Quantifying the relationship between skid resistance and crashes for Iowa roadways: A framework for a skid resistance policy

Thumbnail Image
Date
2020-01-01
Authors
Abdullah, Wasama
Major Professor
Advisor
Omar O Smadi
Committee Member
Journal Title
Journal ISSN
Volume Title
Publisher
Altmetrics
Authors
Research Projects
Organizational Units
Journal Issue
Is Version Of
Versions
Series
Department
Civil, Construction, and Environmental Engineering
Abstract

The lack of sufficient friction at the tire-pavement interface is a major contributing factor to traffic crashes. The relationship between surface friction and roadway safety has been recognized since the thirties. Minimum skid resistance guidelines have been the focus of intensive research efforts, but very little research in the U.S. has addressed the different friction demand categories that can be integrated into an effective skid resistance policy. It is crucial to quantify this relationship and determine the level of roadway surface friction needed (i.e., friction demand) to eliminate roadway surface friction-related crashes and reduce the severity of those that have more complex causation. This thesis quantifies the relationship between skid resistance and crashes for Iowa roadways. The correlation between skid resistance measured with a locked-wheel trailer and crash rates for wet, dry and roadway departure crashes is investigated through employing a two-parameter, two-level skid resistance model where factors like roadway geometry, roadway functional classification, traffic volume, speed, and pavement type are considered to determine friction demand investigatory and intervention levels. The research used crash data, skid measurements, traffic volumes and tangents information, from the Iowa Department of Transportation and the Institute for Transportation for the year 2018. Friction measurements are divided into intervals with increments of 2 friction units and the number of crashes for each friction interval were determined for the overall data and for the different analysis categories for which the crash rate models are generated and summarized. The specific crash rates were utilized in the friction demand regression models generation. Regression analyses indicated that there is statistically significant effect of skid resistance on wet, dry, and roadway departure crashes; as expected, skid resistance is a factor in explaining the variation in crash rates. For all sites evaluated, friction is found to be a significant factor affecting wet crash rates except for sites with low traffic where no tangible relationship is detected between the two variables. Friction is found to be a significant factor affecting dry crash rates for most of the sites except for rural interstates. Friction was found to be a significant factor affecting roadway departure crashes at some sites where more roadway departure crashes were successfully matched with crash locations. However, no relationship was detected for urban freeways and expressways, urban principal arterials, rural minor arterials, sites with high speed limit, sites with low traffic as well as Portland cement concrete and asphalt concrete pavements. A larger study is needed to further investigate these findings with stratifying the roadway departure crashes by the surface contamination state. However, as expected, friction data tend to explain only a small portion of the variation in crash rates when considering individual crash sites and a statistically significant effect of skid resistance on the wet, dry and roadway departure crash rate is captured by grouping the crash sites by similar characteristics and better explained the variability in crash occurrence. Generally, based on the data studied, a target skid number (SN40) of 42 and 47 appears to have positive safety benefits with respect to wet and roadway departure crashes maintaining a crash risk of less than 500 crash per hundred million vehicle miles traveled on the network tangent segments.

Comments
Description
Keywords
Citation
Source
Copyright
Tue Dec 01 00:00:00 UTC 2020