More and more, law enforcement agencies turn to technology to swing the odds in their favor -- and the New York Police Department (NYPD) is no exception.
In the early 1990s, New York City developed CompStat, a method of tracking crime trends. More recently, the NYPD plucked now-CIO Jim Onalfo from retirement to upgrade its computer systems -- and to develop a Real Time Crime Center.
In crimes such as kidnappings, it's crucial for law enforcement to get on the perpetrator's trail quickly -- more often than not, delay means death for the victim. The NYPD's Real Time Crime Center gives investigators the jump-start needed in those critical first 48 hours after the commission of a crime -- when getting the right lead can mean the arrest of a perpetrator before the trail goes cold.
Onalfo said the center, which went live in July, is another step toward making NYPD detectives and officers more efficient because it takes the grunt work out of investigating crimes, and allows detectives to do what they're good at -- forensic analysis and solving crimes.
"It allows us to get information as a detective is going to
a case," he said.
Onalfo said CompStat worked well at looking back at crime trends during the previous week or two. The Real Time Crime Center, however, gets a snapshot of events immediately after they occur. It instantly provides 911 information and GIS maps for detectives, displayed on about 20 Mitsubishi panels that constitute what's known as the data wall.
"We can throw information up on these screens -- if you have a case where you need to have five or six different pieces of information, you could never see it on one computer screen, so it goes perfectly up on the data wall," Onalfo said.
The Real Time Crime Center gives detectives a picture of what's going on in the city, providing a situational overview and allowing police to construct a more coordinated strategy.
"Anything that enhances the decision-makers' situational awareness is beneficial," said Matt Snyder, technology administrator for the International Association of Chiefs of Police. "Anything that removes the clouds on the battlefield is a great thing."
Before the crime center was built, each precinct worked separately, with little or no coordination between them. Now the center gives detectives and other personnel from each precinct an understanding of what the others are doing, via feedback directly from the center.
The crime center funnels critical information from multiple databases into one data warehouse, and combines that information with MapInfo's MapXtreme technology.
Together, the real-time data and GIS maps aid officers in the field by getting detailed information to them quickly. Before the center, detectives collected evidence and analyzed it, which could take days or even weeks because of the difficulty in finding information from several different databases.
The data warehouse now provides immediate access to all those records -- probation and parole records, complaints, 911 call histories, and state and federal crime records -- by searching one database.
"[The detective] would go to the scene, collect information, collect the clues, do all the investigative work. Then he'd go back to the precinct, and through very archaic techniques, would drudge through databases, many of them one at a time," Onalfo said.
Now the investigation begins as soon as the 911 call is completed. The collected information is sent to the center, where 26 detectives begin digging up data about the area. The responding officer will receive data about the incident location via a laptop before he or she even arrives at the scene, including the location's criminal history, whether any known criminals live in the area and the identification of anyone who may fit the description given by the 911 caller.
Detectives in the crime center carry out five key tasks as soon as they receive 911 information:
911/311 data review: analyzing any previous 911 and 311 information that relates to the current call.
Location analysis: reviewing data to garner intelligence on any previous crimes or events at the location.
Victim analysis: developing intelligence on the victim, such as the victim's known associates, enemies, previous addresses or involvement in crimes.
Pattern analysis: reading the data, including GIS maps, to look for trends in crimes that might relate to the current call.
Suspect analysis: using the data to find ties to the suspect -- relatives, tattoos, previous addresses, etc. -- to possibly tie the suspect to other crimes.
"When [the officer] gets there, he now has more information about the scene, which will give him a running start. That's a huge difference," Onalfo said. "First we tell them what's been going on around that location -- here are the parolees or perpetrators who live in the area, here are similar crimes that have occurred in that area. Think of it as a super detective help desk."
The crime center was the brainchild of Police Commissioner Raymond W. Kelly, who hired Onalfo in 2003. "He said, 'I want to help the department do things by taking advantage of technologies that we don't use today,'" Onalfo said of Kelly.
"The first piece was taking the old databases and putting them into a modern database with easy accessibility. So we built the data warehouse. The second piece was to buy and install the data wall, and build the room where the Real Time Crime Center houses the detectives."
The crime center, which cost $11 million, is located directly adjacent to the NYPD's Emergency Operations Center (EOC) in police headquarters. In the case of an emergency, it can be converted from an intelligence gathering and analysis center for the NYPD to a regional EOC, with involvement from multiple agencies.
The ability to place current 911 call information on GIS maps and analyze them in the Real Time Crime Center gives detectives a view of the city they've never had before.
The crime center relies on MapInfo's MapXtreme, a Java-based application that allows the NYPD to visualize crime data on a map to determine relationships and trends that are more difficult to spot in typical paper reports.
"You can put a map out on everything that's happening from a 911 perspective and get an idea of what the situation is within the city," said Jim Karpen, NYPD project manager at MapInfo. "The only way you could do this is with the data warehouse. Previously it was the precincts doing their own thing."
Karpen said the company also provided customized analysis and pattern recognition functionality for the NYPD, which collects point data, aggregates it and depicts the information on a map, illustrating crime hot spots.
"From an infrastructure point of view, it's scalable and intended for Linux-based servers," he continued. "But they're actually running this on a mainframe environment [IBM zSeries], which is very unique for us and for IBM. But it's Linux based. We had to prove to them that our MapXtreme Java technology was open enough to port that."
Karpen said the technology gives users more than GIS mapping.
"We think of GIS as more than creating maps. That's already been done," he said. "What we're trying to bring out with MapXtreme technology is location intelligence. Doing your standard queries and requests for information, and now, instead of just getting the information in report format, bringing in the geographic aspect.
"Police Chief Kelly can get up in the middle of the night, sign in, see what's going on and view the maps if he needs to."