Citation

Mirjalili R, Barati H, Yazici A. Transp. Res. Rec. 2023; 2677(1): 694-707.

Copyright

(Copyright © 2023, Transportation Research Board, National Research Council, National Academy of Sciences USA, Publisher SAGE Publishing)

DOI

10.1177/03611981221101034

PMID

unavailable

Abstract

This study presents a quantitative analysis of the snow storm resilience of New York City (NYC) by utilizing network science-based system performance functions and publicly available datasets, that is, New York Department of Sanitation?s snow removal operation and NYC traffic speed data. Several graph theory based metrics and some heuristic metrics are utilized to calculate the temporal changes in transportation network functionality along eight snow events. The NYC transportation network is updated based on snowplow movements, and the performance indexes (PIs) for all metrics are calculated throughout the snow storm timeline. Since PI graphs rely on network topology but not the actual traffic conditions, the times that the system bounces back to ?regular? conditions (i.e., time-to-recovery/resilience) are calculated based on the similarity between hourly speed distributions on NYC roads. Bhattacharyya distance and Kolmogorov?Smirnov test are used as measures for distribution similarity. Accordingly, the PI values that correspond to the recovery times are also identified. Within the limitations of the size of the snow storm sample, the findings show that less data-intensive graph theory metrics can be used to estimate the transportation network performance?an estimation that would require extensive and detailed data otherwise. Accordingly, these metrics can be used to make resilience predictions for future events through simulations on modified network topology, and help make recovery forecasts to inform local governments and businesses on when to resume regular operations.

Language: en