Law enforcement agencies across the U.S. are using advanced prevention programs that rely on a smart analysis of historical crime data in neighborhoods across their cities. And they are finding that they can cut burglaries by as much as 27 percent!
City police departments across the country are turning traditional police officers into “data detectives.” Police departments across the country have adapted business techniques – initially developed by retailers, such as Netflix and WalMart, to predict consumer behavior — to predict criminal behavior. A new IBM Center report, by Dr. Jennifer Bachner at Johns Hopkins University, tells compelling stories of the experiences three cities — Santa Cruz, CA; Baltimore County, MD; and Richmond, VA – are having in using predictive policing as a new and effective tool to combat crime. Click here to see more on the study completed by Dr. Bachner.
In 2010, IBM revealed how it aided the Memphis Police Department (MPD) by providing predictive analytics software to reduce serious crime by over 30 percent. Since 2006, Memphis has experienced a 15 percent reduction in violent crimes with the help of IBM’s software. MPD can now assess incident patterns citywide in order to predict criminal hot spots for the proactive allocation of human resources.
Law enforcement agencies such as MDP are employing IBM analytics software to more effectively assign patrols, regulate traffic enforcement and strategically carry out investigations.
NYPD hired IBM Business Consulting Services to help create their Crime Information Warehouse. This facility was designed to provide a single point of information for the entire police department in all five Burroughs. They began by identifying the information elements needed, creating an underlying data model that integrated all the relevant systems. The Universal Database Data Warehouse runs on an IBM System p5 575. IBM’s crime analytics have also proven successful in New York City. Case closings are 25% higher than those nationwide and even though there are now 3,000 fewer officers, crime has dropped 20% since 2002.
Predictive Analytics Reach Beyond Policing
While this report focuses on the use of predictive techniques and tools for preventing crime in local communities, these techniques and tools are being applied to other policy arenas, as well, such as the efforts by the Department of Housing and Urban Development to predict and prevent homelessness, and the Federal Emergency Management Agency’s efforts to identify and mitigate communities vulnerable to natural disasters.
Dr. Bachner says that predictive analytics – in policing or other policy areas – is not a silver bullet. It is an additional tool that supplements existing methods. For example, community policing – where officers build trust and neighborhood-level relations — is an important strategy for deterring crime. “Nevertheless,” she notes, “policing, like many other fields, is undoubtedly moving in a data-driven direction.”
Video surveillance as its used by the great majority of private security services and property owners is a static, after the event record of a security breech that’s already happened, a loss that has already occurred. While of some value to police and security services as forensic evidence, footage of the crime doesn’t return your stolen goods or restore your damaged property.
Integrative security systems based upon real time analysis of ongoing threats and unfolding events turns the static security of traditional passive surveillance into rapid-reaction, proactive security. Computer supported, intelligent video analytics uses real-time image acquisition and modern image analysis to identify crime as it’s unfolding. Using such state-of-the-art tools, countermeasures can be immediately implemented and loss prevention minimized, curtailed or entirely prevented. Command and Control Centers can quickly access and combine local and central video verification tools, and coordinate video dispatch of real- time responders and law enforcement personnel.
Intelligent video analytic capabilities provide fast and essential information. Through real-time information sharing of detailed information, such as license plates and facial images, criminals can be quickly identified. Pre-loss activities flagging the emerging patterns of a crime can be automatically analyzed and loss prevention measures executed.
Analytic software is the key to quickly and proactively analyzing your information and identifying problems. With decision critical video available anytime and anywhere, all the necessary resources can be drawn upon to define and help resolve an emerging crisis. Bomar, IBM and SNAP are on the forefront of creating and employing video analytics for crime detection.