Impact of intersection improvements on roadway safety and traffic operations.

摘要

Intersection modifications that improve highway safety have resulted in decreased traffic fatalities. Crash rates could be further reduced at intersections if transportation engineers could base decisions about intersection improvements upon objective engineering factors such as delay and safety. This dissertation explores the feasibility of modeling crash rates as well as traffic delay at urban intersections with use of available exposure measures in the state of Florida. The importance of traffic operations in safety, although generally accepted, has never been well quantified. Current practices for determining the effect of accident reduction caused by changes in the geometry are not adequate. There is a definite need to quantify the role of intersection modifications with respect to traffic operations and safety. In this dissertation study, an attempt was made to show a relationship between crash rates as well as the traffic delay at signalized intersections and the intersection geometrics such as approach volumes, improvements, presence of median, number of lanes, etc. The physical improvements at the intersections considered in this dissertation include addition of right turn lane, left turn lane, and right of way widening.; Data from four-leg signalized intersections in Hillsborough, Manatee, and Sarasota counties were used. Traffic volumes on the approaches are the routinely available exposure measure. It was noted that in these data sets the same values of traffic volume were often “carried over” several intersections. Using such values of traffic volume combined with other variables as measures of exposure results in correlation between errors of the independent variables, which violates the requirements of standard statistical procedures. Crash rate prediction equations were developed utilizing the multiple linear regression analysis techniques. Similar procedure was used to predict the traffic delay. The dissertation used the data obtained from the county authorities before and after improvements to investigate the effect of intersection geometrics on the traffic delay and crash rates at the intersections. The sample size of the intersections analyzed was 151. These intersections were also analyzed for the reduction in crashes and delay. The analyses indicated that the regression models presented are more practical and easy to use for predicting the crash rates as well as delay as a function of intersection geometrics. Also, models were developed by each crash type based on the individual improvements to evaluate the weightage of each improvement at these intersections. In addition to this, the models were also tested for validation.