This report presents the findings and recommendations based on the Identification of Abnormal Accident Patterns at Intersections. This project used a statistically valid sampling method to determine whether a specific intersection has an abnormally high number of accidents. The department researcher located intersections throughout the state and categorised them by geometric type, stop control type, and traffic volume. A sample of each intersection category was taken and co-ordinates for each intersection were found. Accident reports were obtained for the sampled intersections, and the data was entered into a spreadsheet for analysis. The mean and 90th and 95th percentile were calculated, and the expected value analysis tables were created for each category of intersections. A total of fourteen (14) expected value analysis tables were produced for the various intersection types and compared to those received from other states. Recommendations were suggested for identifying abnormal accident patterns based on information that was received from other state Department of Transportation’s, local agencies, and the South Dakota Expected Value Analysis Tables. (Author/publisher)
Abstract