How PredPol Works: We Provide Guidance on Where and When to Patrol
Using only three data points – crime type, crime location and crime date/time – PredPol’s powerful software provides each law enforcement agency with customized crime predictions for the places and times that crimes are most likely to occur. PredPol pinpoints small areas, depicted in 500 feet by 500 feet boxes on maps – that are automatically generated for each shift of each day.
The algorithms used by PredPol have been published and discussed publicly in peer-reviewed papers. They are based on the observation that certain crime types tend to cluster in time and space. PredPol uses self-exciting point process models to replicate this behavior (Click Self-Exciting Point Process Modeling of Crime).
PredPol takes a feed from each department’s Records Management System (RMS) to collect crime type, location and date/time. This data is collected at least daily and feeds our prediction engine, which is run once a day to create predictions for each beat, shift and mission type.
We initially process several years of data to lay down a “background” level of crime patterns and to understand how crimes propagate throughout the city. This is done using an Epidemic Type Aftershock Sequence (ETAS) Model, which is a self-learning algorithm.
As new crimes come in, they are mapped against existing patterns and events in the city. Based on the propagation patterns uncovered by the initial analysis of the data, we predict when and where similar crimes related to these crimes are most likely to occur.
Every 6 months, we force a “re-learning” of the patterns using all historical and recent crime data. This ensures that new patterns of behavior are picked up by the system as well.
“PredPol does not replace the experience and intuition of our great officers, but is rather an invaluable added tool that allows our police force to use their patrol time more efficiently and helps stop crime before it happens.” Chief Mark Yokoyama (SVG Journal)
Supporting Community Policing
Typically, patrol officers spend a few minutes in PredPol boxes to deter crime when they’re not responding to calls for service or performing other duties. Some of these boxes are places that officers routinely cover, but many others are places that might not otherwise receive attention. This also provides officers an opportunity to interact with residents, aiding in relationship building and strengthening community ties.
Empowering Crime Analysts
Using PredPol’s technology, analysts spend less time making crime maps and more time conducting the analysis and intelligence work for which they are trained. Newer officers are brought up to speed faster in unfamiliar patrol areas, and veteran officers are given additional, high-risk places to patrol that take them beyond their usual target areas.
Law enforcement agencies deploying PredPol are experiencing marked drops in crime due to increased police presence in areas deemed to be at greatest risk.
PredPol goes beyond predictive policing with Command Analytics. Command Analytics is a suite of high-level features designed for command staff, watch/shift supervisors, detectives and crime analysts. Command Analytics is designed to provide quick and simple access to data from records management systems (RMSs), GPS tracking systems, and other departmental data stores.
Our suite offers real time analysis tools for decision making, resource management, and planning through the following features:
- Crime search
- Mission Control
- GPS + Vehicle Pathing
“We told [officers] to go into the boxes and use their knowledge, skills and experience to determine what should be done…They may stay there for just 15 minutes to a half-hour and let people see them walking around the area…Would-be offenders see the police activity and are deterred from committing a crime there. All we are trying to do is deny them the opportunity to commit that crime in that time and place. During our test, we probably disrupted criminal activity eight to 10 times a week….” LAPD Foothill Division Captain Sean Malinowski (Officer.com, January 1, 2013)