Year: 2013
Type: article
Abstract: Work zones are given priority by most government agencies nationwide because of the need for maintenance, rehabilitation, and advancement of the existing road networks.This results in large number of work zones and has had an inevitable impact on the regular traffic flows and traffic safety issues.The main objective of the study is to find the risk factors affecting crash severity (dependent variable) in work zones in the state of Ohio.Year 2010 data was collected from Ohio Department of Traffic Safety in the form of an excel spreadsheet.A total of 24 different independent variables which has 12,275 crash records were used in the development of the Crash Severity Model (CSM).The following steps were employed for the development of CSM.First, the Pearson chisquare statistics test was done to find the variables that have a significant relationship among themselves and the dependent variable.Second, the insignificant variables left from step 1 were selected which were found to have significant effect on crash severity in iv other studies and they were also added along with the significant variables found in step 1 for the development of the regression model.A total of 21 variables were found to have a significant relationship with the dependent variable.Three variables were selected from step 2 based on literatures.A binomial logit model was used to predict crash severity.Results of binary logistic regression showed that forty four categories of seventeen variables were found to be predictive of the fatal/injury crash severity type and also showed that the model fits to the data well with a prediction ability of 72.8 percent.vFor my parents, brothers, and friends.vi I sincerely thank my advisor Dr. Bhuiyan Alam for giving me the opportunity to pursue my research
Author: Vanishravan Katta
Institution: University of Toledo
Language: English
Topic: Traffic and Road Safety
Subfield: Safety, Risk, Reliability and Quality
Field: Engineering
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W127585693