Estimating road network accessibility during a hurricane evacuation: A case study of hurricane Irma in Florida. (June 2020)

• Record Type: Journal Article
• Authors: Zhu, Yi-Jie; Hu, Yujie; Collins, Jennifer M.
• Is Part Of: Transportation research. Volume 83(2020)
• Publication Date: 2020-06
• Subjects: Road accessibility -- Hurricane -- Evacuation -- Behavioral response curves -- Distance decay; Transportation -- Research -- Periodicals; Transportation -- Environmental aspects -- Periodicals; 354.76
• DOI: 10.1016/j.trd.2020.102334
• Languages: English
• ISSNs: 1361-9209

Abstract

Highlights: Proposes a method to measure road network accessibility during a hurricane evacuation. The method accounts for both dynamic evacuation demand and hurricane characteristics. The method is applied to a case study of Hurricane Irma in the State of Florida. Results can help emergency agencies identify congested roads and vulnerable neighborhoods. Abstract: Understanding the spatio-temporal road network accessibility during a hurricane evacuation—the level of ease of residents in an area in reaching evacuation destination sites through the road network—is a critical component of emergency management. While many studies have attempted to measure road accessibility (either in the scope of evacuation or beyond), few have considered both dynamic evacuation demand and characteristics of a hurricane. This study proposes a methodological framework to achieve this goal. In an interval of every six hours, the method first estimates the evacuation demand in terms of number of vehicles per household in each county subdivision (sub-county) by considering the hurricane's wind radius and track. The closest facility analysis is then employed to model evacuees' route choices towards the predefined evacuation destinations. The potential crowdedness index ( PCI ), a metric capturing the level of crowdedness of each road segment, is then computed by coupling the estimated evacuation demand and route choices. Finally, the road accessibility of each sub-county is measured by calculating the reciprocal of the sum of PCI values of corresponding roads connecting evacuees from the sub-county to the designated destinations. The method is applied to the entire state of Florida during Hurricane Irma in September 2017. Results show that I-75 and I-95 northbound have a high level of congestion, and sub-counties along the northbound I-95 suffer from the worst road accessibility. In addition, this research performs a sensitivity analysis for examining the impacts of different choices of behavioral response curves on accessibility results.