Know what’s next.
In nearly every industry, using data and mathematics, coupled with theory and trending, professionals are leveraging the power of prediction to make financial decisions and to guide resource allocation. Today’s decision-makers are held accountable for knowing what data needs to be gathered, the appropriate analysis of the data and the interpretation of the results of that analysis as well as to be able to translate those results into both tactical prediction and strategic forecasting. Decision-making is no longer an educated guess, but a scientific approach on which improvements can continually be made. Thus, as an industry gathers more data and works to perfect the mathematical and analytical approaches to that data, more accurate and useful predictions for decision-making result in more efficient and effective use of resources.
In the field of law enforcement, prediction is not new. Crime analystsc have been making predictions for decades. Using data from a specific crime series (two or more crimes believed to have been committed by the same offender), and applying what is known about criminal behavior, analysts have predicted future crimes down to the day, time and location quite successfully. However, though successful at this type of tactical crime series prediction quite often, it’s certainly not perfect – no analyst has a 100% prediction accuracy rate. In fact, analysts have always had concern over predictions they have made where the date and time passes with no crime occurring, at least as far as can be determined. Were the analytics applied incorrectly? Was data missing? Or was a change in criminal behavior responsible for the outcome? Before predictive analytics became such big news across multiple industries in the past few years, these unsuccessful predictions often discouraged police commanders from deploying based on an analyst’s recommendations. The decision-makers questioned whether the prediction was anything more than an educated guess- anything more than the commander could have predicted himself by studying the crime data and making a logical suggestion for deployment.
However, as the science of crime analysis grew, computers became more powerful and analysts had access to more data sets and the ability to leverage those data sets against the crime incident data being gathered, crime predictions became more scientific, more commonplace and more respected. Having prediction in the news in other fields shed light on the success that could be gained through appropriate data gathering, analysis and interpretation and confidence in the value of prediction rose.
Additionally, access to more data over longer periods of time allowed analysts to move from only making “next (crime) event predictions”, to leveraging the science to make forecasts as to what could be expected strategically over the next days, weeks, months or even years. That is, they use the power of predictive analytics to not only forecast the volume of crime that will happen when, for instance, an empty plot of land is developed into a shopping mall, but also the types of crime that will occur and over the timeline those crimes will take place. Simply put, the crimes that would take place during the construction period attracted different offenders than the crimes that would take place once the mall was open for business. And whereas construction crimes mostly took place after business hours and on weekends, the crimes that could be forecasted once the mall was open for business would take place on weekdays while the mall was filled with workers and shoppers and targets were readily available. Using this information, appropriate staffing and deployments could be implemented with increased nighttime patrols and anti-construction theft strategies in place for the two years of construction and a shift to increased daytime efforts with the mall established to combat vehicle burglaries and thefts and to engage the stores in anti-shoplifting and fraud tactics.
Leveraging predictive analytics in the field of policing has allowed agencies to address active tactical crime series and long-term jurisdictional problems and changes in a meaningful way. As the field has moved from almost entirely “reactive” to more and more “proactive”, predictive analytics have played a key role in preventing crime, apprehending criminals and creating a safer environment for the citizens. Whether predicting a date, time and location that an active offender might strike or quantifying the exact need for additional police personnel to support a rapidly developing area of the city, predictive analytics in policing are playing a bigger role than ever in decision-making and accountability. Using the science to make decisions and allocate resources has been widely embraced and accepted. Removing “good guesses” and “hunches” and replacing them with “statistically likely” and “scientifically supported” is a game-changer in law enforcement.