摘要: |
Accident reduction factors (ARFs) are mechanisms that the California Department of Transportation employs in calculating Traffic Safety Index values, which are used to prioritize safety-improvement projects on California state highways. Many factors that affect motorist safety have changed over the nearly 30 years in which ARFs have been used in the Traffic Safety Index Program. It is therefore appropriate to review and update, as needed, these ARFs. The research focused on four ARFs that are currently in use (rumble-strip installation, shoulder widening, superelevation correction, and curve correction). An ARF was developed for a fifth treatment category (wet-pavement treatments). Data were gathered for all projects proposed for funding in the state of California's Safety Improvement Program from 1985 through 1995. Projects completed from 1988 through 1992 were considered for inclusion in a before-and-after study that employed empirical Bayesian statistical analysis. A Bayesian statistical software package, BEATS (Bayesian Estimation of Accidents in Transportation Studies), was used in the analysis. The study reviewed scope of work for each of the projects of interest that were completed from 1988 to 1992. Thirty projects--the most frequently occurring individual treatments and treatment combinations--were categorized by treatment type and analyzed. ARFs of sufficient statistically significant strength were found for wet-pavement treatments, rumble-strip installations, and shoulder-widening projects. Data for shoulder widening, superelevation correction, and curve correction projects also are presented, but small sample size hampered statistical significance for these projects. Results of the study revealed the importance of improving curve radius during superelevation correction and lane- and/or shoulder-widening treatments. |