Computer Modeling Predicts Crashes: Data Help Steer Police To Areas Most Needing Patrol Presence
The Ohio Highway Patrol (OHP) has deployed a cutting-edge statistical modeling system that allows it to predict where and when traffic accidents will occur. The system was developed by the Ohio State University Statistical Consulting Service. “Troopers traditionally have a good idea where crashes have occurred,” says Lt. Rick Zwayer, a spokesman for the patrol. “But this model flags certain areas, times, and even days.” The Ohio State system uses data from crashes that occurred across Ohio from 2001 to 2005 to predict crashes. Among the system’s forecasts are several seemingly obvious findings: that most alcohol-related crashes occur in the evening; the rate of crashes increases in bad weather; and the number of crashes is highest in the afternoon rush hour. The system’s forecasts allow officers to focus their traffic enforcement attention on areas that have high rates of accidents. Data for 2005 fatal crashes is available on the OHP’s Web site and can be accessed by the general public.
Abstracted by the National Law Enforcement and Corrections Technology Center(NLECTC) from the Toledo Blade (OH) (07/09/06); Patch, David .