- Media miss the point re the "lone controller" in Lexington:
Underlying staffing problems bear the union label
- New ATC report holds lessons for U.S. ATC reform:
Scholar says Canada and U.K. point the way
- Alarming growth in controller errors:
Could lenient training regime be at fault?
- FAA to accelerate ADS-B implementation
But cost savings still too far in the future
- News Notes:
ATCA, NATCA, AOPA, Reason Foundation
- Quotable Quote
A tragic case of pilot error that led to a number of deaths when a regional jet took off on the wrong runway has somehow morphed into a tale of control towers that are short-staffed on the midnight shift. Trying to stave off congressional intervention and get the media off its back, the Federal Aviation Administration quickly added a second controller to the midnight shift in the Lexington, KY airport control tower. That's the last thing the airport needed.
First, note the built-in media assumption that it's the controller's job, after clearing the plane for the correct runway, to watch over it like a mother hen to make sure it does what it's been told to do. Not so�never has been.
Second are the cries of alarm over there being only one controller on the job during the midnight shift, when there is little or no flight activity. You'd never know from most media reports that this practice is very common, and not just at little airports like Lexington. There's only one controller in the tower overnight at Ft. Lauderdale, which has an order of magnitude more airline activity than Lexington. According to the Official Airline Guide's latest issue, exactly seven flights land at FLL after midnight, the latest one at 12:35 AM. After that, nada.
As for the sleep-depriving shift schedule, a former controller and manager tells me that controllers used to work alternating weeks of five day-shifts and five evening shifts, with an occasional week of midnight shifts. That type of shift schedule is much easier, physiologically, than the common practice today of two evening shifts, two day shifts, and a midnight shift on the last day of the work week after just nine hours off. That gives the controller a very long weekend—e.g., from 6:30 AM Friday till 4:00 PM Monday. The controllers' union fought for this schedule, and the FAA long ago gave in.
Finally, how would a businesslike ATC enterprise deal with a situation of practically no flight activity at smaller towers? By consolidating the function into a few, larger facilities. A tower controller in upstate New York (close to retirement) tells me the following:
"Last night while working the 2-10 shift, we had two positions open in the radar room solely because we had plenty of people. But in two separate hours of watching the scope, I worked a total of eight airplanes, two in one hour and six in the other. Why not combine our facility with [X], close that place down, and even consider making one major upstate facility, say in Rochester or Syracuse, to handle all of the radar operations upstate? You could reduce the radar staff by half, and even incorporate airspace up to 17,000 feet, helping to reduce problems of staffing at the centers . . . . This simple idea could be mimicked around the country, but the union will not listen to reduced staffing. I'm basically committing treason when I mention it in the tower."
Yet the media swallow the union press releases about serious staffing shortages at towers around the country.
The IBM Center for The Business of Government has published a report that should be required reading for everyone who will be involved in next year's debates over reauthorization of the FAA. "Reforming the Federal Aviation Administration: Lessons from Canada and the United Kingdom" reviews the problems that have plagued FAA for decades, as well as the recommendations for reform made repeatedly over the past 20 years by a series of special commissions. While it acknowledges the creation of the Air Traffic Organization within FAA, two years ago, as a step forward, it goes into some detail on the fundamental problems that remain to be solved.
In brief, those three problems (no surprise to readers of this newsletter) are:
- "Air traffic control funding has a fundamental disconnect between the factors that drive the costs of providing the services and the factors that drive the revenues used to provide those services.
- "Air traffic control modernization programs continue to be hampered by the poor performance and high costs of [FAA] capital investment programs.
- "Air traffic control lacks organizational independence, which prevents resources from being used in the most effective ways and which also results in self-regulation of the air traffic control system."
After discussing these problems in some detail, as well as recounting the mostly ignored recommendations of expert bodies like the Mineta Commission, the report then explains how similar long-standing problems have been addressed in Canada and the United Kingdom, via fundamental institutional and funding reform of their ATC systems. While both Nav Canada and the UK's National Air Traffic Services have had their growing pains, the report shows how organizational transformation has occurred, thanks to the financial and managerial independence that has been granted to both entities.
This powerful report was written by Clint Oster of Indiana University and John Strong of the College of William and Mary. Oster is one of the most respected academic experts on aviation. He's been on numerous commissions and expert panels for entities like the National Academy of Sciences and the Government Accountability Office. He was the research director for the Aviation Safety Commission of 1987-88.
Fortunately, the IBM Center has posted the report on-line here. Download it today and take a look.
The April-June issue of The Journal of Air Traffic Control includes a very disturbing article. "ATC Certification and Quality Assurance" discusses how controller training changed after the rebuilding of the workforce following the 1981 strike. And it suggests that a decline in training standards and rigor may lie behind the rise in both runway incursions and operational errors during the past decade.
Before the strike, training took place in phases, each of which was pass/fail—i.e., trainees knew that if they didn't measure up at each milestone, their would-be career as a controller was over. It also took a long time, building up skill levels with carefully supervised on-the-job training with the end goal of being certified as a Full Performance Level (FPL) controller. Those who failed at that stage either left the service or were assigned to a Flight Service Station (which does not control traffic) or one of the lowest activity-level towers.
Something had to change after the strike, in order to rebuild the workforce as quickly as it was possible to do safely. Trainees no longer had to have aviation experience, but the pass/fail system was retained, even though on-the-job training was accelerated such that controllers reached FPL in about half the time. Pre-strike staffing was finally reached by the early 1990s.
At that point, recounts author Ed Drury (a retired air traffic manager with 40 years' experience), FAA headquarters decided that it was not cost-effective to put lots of resources into candidates who ultimately washed out. So they abolished the pass/fail system and created a new process, called Train-to-Succeed. Trainees could progress at their own pace without fear of washing out, leading some to term the program "train �til you die." Drury acknowledges that "the vast majority of new recruits" eventually did OK. But he also discusses stories about "trainees being reassigned from one supervisor to the next, until a supervisor was found who would sign off on certification." But also, "there was a cadre of trainees, throughout the system, which no one would certify, and they were ultimately reassigned to lower activity towers."
But it gets worse. When FAA began, in the mid-1990s, to contract out its lowest-activity towers, the new union (NATCA) did not want its members working for private companies. It negotiated with FAA to have those controllers train to certify at higher-activity facilities under Train-to-Succeed. "Some controllers even returned to facilities where they had previously failed."
It's at this point that Drury presents the data on operational errors (made by controllers) and runway incursions caused by controller operational errors. To provide some context, he also lists the total number of instrument flight rules (IFR) operations each year, as a basic measure of overall controller workload. I converted all of these numbers to rates per million IFR operations, which makes the trends easier to see. In the first half of the 1990s, OEs ran at about 12 to 13 per million operations. But for the 2000-2004 period, the rate leaped to the 16 to 18 level. The runway incursion data don't go back as far, but also show an upward trend, averaging 1.2 from 1997-2000 but 1.37 from 2001-2004. And mind you, during the last five years or so, FAA has focused increasing attention on reducing both operational errors and runway incursions, obviously without much success. Drury's view is that "the certification and performance standards resulting from Train-to-Succeed are at the heart of this problem."
I'll leave you with Drury's closing sentence. "As the FAA proceeds forward with plans to hire and train the next generation of Air Traffic Controllers, this may be the perfect opportunity to take stock of what works and what doesn't."
In a series of bold announcements this spring and summer, the FAA has made a major commitment to the technology that many consider one of the key building blocks of a network-centric system that should replace today's labor-intensive, radar-based air traffic control. Automatic Dependent Surveillance-Broadcast (ADS-B), as discussed previously in this newsletter, would replace today's airborne transponders (required on all planes in controlled airspace) with a box that broadcasts the plane's GPS location, rather than just its identity and altitude. It would also be able to receive and display both traffic information and real-time weather information on a cockpit display.
In May the agency unveiled an overall schedule, as well as a bold departure in contracting. Rather than defining detailed hardware specifications and purchasing the ground stations from the winning bidder, instead the agency will issue performance specs and enter into a 15-year service contract, under which the winning bidder will develop, produce, install, and operate the ADS-B ground stations, providing the information services to all equipped aircraft. This would likely be a much larger outsourcing than FAA's 2005 contract with Lockheed Martin to operate and modernize its Flight Service Stations.
In addition to facilitating a shift to GPS-based navigation, ADS-B (and the rest of the network-centric approach) offers the promise of permitting the retirement of thousands of obsolescent and expensive-to-maintain ground-based systems, especially radars. But for those cost savings to be realized, all aircraft in controlled airspace must be equipped with ADS-B, so that they show up in the system. Here is where the FAA's plan disappoints. There is no requirement or target date for equipage, and the proposed four-segment deployment schedule for the ground stations would not complete the removal of "legacy" ground equipment until FY 2023.
One very knowledgeable reader of this newsletter suggested recently that the FAA make equipage mandatory for all aircraft produced in 2009 and later. Since all new Airbus and Boeing airliners have come equipped for some time now, and with large fleet-replacement orders continuing, he expects such a mandate would lead to the majority of the airline fleet being equipped by 2014. Turnover in the business jet fleet is lower, but such a requirement would help there, as well. In addition, of course, if ATC funding is shifted from user taxes paid to the Treasury to user fees paid directly to the Air Traffic Organization, then the ATO could offer discounts on those fees to equipped planes, as both Eurocontrol and Nav Canada already do.
That still leaves the problem of non-jet general aviation. Even the Aircraft Owners & Pilots Association (a big supporter of ADS-B) acknowledges that mandatory equipage is inevitable, at some point. But they want this put off long enough so that the on-board equipment comes down the learning curve in price to the equivalent of today's transponders. Very large-volume production might lead to such pricing, and Australia's commercialized air navigation service provider actually issued an RFP along these lines earlier this year�but rescinded it as premature in July. I hope somebody at the ATO is crunching the numbers to see at what point the cost savings from retiring X thousand "legacy" systems on the ground 10 years earlier would offset the cost of an ATO program to bulk-purchase ADS boxes for GA planes.
Outsourced Flight Service Gets Good Grade from AOPA. The largest non-Defense Department outsourcing in federal government history was last year's contract between the FAA and Lockheed Martin, under which the latter is consolidating and modernizing the agency's Flight Service Stations (which provide various flight-planning services to general aviation pilots). Last month the Aircraft Owners & Pilots Association, after meeting with LM project people, gave their performance thus far a grade of B+. As of now, the company is running the system with 500 fewer employees, yet providing quicker call-answering than before. It is recruiting and training new people (especially private pilots) as FSS Specialists. So far, so good for this form of privatized service.
NATCA Ousts Old Leadership. In Issue No. 35, under the headline "Three Strikes and You're Out," I noted that three times in the past few years controllers' union president John Carr opted to spend a small fortune on militant, confrontational campaigns to get Congress (and the public) to intervene on its side of important issues. After losing all three, I asked whether controllers "really want to continue this kind of union leadership, at a time of impending structural transformation of air traffic control?" Evidently they do not, since on July 31st they voted by a 4-3 margin to elect Pat Forrey to replace John Carr. Forrey took over September 1st, noting that the main premise of his approach is: "Dialogue is imperative to solving problems." What a breath of fresh air.
ATCA Selects New President. The board of the Air Traffic Control Association announced in late July that it had unanimously selected Pete Dumont as the organization's new president. It's a sign of the times that in addition to a background as a military air traffic controller, Dumont comes to ATCO from his position as vice president of Serco, one of the world's leading private-sector providers of air traffic control services. Serco runs many of the towers in the FAA's Contract Tower program. Congratulations, Pete!
Executive Reading. My article "Solving Airspace Gridlock" appeared in the July/August issue of Chief Executive magazine. In it, I lay out for corporate CEOs what is at stake for them in the upcoming battle over how to fund and modernize air traffic control�and what lies in store for all aviation if we fail to expand the system's capacity in a timely manner. The article draws on the analysis of the impact of ATC user fees on business jets in my recent Reason policy study (www.reason.org/ps347.pdf). We've posted the article on the Reason Foundation website: www.reason.org/commentaries/poole_20060800b.shtml.
Correction. If you downloaded my business jets and ATC user fees study when it was first posted, your copy may contain an error. Figure 3, which graphed wholesale jet fuel prices and business jet deliveries, had an incorrect right-side scale for the fuel prices. What was labeled as dollars should have been cents. That error has now been corrected.
"I always find it interesting when the discussion gets to ATC that the conversation turns to technologies based on ground command and control. It is that very system that was designed and implemented based on assumptions that are no longer valid. The system was designed assuming position uncertainty of the aircraft and the other aircraft in proximity. There was a need for a third party to provide, through procedures or technology, in this case Radar, an assured separation from other aircraft. These and other assumptions made during the �50s and �60s are as obsolete as the thinking that continues to prevent the next-generation system to be established. The airborne platform can assume much of this assured separation in a much more efficient manner; the ground system could act as a safety net function, but in a much less costly manner. The whole concept of airways and the ground navigation that supports it are expensive legacies of the outdated assumption. The costs to the government and the users are enormous. Does anyone believe that there is a need for 20 centers and 135 approach controls?"
—Neil Planzer, Boeing, commentary on Mifnet (used with permission)