FAA Revamps Air Traffic Systems

At the Federal Aviation Administration, in Washington, Steve Dash, manager of the Air Traffic Control Telecommunications Services Office, became frustrated with the agencys traditional pay-for-service relationships with contractors.

While vendors could fulfill service requests, Dash said he thought FAA agreements never induced contractors to think about the administrations long-term welfare.

"The vendor would be more than happy to deliver services wherever we wanted them, whether they made sense or not," Dash said.

Its a well-known dilemma. You cant rely on the contractor to tell you when and if youre making a mistake.

When it comes to determining what services to request, Dash said, "the burden is on the government to make the smart decision."

Dash wanted to break the cycle. The FAA had a huge project coming up, and he wanted a vendor the FAA could trust: a true partner with interests that could be aligned with the FAAs.

In late summer 1999, the FAA realized a mission-critical need to increase the availability of telecommunications for air traffic controllers while standardizing equipment spread across 5,000 locations in North America, the Caribbean and parts of Europe, Dash said.

This FAA Telecommunications Infrastructure, or FTI, project underwent another revision after Sept. 11, 2001.

Due to increased security concerns, National Airspace System services would also be physically isolated from the administrative and the mission-support service domains to greatly reduce potential risks and vulnerabilities, explained Tammy Jones, spokesperson for the FAA.

When he looked at the FAAs entire network environment, Dash said, he saw a troubling mixture of leased and FAA-owned legacy networks and subnetworks that his staff was trying to design and manage.

Dash said he believed a contractor could manage it more efficiently, freeing up valuable FAA resources. Systems Management

"The business case is we could do more and better for less," Dash said.

In July 2000, the FAA put the FTI project out to bid and issued a revision in December 2001 that reflected the increased security concerns sparked by 9/11. The RFP (request for proposal) stated a desire to see a business case that would make it conducive for the FAA to form a strong partnership with a contractor.

It was an odd request, according to Bob Coulson, senior business development manager for Harris Corp., an international communications equipment company headquartered in Melbourne, Fla.

Coulsons colleague John OSullivan agreed. "How do you establish a partnership in an arms-length environment?" asked the vice president and program manager for the FTI project, as he and his team were in the throes of putting together a bid to win the FTI contract.

Coulson said the FAAs RFP showed an obvious need for a 100 percent technologically solid, low-risk solution. However, the agency also wanted an economical service.

One way to save money was to not overprovision availability services for an entire location but instead deliver services based on each user need within that location.

Dash explained that, historically at the FAA, "there was a pattern that a facility type drove the solution rather than the specific user within the facility, and it resulted very often with some users within a facility having a higher level of availability than [what was] required." Being able to create that user-by-user provisional flexibility was a significant procedural change for the FAA, said Dash.

After much deliberation, Harris introduced a proposal unlike traditional contractor or integrator relationships, said Coulson. Both Dash and Coulson described the agreement as one they hoped would fulfill requirements of solid technical solutions and economical decisions. Instead of a pay-per-service relationship, Harris revenue would be determined by the quality of its decisions and services. Simply put, the FAA would pay Harris based on the companys performance, Dash said.

Under the contract granted in July 2002, the FAA agreed to reward Harris based on "measures of effectiveness." Harris amount of pay would depend on operational metrics such as network availability, latency, security, help desk performance, and order handling and processing, said Coulson.

"To be a true partner with the FAA here, our vested business interest always has to be about enhancing the services we provide to the FAA," said OSullivan.

As an example, OSullivan said the most notable improvement to the FAAs communications network has been the upgrade of the IP backbone.

Click here to read about the disappearance of telco pricing information.

According to Mark Graham, FTI chief systems engineer for Harris, the new equipment and installation has greatly reduced latency and improved throughput.

"Were evaluated every six months. If the IP backbone continues to exceed service-level agreements in terms of latency and reliability, then that would show up in our rewards fee," said Graham.

Building the entire communications infrastructure for the FAA requires a lot of manpower. Internally, Harris has 750 people devoted to the project, with at least 250 more performing installations via telecommunication partners.

The FTI project is far more encompassing than the previous VSCS (Voice Switching and Control System) project. "[Unlike VSCS], FTI spoke to not just voice but also radar surveillance, navigation systems, weather systems, landing systems and everything else that makes an air traffic control system work," said Graham. "[But] FTI is very much like VSCS in that we have redundancy on top of redundancy."

Upon completion of the FTI project, the FAAs voice and data communications environment will have so many levels of redundancy that it is expected to be able to withstand any technical failure—from a burnt-out chip on a circuit board to complete loss of service from a long-distance provider, said Coulson.

In June, Harris and the FAA finished Phase 1 of the FTI project by upgrading 27 major nodes at major facility locations. These nodes are significant hubs, such as large air traffic control centers that connect to many other locations.

Click here to read about concerns about the nations data theft laws.

Completion of this phase has set the stage for the team to complete Phase 2 by December 2007. That phase calls for upgrading the legacy networks at the remaining 4,000 locations, which could be anything from radar on top of a hill to a nondirectional beacon in a cornfield to an isolated radio site off the coast of Alaska.

Harris is more than happy to be the one opening up these new opportunities, Coulson said. "Our rewards depend on how well we help the government perform their job," he said. "Their profit means our profit. Their goals mean we do better."

David Spark is a free-lance writer based in San Francisco. You can contact him at david@davidspark.com.

Check out eWEEK.coms for the latest news, views and analysis on servers, switches and networking protocols for the enterprise and small businesses.