Esri Blog

Minneapolis Police Target Gun Violence Using Location Analytics

In an effort to reduce citywide crime rates, the Minneapolis Police Department (MPD) is narrowing incidents of crime to specific neighborhoods and even streets that need added attention.

One example of MPD’s geographic analysis involves violent street corner gangs. Over the last five years, traditional gangs in Minneapolis have been replaced by small, splintered street-corner cliques. These cliques at first appeared to be random, but over time, patterns of crime were identified as place-based behaviors that could be anticipated and predicted.

“Traditional gang offenders age out of street crime and gun violence with relative statistical certainty,” says Sergeant Jeff Egge, MPD Strategic Analysis Unit supervisor and inductee into the Center for Evidence-Based Crime Policy Hall of Fame. “The trend has been toward hybridization of gangs and loyalties.”

The new cliques center on specific streets that have evolved into retaliation zones. MPD uses a geographic information system (GIS) to identify and monitor these high-risk crime zones for deterrence efforts and prevention.

“The geospatial rendering of our data makes our work more efficient and valued by the organization by bringing life to crime trends and patterns,” Egge said. “With effective and advanced data and analytics, we began to understand more about what’s driving crime in these chronic crime places.”

Mapping Criminal Activity

An early pioneer in evidence-based policing, the MPD has a strong culture of evaluating how and where to fight crime. MPD’s Strategic Analysis Unit was created after the city completed 1,000 weeks of CompStat evidence-based policing and identified the need for relentless follow up and assessment to deliver effective tactics.

Since 2007, the Strategic Analysis Unit has worked to combine intelligence links, crime identifying factors, and tactical data—working together to give context to the city’s crimes. Intelligence for analysis are gathered from officers on patrol, group intervention teams, community engagement personnel, public health, county, state, and federal partners. Each precinct has an analyst who maps incoming data to uncover crime patterns alongside corresponding police efforts.

A number of factors in locations such as commercial corridors, bus stops, convenience stores, problem properties, vacant lots and group reentry houses provide opportunity magnets for high-risk offenders.

“Mapping brings to life the dynamics that facilitate the opportunity structures of crime,” Egge said. “Tactical maps guide intervention to change the trajectory of criminal activity.”

Depicting gun violence in 3D is one way to creatively illustrate concentrations, inspire stakeholders and foster commitment. This example was inspired by Dr. Lawrence Sherman, one of the researchers who worked with Minneapolis to better understand hot spots.
Precinct Five violent crime analysis
Analysts regularly update precinct maps, like this one of the Fifth Precinct, to drill in and persist analysis of hot spots that experience the majority of the city's violent crime.

Addressing Gun Violence

Tactical interventions have produced results for most crime, however, gun violence remains a concern in specific areas.

Minneapolis Police Chief Medaria Arradondo tells the Star Tribune that while police do a good job of apprehending perpetrators of gun violence, more could be done in crime prevention.

Looking more closely at gun violence, the Strategic Analysis Unit found 386 blocks with 10 or more shootings over 25 years and 31 blocks with at least one shooting per year.

“Since gun violence has remained so steady on certain blocks for a quarter century, there needs to be a tailored ‘Case of Places’ approach to make an impact,” Egge said. “We have worked with our criminologists to understand the social problems around the hot spots and have developed tailored enforcement and prevention strategies that translate this research into practice.”

The Strategic Analysis Unit created a dashboard to display data about gun violence, including real-time alerts of gunshots collected from an array of acoustic sensors provided by ShotSpotter, an Esri partner. The sensors detect the sound of gunshots and triangulate the location of where shots were fired. These locations are displayed on the dashboard alongside details from 9-1-1 calls to create a central place for MPD to view data and formulate a shared understanding of events.

“When Shotspotter was first installed it didn’t include the triangulation that pinpoints where guns were fired, which aids the geospatial analysis of gun crime hotspots,” Egge said. “Places like, GVR, Little Earth and Penn were off their rockers and the activity pushed us to make an update.”

The Strategic Analysis Unit also developed a public-facing Shots Fired Map using Esri’s ArcGIS Online platform to report and display gun violence to citizens.

“The map has been a valuable tool to help stakeholders and community partners visualize this problem,” Egge said.

The interactive Shot Fired Map provides transparency for citizens and local leaders on problem spots, including the ability to report incidents on the map.

Crafting a Predictive Strategy

The MPD is now boosting its crime-fighting efforts with the use of maps and location intelligence to track and prevent crime—an approach known as predictive policing.

To conduct predictive policing, MPD aimed to address high-risk offenders by examining the associative nature of people and locations of reported crime and arrests. To improve the consistency and predictability of the mapping, and to predict the possible locations of violent crime, analysts compared the maps to locations of problem addresses, parks, street corners, and top 9-1-1 call locations. Scott Wolfert, a crime analyst with the city of Minneapolis, describes this as the “geospatial analysis of crime concentration.”

“Crime forecasting and predicting patterns of crime is not without risk,” Egge said. “Tactics must be undertaken with consent of the taxpaying public.”

Analysts built a predictive model of where they expect gun violence to increase by pulling together three layers of data: places suspects discharged weapons; places suspects were more prone to be arrested with guns; and places where victims were shot or shot at.

Crime occurs when a motivated offender meets a victim at a location lacking suitable guardianship. This is known as the crime triangle based upon the routine activity theory developed by Marcus Felson and Lawrence E. Cohen in 1979.

“Forecasting crime is a product of the crime triangle multiplied by factors and fractions known to facilitate crime over time such as poverty, trauma, weather, and fear of apprehension,” Wolfert said.

Attacking an Anomaly

In Minneapolis violence generally increases in warmer weather, with crime incrementally creeping upward from early March until late July or early August. For example, a rash of youth violence across the Twin Cities on Mar. 1, 2019, sparked a heightened sense of concern.

“We had three shootings in Minneapolis and then one in St. Paul all within the same 24-hour period, and then it stopped,” Wolfert said. “That was very different than traditional violence, especially gang violence, so it requires a different analysis.”

The investigation continues on the incidents of March 1. Analysts are working to detect gun violence spikes and keeping an eye on gang retaliation zones that may see ripple effects from these shootings. Any new similar incident will trigger a more urgent predictive methodology to examine and identify criminal networks, evidentiary links, motives, and geospatial patterns involving high-risk people and locations. Analysts work closely with officers in the field to learn from all incidents and fine tune tactics.

Analysts localize crimes to micro hotspot grids that can be monitored by a single patrol car. Once probable patterns have been isolated into manageable one-, two-, or three-block areas, a report is produced for patrol supervisors that includes a map, recommended times and locations for intervention, problem persons in the geographic area, and other facts to improve an officer’s situational awareness.

The analyst team reviews data for different time periods, like a rolling 14-day or one-year period, looking at things like the number of ShotSpotter activations or the total number of 9-1-1 calls coming in. The data is then used to adjust policies and distribute resources, such as police coverage in the field.

“An overall decrease of 40 percent in Part I crime (murder, nonnegligent homicide, rape, robbery, aggravated assault, burglary, motor vehicle theft, larceny-theft, and arson) since 2000 is a credit to many of the strategies employed during peak times of violence,” Egge said.


Learn how law enforcement maximize the power of location intelligence, download an eBook on Data-Driven Policing.

About the author

John Beck is the Director of Law Enforcement Solutions at Esri where he is responsible for helping police agencies worldwide understand and implement GIS for every mission. Before joining Esri, Beck was a police officer and crime analyst in Nevada. In his role at Esri, he helps police agencies apply GIS to crime analysis, strategic planning, patrol operations, investigative support, and citizen engagement. John has also worked with agencies to implement GIS for open and transparent policing and to tackle hard problems like the opioid epidemic and homelessness. He is also helping police realize the value of new geospatial technologies including machine learning and big data analysis to gain a real-time understand of crime patterns and make better informed decisions. John earned undergraduate degrees in geography and anthropology and a master’s degree in criminal justice from the University of Nevada.

Next Article

Relief Workers Rely on Drone Imagery to Help Bangladesh Refugee Camp

Read this article