TY - JOUR
T1 - Development of crash modification factors for installation of high-tension cable median barriers
AU - Russo, Brendan J.
AU - Savolainen, Peter T.
AU - Gates, Timothy J.
N1 - Publisher Copyright:
© 2016, National Research Council. All rights reserved.
PY - 2016
Y1 - 2016
N2 - Cross-median crashes are among the most hazardous collision types on freeways. The primary countermeasure to reduce such crashes is the installation of a median barrier, with a cable barrier being a widely used alternative. As of 2014, approximately 317 mi of cable median barriers had been installed on freeways throughout Michigan. To assess the safety impacts of this installation program, a comprehensive before-and-after crash analysis was conducted; this analysis included a manual crash report review to identify all median-related crashes. Safety performance functions were developed for roadway segments before and after cable barrier installation, as well as for similar control segments where no barrier had been installed. The empirical Bayes (EB) method was used to develop crash modification factors as a function of median width. To explore the effects of regression to the mean (RTM), the EB results were compared with those from a naïve before-after analysis. The results showed that cable barriers significantly reduced the frequency of fatal and severe injury crashes, although substantial increases in less severe crashes were also observed. The cable barrier was found to have greater impacts on crashes across nearly all severity levels when used on narrow medians (i.e., 26 to 50 ft) compared with wider medians (50 to 94 ft). In addition, reductions in fatal and serious injuries were less pronounced when compared with prior research (especially for wider medians), which was largely attributed to RTM effects. Ultimately, this study provides important guidance to deal with the expected safety impacts (and related costs) associated with cable median barrier installation.
AB - Cross-median crashes are among the most hazardous collision types on freeways. The primary countermeasure to reduce such crashes is the installation of a median barrier, with a cable barrier being a widely used alternative. As of 2014, approximately 317 mi of cable median barriers had been installed on freeways throughout Michigan. To assess the safety impacts of this installation program, a comprehensive before-and-after crash analysis was conducted; this analysis included a manual crash report review to identify all median-related crashes. Safety performance functions were developed for roadway segments before and after cable barrier installation, as well as for similar control segments where no barrier had been installed. The empirical Bayes (EB) method was used to develop crash modification factors as a function of median width. To explore the effects of regression to the mean (RTM), the EB results were compared with those from a naïve before-after analysis. The results showed that cable barriers significantly reduced the frequency of fatal and severe injury crashes, although substantial increases in less severe crashes were also observed. The cable barrier was found to have greater impacts on crashes across nearly all severity levels when used on narrow medians (i.e., 26 to 50 ft) compared with wider medians (50 to 94 ft). In addition, reductions in fatal and serious injuries were less pronounced when compared with prior research (especially for wider medians), which was largely attributed to RTM effects. Ultimately, this study provides important guidance to deal with the expected safety impacts (and related costs) associated with cable median barrier installation.
UR - http://www.scopus.com/inward/record.url?scp=85012098046&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85012098046&partnerID=8YFLogxK
U2 - 10.3141/2588-13
DO - 10.3141/2588-13
M3 - Article
AN - SCOPUS:85012098046
SN - 0361-1981
VL - 2588
SP - 116
EP - 125
JO - Transportation Research Record
JF - Transportation Research Record
ER -