysis could capture congestion dynamics, spill-backs, and queues propagation in the network. Here, we combine both approaches mentioned above to come up with a classification (or ordering) of the different links and nodes inside an urban zone. This classification can then be used for real-time traffic control purposes, e.g. a city can choose the intersections that need to be instrumented and thus reduce the operational cost of online traffic management. Finally, we discuss the correlations between the different indicators and how one could take this into account in the decision making process.