Impact of topology and congestion on link criticality rankings in transportation networks

Abstract Due to the lack of ground truth on the link criticality rankings of transportation networks, it is not possible to identify the most accurate metric. While several studies provide reviews of the existing criticality metrics, the consistencies of calculated rankings by these metrics are yet to be explored. In this paper, link criticality rankings with five different metrics are compared under 150 test networks of different network topologies and congestion levels. Accordingly, the consensus metric that provides a ranking that is closest to the aggregate ranking is identified for each test network. The results show transportation network efficiency metric is the most frequent consensus metric and link flow can serve as proxy for criticality analysis under certain network characteristics. The level of congestion and network transitivity are the most important features to determine the consensus metric. The proposed framework is then tested on Sioux Falls, Eastern Massachusetts, and Anaheim networks.