Analysis of Accident Reduction Potentials of Pavement Markings

Abstract

The major objective of this research was to apply a probabilistic procedure to evaluate the safety effectiveness of pavement marking of undivided rural roads. A 100 sites were randomly selected from the population of improved roads in the state of Indiana. In the analysis, accident reduction factors at the individual site level were estimated as a proportion based on the expected accident rates. To eliminate the effect of regression to the mean, a Bayesian approach was adopted to estimate expected accident rates in the before and after periods. Results of the analysis indicated that the beta distribution adequately fitted accident reduction factor frequencies. Also, accident reduction factors at individual site level were found to be positively correlated with accident rates in the before period. When pavement marking was considered at all sites irrespective of their accident experience, no statistically significant safety impact was observed. However, pavement marking at hazardous sites provided a significant level of accident reduction. Thus, it can be concluded that pavement marking does improve the safety of hazardous sites on rural undivided highways.