WP's Police Tech


What is Predictive Policing?

In Minority Report precognitive’s used special, extra sensory, abilities to see future events and lead police to future violent criminals before they were able to act. While conjuring up images of Minority Report style police units, predictive policing has no rooms filled with ‘pre-cogs' lying in a pool of water sensing what crimes are about to happen. Instead, predictive policing uses data analysis, much like the business world or insurance industry. Predictive policing is actively being pursued by an increasing number of law enforcement agencies as it is seen as a way of reducing crime and lowering agency costs.

The National Institute of Justice defines predictive policing as - taking data from disparate sources, analyzing them and then using the results to anticipate, prevent and respond more effectively to future crime (( Source - NIJ. )) .

Wal-Mart has used the same method when it came to pre-storm planning for what needed to be stocked in its stores. What Wal-Mart found out was that when severe weather happened they ran short on duct tape, bottled water, and Pop-Tarts. Not just any flavor of Pop-Tarts but strawberry flavored to be precise (( Source - Police Chief Magazine. Pop-Tarts. )) .

Like the Wal-Mart example, law enforcement is analyzing data to forecast where crime will occur and using prepositioned resources to stop or negate crime in a given area. Like Wal-Mart wanting to have Pop-Tarts and duct tape on hand, police agencies hope to gain the upper hand on crime.

The predictive-policing model envisioned by the Los Angeles Police Department and Police Chief William J. Bratton builds on and enhances the promise of Intelligence-Led Policing. With new technology, new business processes, and new algorithms, predictive policing is based on directed, information-based patrol; rapid response supported by fact-based prepositioning of assets; and proactive, intelligence-based tactics, strategy, and policy. The predictive-policing era promises measureable results, including crime reduction; more efficient police agencies; and modern, innovative policing. Predictive policing already has been shown to enable doing more with less, while significantly improving policing outcomes through information-based tactics, strategy, and policy. (( Quote Source - Police Chief Magazine. ))

The data used is pulled from crime statistics, reports, GIS information, weather, community events, health, school information and any other data they have access to. Once the data is gathered is where the ‘predictive’ in predictive policing comes in.  Agencies use predictive analytics to see the patterns and narrow the data down to subsets and areas of concern. An analyst combs through the information to make vetted predictions about where crime is most likely going to occur. The types of crimes in a given area and the types of units needed are based on the patterns discovered.

Predictive policing is a relatively new law enforcement concept that integrates approaches such as cutting-edge crime analysis, crime fighting technology, intelligence-lead policing and more to inform forward thinking crime prevention strategies and tactics. (( Quote Source - DOJ Article. ))

  Predictive policing isn’t new. In-fact, it is a new way to get more out of the data at the agency’s fingertips. In the NIJ article quoted above, Chief Tom Casady, Lincoln Nebraska Police pointed out  -

"Are we doing anything new or innovative with this data or are we just doing it better and quicker?" asked Chief Tom Casady of the Lincoln, Neb., Police Department. Casady argued that the idea is not new. "It is a coalescing of interrelated police strategies and tactics that were already around, like intelligence-led policing and problem solving. This just brings them all under the umbrella of predictive policing," he said. "What is new is the tremendous infusion of data," Casady added (( Source - NIJ. )) .

At the 2010 IACP convention Casady stated - “We’ve got to use our resources more effectively,” Casady explained, “and that means targeting our efforts more intensely on efforts that do not involve simply driving around waiting for something to happen. We’re going to be forced to do more with less, and predictive policing has the potential to help us be more productive and more efficient (( Quote - PoliceOne article here. )) .”   Some argue that predictive policing is redundant, in that it duplicates what officers already know. While the criticism is valid - predictive policing’s underlying tenants aren’t new - what is new is the adoption of methodologies to analyze the data in more ways. Today, crime data is teamed with mapping software to produce a 21st century version of the { Daily Occurrence} sheet, but it mostly confirms what most LAPD cops already know. The city of Los Angeles is patrolled by officers working out of 21 patrol divisions, each of which is further divided into 40 or 50 reporting districts. Every cop in the city knows which divisions see more crime than others, and every cop at each station knows which reporting districts see more crime than others within the division. Any cop transferring from one division to another knows within a week or two where he will be spending most of his time at his new assignment. He doesn’t need a computer to tell him where the trouble is (( Quote Source – Pajamas Media )) .   Predictive policing is getting more information to officers on the street.

"Those of you who have been in policing a while know well what it means when a quarter-million-square-foot big box retail facility goes in on a street corner in your city, for example. You know what that’s going to mean in terms of demand on police services and crimes like theft and fraud and identity theft. You know that because you know exactly how it happened at the other big box retail store across town. So we know a lot — where crime is likely to occur, who is likely to commit those crimes, and who are the people most prone to be victims. With predictive policing, we’re using that knowledge about potential hot spots, crime victims, and the criminals themselves to make deployment decisions that actually prevent crime, not just respond to it (( Quote - PoliceOne article here. )) .”   The influx of analyzed data is of particular help to younger, newer officers. Rather than taking years to gain the knowledge that certain areas have a spike in crimes at certain times or at certain times of the year or during / after  particular events, officers are able to look at the compiled data and see - like a GIS map - where the hot spots are likely to be. “LE can quickly link crimes, locations, dates and time and provide this information directly to officers in the field.  While we are still developing this technology (and also developing the culture to exploit it), it is quickly becoming possible for officer’s in the field to gain remarkable insights into the criminal activity in their area of responsibility.  One reason for LE’s historic deference to seniority is that prior to the advent of this technology it took substantial amounts of time to gain an understanding of crime and crime trends in a given area.  It is now possible for officers to gain the informational aspects of the benefits of seniority (although not necessarily the experiential or relational benefits) in relatively short order.  As younger officers, more adept as a group in exploiting this technology, come of age I would expect this to provide an ever increasing benefit to police agencies.” - Sergeant Greg Stewart, Portland Police Bureau / Crime Analysis Unit   More data is always better, as long as it doesn’t weigh you down or distract. With more data sources, and being able to effectively analyze that data, law enforcement will be able to determine, with a high degree of success, the area where a serial bank robber or rapist will strike next. Take these two examples - Ryan Hughes, Minneapolis Police Department -

Ryan Hughes, data analyst for the Minneapolis Police Department, was tracking a serial bank robber in his city (( Source - Article in the Standard-Examiner. )) . Using free software provided by the NIJ, Hughes and his team made a prediction of where the next robbery would be. The estimate of the robber’s next target turned out to be a mile off. While a mile may seen huge, to a patrol car loitering a mile away is closer than officers who are five miles away.      

George Mohler, Santa Clara University - Another method discussed online by George Mohler, an Assistant Professor of Mathematics and Computer Science at Santa Clara University, is Geographic Profiling. Mohler uses the same mathematical formulas that seismologists use to predict the distribution of aftershocks from an earthquake but applied to predicting criminal occurrences based on the data. Like seismology, Geographic profiling is the problem of estimating the residence (or place of work) of a criminal offender given the locations of crimes committed by the offender.  We have developed an agent-based, Bayesian method for geographic profiling that calculates a prior distribution of residences using housing/population density and a prior distribution of foraging parameters using historic crime data.  The method attempts to take into account how criminals interact with their heterogeneous environment. The figure at right is a geographic profiles for a solved burglary series in Los Angeles.  The circles are the crime locations and the square is where the burglar lived. Mohler believes that police need to start thinking of crimes the way seismologists think of earthquakes and aftershocks (( Source - LA Times article. )) . Using L.A.P.D. burglary data to identify a series of random, initial offenses in a sector of the city and adapting algorithms used to forecast aftershocks, Mohler predicted that 17 percent of the city’s burglaries would occur in a 5-percent area of the city over the next year (( Source - New York Times article. )) .   Like many law enforcement tools and innovations — it could be a huge benefit, if we use it right (( Quote Source - Change.org. )) .   Predictive policing has already been shown to work in practical applications that are saving departments money and time - Richmond, Va - Reducing Random Gunfire in Richmond. Every New Year’s Eve, Richmond, Va., would experience an increase in random gunfire. Police began looking at data gathered over the years, and based on that information, they were able to anticipate the time, location and nature of future incidents. On New Year’s Eve 2003, Richmond police placed officers at those locations to prevent crime and respond more rapidly. The result was a 47 percent decrease in random gunfire and a 246 percent increase in weapons seized. The department saved $15,000 in personnel costs.   Arlington, Tx - Connecting Burglaries and Code Violations in Arlington, Texas. The Arlington, Texas, Police Department used data on residential burglaries to identify hot spots and then compared these locations to areas with code violations. According to Chief Theron Bowman, officers found that every unit increase of physical decay resulted in almost six more residential burglaries in the city. Thus, neighborhoods with greater physical decay could expect greater increases in residential burglaries. Arlington subsequently developed a formula to help identify characteristics of these “fragile neighborhoods.” The police department and other city agencies now work more efficiently in the neighborhoods to help prevent crime (( Examples from DOJ article here. )) .   The Portland Police Bureau is implementing predictive policing as a way to reduce crime. In 2010 some of the bureau’s patrol areas have experienced 18% - 25% in requests for service at a time when budgets are getting tighter. PPB Chief James Craig is creating a crime suppression unit that - will be made up of existing officers, who will use a variety of information to predict when and where crimes are more apt to happen. Craig said that information will include regular analysis of when certain crimes happen, whether it be certain seasons, months, days or hours (( Source - The Forecaster. )) .   According to a report in ChanelWeb, one unnamed Virginia agency has reduced crime by 49%. In Chicago, the number of crime victims has dropped 77% even though the agency has lost 1,300 officers . Outgoing Chief Jody Weis attributed the reduction to “predictive policing through the use of statistics” (( Source - Chicago Now article. )) . “With predictive policing, we have the tools to put cops at the right place at the right time or bring other services to impact crime, and we can do so with less. ” - Chief George Gascón,  San Francisco Police Department   Why just count crime when you can anticipate, prevent and respond more effectively?  The Predictive Policing Model enables public safety and security executive staff to leverage predictive analytics in support of meaningful, information based tactics, strategy and policy decisions in the applied public safety and security environment.  As the public safety and national security communities increasingly are asked to do more with less, the Predictive Policing Model represents an opportunity to prevent crime and respond more effectively, while optimizing increasingly scarce or limited resources, including personnel. The Predictive Policing Model leverages predictive analytics to enable information based approaches to public safety and security tactics, strategy and policy.  Predictive analytics tools, techniques and processes support meaningful exploitation of public safety and security data necessary to turn data into knowledge and guide information based prevention, thwarting, mitigation and response. The ability to identify and characterize threats, and anticipate crime represents a game changing paradigm shift in operational public safety and security.  A fluid, agile force with the ability to use intelligence to guide information based operations can penetrate their adversary’s decision cycle – drug dealer, gang member, counterfeiter, thief or terrorist – affording unique opportunities for prevention, thwarting and information based response.  Ideally, we will be able to prevent crime.  At a minimum, being able to mount information based responses in support of consequence management can significantly mitigate those incidents that do occur; ultimately, changing outcomes for those we serve (( Source - MC2 Solutions, LLC. )) . - MC2 Solutions. A Predictive Policing vendor