The New Jersey Department of Transportation (NJDOT) is enforcing its access management standards within the limits of most of its highway improvement projects. In order to identify the impact of access points on accident rates, a study was conducted that concentrated on NJ State highways Rt. 27, 28, 33, and 35. The study concentrated primarily on the impact of major geometric and traffic flow characteristics on accident rates at the macroscopic level. The principal variables that were taken into consideration included the Annual Average Daily Traffic (AADT), number of lanes, shoulder existence, median existence, speed limit, and access points per mile (access density). The analysis included a comparative analysis between the occurrence of accidents at signalized intersections (intersection-accidents) and between intersections (section-accidents). In addition, a field study was conducted on one NJ state highway section where the main objective was to provide insights into the microscopic traffic flow characteristics such as the speed profile along the roadway, and the impact of turns from/to access points on the speed of the vehicles traveling on the mainline. The principal conclusions are: approximately 30% of the total accidents on the study routes were reported to occur between signalized intersections, approximately 7% of the accidents were attributed to vehicle maneuvers from/to access points, accident rates are better represented by a log-normal distribution, access density is a contributing factor to the occurrence of accidents between signalized intersections although not a necessary one, the highest percentage of accidents occurred during the evening peak from 3:00 to 6:00 PM, driver inattention is the primary factor in accident occurrence for both the section (37%) and signalized intersection accidents (33%), regression models were developed for 4-lane highway with shoulder, 2-lane highway without shoulder, and 4-lane highway with median that included a combination of the AADT and access density and/or the speed limit as independent variables. A more comprehensive statistical analysis of all NJ state highways and the US is recommended that could identify similarities and differences among various types of highways. This analysis should include sub-hourly traffic flow rates, the distribution of the speed and more detailed geometric characteristics. The development of a detailed GIS based database that would include accidents, traffic flow, geometric and weather characteristics would provide a universal basis for conducting similar analyses. A microscopic simulation model is needed, which can capture the traffic flow characteristics of multilane highways/arterials, especially capturing the impact of access points. (Author/publisher)
Abstract