Comparison Between Traditional Methods and nd the EB-SPF Method for Identifying High-Crash-Risk Intersections

摘要

Identifying high-crash-risk locations, called hot spots, is an important step in improving roadwaysafety. Use of the empirical Bayes (EB) method coupled with the use of safety performancefunctions (SPFs) is regarded as the state of practice in identifying such locations. To apply theEB-SPF method, however, requires considerable resources in preparing data as well as statisticalexpertise. As a consequence, many highway agencies still rely on traditional methods that usecrash frequency and crash rate to identify locations for potential safety improvements withoutknowing the accuracy of such methods.This study examined four traditional methods commonly used to identify potentiallocations for safety improvements and compared them to the EB-SPF method. The four methodsevaluated were crash frequency, crash rate, rate-quality control, and equivalent property damageonly methods. The study was limited to four-leg intersections with either a traffic signal or two13way stop control, and 2004-2008 data were collected at 1,670 such intersections. The studyfound that the crash frequency method performed the best among the four methods in correctlyidentifying the top 1 percent of unsafe intersections. However, the method tended to flagintersections as top hot spots incorrectly. The rate-quality control method performed the bestwith respect to identifying the top 5 and 10 percent of unsafe intersections. The findings areexpected to help highway agencies that continue to use the traditional methods choose the mostappropriate method so that scarce resources available for safety improvement can be investedeffectively.