As tensions have risen in many cities between law enforcement and some citizens, the risk of officer ambush has increased. The National Law Enforcement Officer Memorial Fund reported THAT, “Of the 50 firearms-related fatalities in 2014, fifteen officers were shot and killed in ambush attacks, more than any other circumstance of fatal shootings in 2014.”
Whether these are premeditated or spontaneous ambushes, officers responding to 911 calls could be better prepared with better up-to-date information about where they’re going and what, and who, they might find once they arrive on the scene.
In today’s environment, the data exists to give law enforcement officers this additional layer of protection through better situational awareness. Traditionally, the problem has been critical information is not presented to officers in an integrated way or that’s tailored to the specific officer’s needs at a time when it is critical or while on routine patrol. Nor has it been available in near real-time to provide an officer the latest and most relevant information that could signal potential threats prior to arrival in the area of a call for service.
Integrating multiple data sets can provide an overall risk assessment. That information could include CAD data, geospatial data, the last 24-72 hours of activity in the area — particularly violent crime — records management system information, National Crime Information Center records … and a hot list for that area that includes wanted subjects, stolen vehicles, gang activity, etc.
But integrating all this data, and analyzing it to provide useful information in real-time, can be a significant undertaking. So, what’s the answer?
Enter Event Stream Processing (ESP). ESP analyzes high-velocity big data while it’s in motion, helping a user know what requires action, and what can be ignored. ESP can process millions of events per second from many data streams, including structured and unstructured data from incident reports and other data like CAD, video or social media.
An agency’s IT department can easily define, test and refine event stream models, alleviating the need to hire specialized programmers. Streaming visualization, alerts and notifications support immediate situational awareness so officers can plan for and react to what’s happening in real time to better prepare them for what they are responding to.
Instead of an officer having to request information, it’s routinely provided. An officer responding to a call at an address would get immediate feedback with the latest information on the residence. Are firearms likely to be present? Is there a history of violence with the person or at the location? The information would be prioritized to identify the most critical threats. The officer could drill down into the report to learn more if desired, but the goal is to present the critical information to them with little effort on their part.
Situational awareness is not just useful for officer safety. It can help an officer with their initial reports and investigations. For example, an officer is dispatched to a possible child abduction. The system would give them immediate reports of registered sex offenders with vehicle information in the area. They could already be on the lookout for these vehicles as they proceed to the address.
Perhaps summary-level information (due to legitimate privacy concerns) from Child Protective Services indicates a history of domestic abuse at the address. Certainly, that would assist in the forming of questions that may lead to locating the child sooner — or identifying possible suspects.
Responses to calls from densely populated places like apartment complexes carry additional risk. Even if the person who called 911 is in need of legitimate help, there may be other residents engaged in criminal activity that lead to a confrontation with the responding officers.
With a real-time, integrated system, an officer en-route to an apartment complex could see, for instance, that there are 28 people with misdemeanor warrants for things like marijuana possession and urinating in public; 24 people with histories of non-violent felonies; and 17 violent felons with rap sheets including assault with a deadly weapon, attempted murder and armed robbery.
Before the officer arrives, the officer would also know who among the residents is wanted and who the ones are with a propensity for violence, or what weapons offenses and violent crimes have occurred recently in the building. The system would also provide mugshots or physical description and the apartment numbers of high risk residents so the officer would know who may pose a legitimate threat if they observe them while responding to a call for service. A CAD operator may decide the overall risk assessment of the building is too dangerous, and request additional units respond to provide adequate backup.
NYPD has been on the cutting edge of addressing a situation like the one above. However, officers have to request information on a particular location. It is imperative that the data be as near real time as possible — and updated as quickly as technically possible.
Currently, there are no systems for law enforcement that automatically prepare, analyze and present results in real-time, providing the best most up-to-date information and a clear picture of the situation and area for responding officers. The technology has not been able to handle the number of transactions quickly enough to evaluate and bring the different types of data together, perform analysis and push out results in a timely manner required by law enforcement. This is changing, and new technologies developed for other industries have made this capability a reality.
ESP opens up a world of possibilities. Imagine being able to monitor social media, blogs and news feeds for potential emerging events or issues related to critical infrastructure and key resources and/or criminal precursors to terrorism or an officer assault. By combining geospatial analysis, we could automatically collocate calls for service with IP addresses of social media threats. Maybe a person in the area of a 911 call has tweeted their intention to commit suicide by cop.
This information could be passed to the responding officers immediately and dynamically without them having to make a request to search social media while responding to a call of a man with a gun. Normally, this information would be discovered after the suspect was fatally shot and their identity is determined and social media sites are checked.
Other possibilities include monitoring license plate reader (LPR) data feeds. Following an armed robbery, the data from LPRs in the vicinity could automatically be sent immediately to nearby officers, including plate numbers, types of automobile and descriptions of vehicles departing the area.
Or, automatic vehicle locator data for cars that have been reported stolen or are owned by wanted felons could generate alerts for officers nearby and provide possible locations of these wanted subjects.
This tremendous amount of disparate data represents an incredible opportunity to improve situational awareness, protect officers and reduce crime. But it takes high-powered processing, data integration and analytical software to generate the reports and alerts officers need.
The use of available data is unending and it’s just a matter of determining what is important and can be made available to officers, commanders and anyone responsible for the safety of the public and each other.
Providing better and more accurate and real time information to officers is something that is a critical element of policing — now and into the future.
Law enforcement over the last five to 10 years has embraced the use of data for predicting criminal activity, allocating resources to reduce crime rates and building intelligence to better identify those individuals and organizations that are responsible for the majority of crimes in a specific jurisdiction.
It is time to provide responding officers with information that allows them to make better informed decisions and provide for their own safety, while at the same time reducing crime and making our communities safer.Dale
Peet is a 23-year veteran of the Michigan State Police and the retired commander of the Michigan Intelligence Operations Center, Michigan’s largest and primary fusion center for homeland security. He now serves as a principal law enforcement consultant at SAS. He previously wrote, Combatting Human Trafficking Through Early Victim Identification, Disruption of Criminal Networks, Social Media Analytics Can Aid Law Enforcement, First Responders in Civil Unrest, Disasters, Investigations, Predictive Analytics and Geospatial Tools for Crime Forecasting, Building a Better, More Effective Fusion Center, and, Fusion Centers and Suspicious Activity Reports – Best Practices.
