Crime forecasting software ready for market

An innovative software program could soon be helping the military and police accurately forecast where and when crime is likely to occur. The software – called Daily Crime Forecast – uses an algorithm to assess crime incident reports from existing data to pinpoint where and when future crimes are likely to occur.

Edmonton Transit has been testing the software since January 2006, and the results, according to security analyst Stephane Contré, have been impressive. Two years after first being introduced, officer-initiated calls – when an officer initiates a call on his own after seeing an offence or problem – rose 159 per cent, while reactive calls – when an officer is dispatched after a victim or complainant calls for assistance – dropped 52 per cent.

Now, thanks to the assistance of the Edmonton-based Northern Alberta Institute of Technology and its novaNAIT centre for applied research and technology transfer, Contré is poised to commercialize the Daily Crime Forecast .

Late last year, Contré entered and won the inaugural novaNAIT Technology Commercialization Challenge – a competition giving innovators in the Edmonton, Alberta area an opportunity to pitch their technology, product or business ideas to a panel of experts for a chance to win up to $10,000 in novaNAIT services.

Stuart Cullum, novaNAIT executive director, said that while picking Contré from the 35 applicants, including seven other finalists, was difficult, "What we liked about Stephane's software was that it had a very clear and practical application. It was already being used in a policing context and was piloted within that context and it reaped significant results."

novaNAIT offers a wide range of services, Cullum said, including acting as an access point for industry-focused applied research and offering business incubation, product development and commercialization services and supporting applied research in areas including boreal forest reclamation, robotics, simulation, health education, machining and manufacturing, and green chemistry. "In Contré's case, we helped him develop a commercialization strategy. For others, it may be taking an invention from a concept to a prototype through our Prototype Development Program."

For Contré, the commercialization strategy consisted of assessing potential markets for his product, identifying the most promising markets, in this case the military and police services, and offering suggestions for how to break into those markets. novaNAIT, Contré said, both focused his attention on the most promising markets going forward and helped "solidify what our vision should be, what our mission should be, to make this successful."

Contré is now in negotiations to test his crime-forecasting software with a police agency in the United States and with the Canadian military. As well, he's submitting his product to the U.S. military in response to a request for proposals to assist in countering improvised explosive devices. (A reservist with a local military intelligence company, Contré effectively applied the Daily Crime Forecast to roadside bombing situations while on a recent tour in Afghanistan.)

He said the idea for the crime-forecasting software was born nearly a decade ago when he was working as a cop in Ottawa. Officers were given crime hot spot maps that looked at where crime had previously happened. "In a sense," he said, "we were always fighting yesterday's battle instead of anticipating what was likely to occur today."

But it took a few years before the Daily Crime Forecast started to take shape. While working as a security advisor for an oil company in the central African country of Chad between 2003 and 2005, Contré noticed a pattern to the highway robberies workers encountered as they travelled the 200-kilometre route between the drilling site and the company headquarters. "As it worked out, the robberies were highly correlated with the times when local army soldiers were off-duty," he said.

Recognizing that a pattern existed between the time of day and day of week and the crimes led Contré to develop a program that today, after 81 drafts, is two times more predictive than other spatial and temporal programs in use. "It's like fishing with a fish finder," said Tana Vea, an Edmonton Transit training officer who has seen the benefits of using the software within the transit system. "It tells you were things will happen and, because you're there, you see a positive outcome."

"Ultimately," Contré said, "I would like my product to help communities increase the safety and security of their citizens and provide a better quality of life for all by effectively targeting crime."