Innovative Program Identifies Auto-Theft Patterns

December 7, 2004 — An auto-theft recovery program developed through the UCF Public Safety Technology Center (PSTC) is identifying theft patterns that enable law enforcement officials to increase patrols in high-theft areas and drop-off locations. The program may significantly reduce both auto-theft crimes and the cost to insurance companies for compensating owners of stolen vehicles.

“Every 24 seconds a motor vehicle is stolen in the United States,” said criminal justice and legal studies Associate Professor K. Mike Reynolds, who also serves as the director of the PSTC. “Law enforcement officers work daily to locate and retrieve stolen cars, while insurance companies spend billions of dollars each year compensating auto owners for vehicles that are never recovered.”

With combined resources from the PSTC, State Farm Insurance, and Orange County Sheriff’s Office (OCSO), an artificial intelligence clustering algorithm program was created by Olcay Kursun, a UCF postdoctoral fellow in computer science. Using a database of recent auto-theft information provided by the OCSO, the program is able to identify and use patterns to predict the drop date and the drop location of stolen cars. The program produces cluster analysis, which can be visualized as a series of dots on a map and shows where theft has previously occurred.

Law enforcement officers are currently testing the program by using it to identify day-to-day trends in auto theft locations, daily hot spots of theft and preferred-drop locations. By pinpointing popular “dump sites,” officers are able to stake out a site and attempt to recover the vehicle and apprehend and arrest the criminal.

Assigning law enforcement to common locations allows patrol officers to increase patrols in high theft areas and assist in crime trend reduction strategies. Information produced by the program will be transferred into a daily report that is to be transmitted to the OCSO for use by the auto-theft detectives and auto-trap unit.

According to public affairs doctoral student and former OCSO staff member Joe Saviak, “Certain types of car thieves follow patterns. If you can identify and predict those patterns, you can catch the thief and recover the vehicle."

This project was funded by a $12,000 grant from State Farm Insurance. Along with the police data-sharing system, the auto-theft recovery program is housed in the Public Safety Technology Center located on the UCF campus.