Urban arterials by their nature carry heavy traffic volumes and generate la
rge numbers of motor vehicle crashes. The present study involved review of
police crash reports to identify precrash events and driver actions for a s
ample of crashes on urban arterials and describes a method for reducing suc
h crashes based on analyses of collision patterns and identification of loc
ations with excessive numbers of crashes of a particular type. Police-repor
ted crash data were obtained for three urban arterials in the Washington, D
C metropolitan area. A total of 2013 crash reports were analyzed. Seven cra
sh types accounted for nearly 90% of these reports. On each arterial studie
d, several locations with excessive numbers of crashes of a particular type
were identified, and corresponding engineering countermeasures were recomm
ended. Differences between the approach employed in this study and traditio
nal blackspot analyses are discussed. (C) 2001 Elsevier Science Ltd. All ri
ghts reserved.