Failure To Recommend Corrective Action a Lost Opportunity, Controllers Union Says

Sun, Oct 14, 2007 — David Evans


The National Air Traffic Controllers Association (NATCA) continues to be dismayed by the fact that the Federal Aviation Administration (FAA) permitted only one controller to be on duty in the tower at the time of the Comair flight 5191 accident on 27 August 2006 (the midnight shift), and that the National Transportation Safety Board (NTSB) did not issue any recommendations addressing specifically the issue of tower staffing (see Air Accident Digest, 4 September 2007, p. 1, “Pilots Faulted in Comair Crash”).

Indeed, the NTSB did not hold the FAA accountable as a contributing factor in the probable cause statement.

Darren Gaines, a member of NATCA’s Air Safety Investigations Committee, gave a presentation on 11 September 2007 in which he pointed to past NTSB accident investigations that, to his view, properly pointed to issues of FAA oversight of local traffic services and of FAA quality assurance (and the lack thereof). Gaines pointed to the NTSB investigation of the fatal 1991 crash of a USAir B737 at Los Angeles, which collided with a Skywest aircraft on the ground, and the fact that the probable cause statement panned the FAA for not providing adequate policy direction and oversight.

“The published probable cause [of the Comair accident] is meaty … but there are issues not in the probable cause that need to be discussed,” he said. “It’s not healthy having one person on the midnight shift. It’s not about workload – it’s about an additional level of safety.”

“The [lone] controller was in his last day of the workweek, on his last hour, operating in the trough of his circadian rhythm,” Gaines pointed out.

Speaking in terms of “our” because he is an air traffic controller, Gaines said, “Our performance is degraded, our reaction time is reduced, our reflexes aren’t the same. For these reasons alone, operating down there [in the region of circadian low] we need somebody else. We need that extra layer of safety … we need those eyes on the runway out there.”

The controller on duty at the time of the accident was later interviewed by NTSB investigators. He indicated that when a second controller was available on the midnight shift that they were both in the tower.

Gaines pointed to the Swiss cheese model of accident causation, in which holes in the cheese represent vulnerabilities that can lead to an accident. In this particular case, Gaines said, “What’s the element that’s missing? The other controller in the tower. That would plug he holes at the highest level.” (See Figure A).

FIGURE A: In the Swiss cheese model of accident causation, the holes represent poor defenses or absent defenses. NATCA asserts that holes in the first two slices of cheese can be laid at the feet of the FAA. Source: NATCA

“Working your way back [from the accident], you can stop at the [lone] controller … but I scratch my head when the FAA allows one man in the tower,” said Gaines.

“The NTSB essentially gave the FAA a ‘free pass’ to continue short staffing towers,” Gaines alleged. “We controllers keep getting beat-up in NTSB reports for ‘not following procedures’ (all of the weather accidents, Crossfield, etc. [see brief below]), that may have prevented accidents,” Gaines said. “Yet the FAA is not held responsible for not following procedures (2-man mid-watch), when following this guidance may have prevented he accident.”

“What if this airplane had taken off with one pilot on board? That would be criminal. We would never tolerate that. We wouldn’t even question redundancy in the cockpit. We know it’s needed, it’s required, and it needs to be there,” and the same ethic applies to the tower, Gaines said.

While the lone controller did issue a legal takeoff clearance that was read back and acknowledged by the flight crew, it should be mentioned that “clear for takeoff” is little more than a homily if it’s premature, formulaic and unmonitored. If a controller “owns” the airfield, he must assume responsibility for whatever happens on and above its surface, within the ATZ (air traffic zone) and, via his radio, exercise positive control. To say that you were “busy doing some lesser task” (what the controller essentially told NTSB interviewers) and therefore didn’t see a lethal situation developing is evading the responsibility inherent in a profession that’s clearly about owning and controlling outcomes.

NATCA has provided additional observations that warrant comment:

NATCA: “Counting strips was something that should have been done by a radar controller.”
Comment: Maybe, but there wasn’t one. Given the circumstances, it is reasonable to expect that a lone controller would recognize his enhanced responsibilities and defer his strip-counting to quiet periods (of which he’d have many at a quiet backwater like Bluegrass Airport). It’s called risk management, and that should be espouse as at least an interim answer (but no solution) to the FAA’s proven inability to properly staff airfields with sufficient trained controllers.

NATCA: “You’ll have complete gridlock of the entire system because there are not enough controllers in every tower to monitor every single aircraft at all moments.”
Comment: Proper supervision entails monitoring (but not necessarily watching constantly) the aircraft’s movement towards the active runway and ceases only when the aircraft starts its takeoff roll. Some in the industry are fans of aircraft calling “Ready” as they approach the hold-short line. This means the controller will be prompted to confirm his expectation of seeing the aircraft at the correct hold-short line.

What has evolved over the years, amongst a welter of loopholes, is the facile practice of issuing a very early takeoff clearance. This can work flawlessly when there is no conflicting traffic and weather or flight crew confusion isn’t a potential factor. Issuing an early clearance also disposes of a pending and distracting task, and allows the controller to get on with other things. But it’s still a shortcut fraught with possibilities.”

NATCA: “Staffing is a safety issue; it is part of the safety net the NTSB talked about, part of the redundancy.”
Comment: When you are short-handed and the redundancy evaporates, you must change your modus operandi. The point at which you cry “enough” and put down your tools becomes that much closer. The ultimate example of that not being done was the Uberlingen Bodensee German mid-air collision (TU154 versus B757 in 2002). The recent trial convicted the air traffic control supervisory chain in SkyGuide (at the time of the accident, only one controller was on duty, and he had to divide his time between work stations to use a second radio frequency). The trial focused on oversight that led to the staffing shortage. One can only say “Hear, hear!”

In effect, one can construct an accident via a poorly composed manning roster that ignores the effects on safety of staffing and other systemic deficiencies. In this respect, NATCA is absolutely correct.

Comments are closed.

Nolan Law Group