In California, all cellular 911 calls were initially routed to the California Highway Patrol (CHP), on the theory that most people calling 911 from a cell phone were on the road. But as cell phone use grew, it outstripped the CHP’s ability to handle all the incoming calls — and many of the callers were not, in fact, on a highway.
By 2007, almost three-quarters of wireless 911 calls were still going to the CHP. And 4.9 million cellular 911 calls in California — 42 percent of them — were either ringing busy or dropped. The CHP was forwarding many of the calls that did go through to local law enforcement agencies’ emergency call centers, also called public safety answering points (PSAP).
“Obviously that causes a delay,” said Danita Crombach, communications manager with the Ventura County Sheriff’s Office. “It’s something that’s been a concern for many of us, because we believe that seconds save lives.”
The situation also was frustrating for callers — both the dropped calls and the transfers.
“The public doesn’t understand that wireless 911 behaves differently than land line 911, and they shouldn’t have to understand,” Crombach said. “They have an expectation that when they call 911, it goes to a certain agency.”
The state started looking at ways to solve the problem. One option was to increase the CHP’s capacity to handle the calls, said Kurt Warner, CEO of Public Safety Network, the contractor that worked on the solution the state ultimately chose. But that would have required new call centers, equipment and employees.
The state then tried routing calls for each cell phone tower sector based on what amenities were located within that particular tower’s coverage area. If it was primarily highway, for example, the CHP would get the calls; in a city, the calls would be routed to the local police department. But that still did not provide optimal routing for all calls.
Seeing RED
This laid the groundwork for the Routing on Empirical Data (RED) Project, which collected and analyzed historical data from wireless calls to determine how best to route cellular 911 calls. The project used data from Phase II wireless calls, which have precise location information.
The idea was to use a Web-based GIS to plot the call data on maps that showed which law enforcement agencies were responsible for which areas, then use the results to determine where cellular 911 calls from each sector should go.
“Let’s look at all the calls that have occurred in the past year,” said Don Reich, vice president of Public Safety Network. “We can map those calls, and we see that the vast majority did not come from the highway but from a housing tract adjacent to the highway.” This means the calls should be routed to the local law enforcement agency and not the CHP, even though there’s a highway in the sector.
“We eliminate the transfer from the CHP to the local agency,” Reich said. “That translates into saving lives, improving health and safety, and saving property.”
At first, the project used nine months’ worth of calls to put together the data. “We realized that wasn’t always enough data to make a good decision,” said Michelle Bland, a CHP statewide 911 coordinator. The call history was increased to 12 months. “Having the additional history and larger call volume allows you to make a better decision. Nine months doesn’t cover a full year, and oftentimes there are seasonal changes.” A popular vacation spot, for example, may get more calls in the summer than in the winter.
“Having a year’s worth of data or more is extremely helpful,” Bland said.
Getting the call data was just the first step, though. For the data to be useful, the team had to know whether the call locations were in the CHP’s or a local agency’s jurisdiction. So the RED Project team created maps, called shapefiles, that showed which agencies were responsible for which areas.
Team members then overlaid the call location data on these shapefiles so they could see which jurisdiction was receiving most of the calls from a given sector. The maps also showed major roadways and other features. All this information was made available online to the stakeholders — state authorities, the CHP, county 911 coordinators, county sheriffs’ departments and local police departments.
“Highly detailed call information about every sector is presented [by the system],” Warner said. The stakeholders can log on and voice their opinions about how calls should be routed. In cases where everyone agrees, no meeting or further discussion is necessary.
“In many sectors, it’s extremely cut and dried,” Reich said. “Those calls are very easily agreed to.” The complicated sectors were those where calls were almost evenly split, or where they were going to a variety of places. For example, in some locations calls were divided between the CHP, local police and county sheriff’s office.
In those cases, the agencies discussed their experience with a particular section, Bland said.
“Is it a high-crash site for the CHP? Are there extenuating circumstances that make it better served by a local agency?” Bland said. “Working with the local agency and local coordinator, we come to an agreement.”
On Schedule for Completion
The RED Project started in July 2008 and is expected to be completed in November 2011. The project cost $7.8 million and is paid for by the State Emergency Telephone Number Account, a state fund that pays for the 911 system.
The project is composed of six phases, of which the first three are complete: Ventura and Santa Barbara; the San Francisco Bay Area; and Southern California excluding Los Angeles. The other Northern California counties (except the Bay Area) make up Phase 4 and were completed at the end of June. Phase 5, which covers counties in central California, is slated for later in the summer. The final phase, Los Angeles, is scheduled for completion in November. So far the project is on schedule.
The state worked with 58 county 911 coordinators and eight wireless service providers, said Karen Wong, deputy director of the Public Safety Communications Office for the California Technology Agency.
A few minimal complications have occurred so far in the implementation.
“Basically nothing was automated when we started the project,” Wong said. Paper maps had to be put online. “Moving from a paper world to a nonpaper world had a few hiccups. Some stakeholders were a little concerned about going to an electronic system. But after Phase 1, we were able to show them that actually the PSAPs were able to better manage their calls.”
Being able to see other jurisdictions’ maps is a benefit. “We’ve been able to prove to them that it is much more efficient,” Wong said. “They can actually see what other PSAPs have — they can look at a county level.”
Some local agencies have been reluctant to accept 911 calls directly, fearing that it would increase their call volume, Crombach said.
“Change is hard, and in some instances there was pushback [from local agencies],” said Bernadette Richardson, a project manager for Public Safety Network. “But once the improvements were realized within jurisdictions, the results spoke for themselves and agencies were anxious to be the next in line for optimization.”
Currently California does not require local agencies to accept wireless 911 calls. “The vast majority do take their calls and it is simply a matter of sending a letter to the state 911 office when they are ready to deploy,” Richardson said. “Many agencies chose to coincide their initial deployment of wireless call acceptance with the RED Project phase in their counties for ease and accuracy of implementation.”
Despite the fact that participation wasn’t required, many local officials saw the benefits. “The vast majority of county coordinators and PSAP managers were excited, cooperative and anxious to participate,” Richardson said.
Positive Results
Even with the project only partly done, it has shown results. The 4.9 million unanswered calls per year statewide in 2007 have dropped to 639,000 since the RED Project began. Unanswered calls are expected to decrease further as the project is completed. And this is happening while the total
number of wireless calls continues to grow.
“The CHP statistically has seen a decrease in call volume and a decrease in transfers in the areas where the RED Project has been completed,” Bland said.
And fewer dropped and transferred calls aren’t the only benefit. “Even though we’re only about halfway through this process, we’ve eliminated almost 4 million busy calls in the system,” Warner said. “Right now you’re starting to see an enormous impact. We’re taking a lot of the bad stuff out of
the system.”
The RED Project is nearing its conclusion, and currently there aren’t plans to continue updating the data. The project’s data is just a snapshot — it doesn’t analyze call locations in real time. “Once we finish a phase, it’s done,” Bland said.
However, the project’s results have helped law enforcement officials get needed assistance to wireless 911 callers more quickly.
“Just to know where the caller was literally standing or in a car was really important,” Crombach said. “It allowed us to route cell sectors to appropriate PSAPs. More and more callers are getting to the appropriate law enforcement agency. We’re not having to transfer them around.”
Margaret Steen is based in Los Altos, Calif., and writes frequently about business and management.
