TRB Pubsindex
Text Size:
x

« Prev Next »

Return to Results | New Search

Bookmark and Share

Title:

Bayesian Spatial Modeling and Risk Mapping of Downed Trees Along Roadways Using Data from Hurricanes Irene and Sandy

Accession Number:

01550575

Record Type:

Component

Availability:

Transportation Research Board Business Office

500 Fifth Street, NW
Washington, DC 20001 United States

Abstract:

Two recent hurricanes namely, Irene and Sandy, crippled the roadway network due to storm flooding and downed trees. Roadways are critical for the operations of the agencies responding to hurricanes as well as the evacuation of the affected people. Therefore, there is a need to improve roadway safety and accessibility in the aftermath of the hurricanes. By better handling hurricane related incidents on the roadways such as downed trees, roadway safety and accessibility can be improved. This paper presents a municipality level zero inflated hierarchical Bayesian model for predicting the downed trees during these two recent hurricanes in New Jersey. First, the most influential factors on downed trees are identified and then a spatial Poisson model is estimated. In order to better visualize the model estimations, several maps are developed based on the modeling results. These maps enable a fast and intuitive understanding of model predictions especially for non-statisticians. This model can be applied to future hurricanes by using known roadway characteristics and weather forecast data. Since the model is developed using hierarchical Bayesian modeling framework, it is also possible to make best/worst case scenario analysis based on the lower/upper bound values of the predicted parameters to facilitate resource allocation during hurricanes. The model is also portable to the other regions since the data used in the estimation is available in other regions of the US as well.

Supplemental Notes:

This paper was sponsored by TRB committee ABE40 Critical Transportation Infrastructure Protection.

Monograph Accession #:

01550057

Report/Paper Numbers:

15-5165

Language:

English

Corporate Authors:

Transportation Research Board

500 Fifth Street, NW
Washington, DC 20001 United States

Authors:

Demiroluk, Sami
Ozbay, Kaan

Pagination:

17p

Publication Date:

2015

Conference:

Transportation Research Board 94th Annual Meeting

Location: Washington DC, United States
Date: 2015-1-11 to 2015-1-15
Sponsors: Transportation Research Board

Media Type:

Digital/other

Features:

Figures; Maps; References (24) ; Tables

Geographic Terms:

New Jersey

Subject Areas:

Highways; Safety and Human Factors; Security and Emergencies; I72: Traffic and Transport Planning

Source Data:

Transportation Research Board Annual Meeting 2015 Paper #15-5165

Files:

TRIS, TRB, ATRI

Created Date:

Dec 30 2014 1:44PM