Date of Award

Spring 5-2016

Level of Access Assigned by Author

Campus-Only Thesis

Degree Name

Master of Civil Engineering (MCE)


Civil Engineering


Per Garder

Second Committee Member

Roberto Lopez-Anido

Third Committee Member

Thomas C. Sandford


The general public perception of bridges is that driving or walking on them is more dangerous than on normal roadway segments. A goal of this research has been to analyze whether that perception is founded on factual knowledge or not. Are bridge segments or roads going under bridges more dangerous or not than regular segments away from bridges, and if so, by how much and in what way do their safety differ? For this analysis, data from police reports were collected, including data provided by the National Highway Traffic Safety Administration (NHTSA) through their Fatal Analysis Reporting System (FARS) and from the Maine Department of Transportation (MaineDOT). The data has been used to statistically analyze whether bridges experience more crashes than normal roadway segments or not. The data is focused on three types of crashes: property damage only, bodily injury, and fatalities. In order to compare the data for bridges to non-bridge (or normal) segments, the crash data is normalized per foot of roadway and by traffic volume.

The results show higher crash rates for bridges than for almost all regular segments away from bridges. The only exception to that, among the segments analyzed, is that the crash rates of minor arterial and other principle arterials in Kennebec County are lower than that of statewide published crash rates. There are many potential reasons why bridges have higher crash rates than other segments. Prime reasons may be: ice and frost conditions negatively affecting bridges, underprepared young and inexperienced drivers not understanding the hazards bridges can produce, and other social factors discussed in this paper that could be contributing to the higher crash rates.