Flood evacuation and rescue: The identification of critical road segments using whole-landscape features

Edward Helderop, Tony H. Grubesic

Research output: Contribution to journalArticlepeer-review

20 Scopus citations


When studying real-world infrastructure systems as networks, one common avenue of analysis is the identification of critical network elements. These features are often defined as the nodes or edges in a given network that play an outsized role in the structure and functionality of the system. When network systems are under duress, particularly during (or just after) large-scale natural disasters, the continuity of networks is heavily dependent on these critical elements. If the critical features are vulnerable to disruption, the robustness of the network is compromised. The purpose of this paper is to introduce a novel geocomputational method for detecting critical road segments in a post-disaster landscape, with an eye toward human mobility and emergency response. This method accounts for the impacts of non-road landscape features on the overall traversability of an area and is contrasted with traditional critical feature analysis. The developed method provides several advantages, including the ability to produce higher-resolution, higher-fidelity criticality metrics for individual road segments. The paper concludes with a discussion on potential benefits to strategic urban planning, evacuation, and rescue planning during extreme events.

Original languageEnglish (US)
Article number100022
JournalTransportation Research Interdisciplinary Perspectives
StatePublished - Dec 2019


  • Emergency response
  • Extreme events
  • Network analysis
  • Storm surge
  • Transport
  • Vulnerability

ASJC Scopus subject areas

  • Civil and Structural Engineering
  • Automotive Engineering
  • Transportation
  • Management Science and Operations Research


Dive into the research topics of 'Flood evacuation and rescue: The identification of critical road segments using whole-landscape features'. Together they form a unique fingerprint.

Cite this