June 5, 1998
A retraining program for air traffic controllers is now underway. More than half of the nation's 18,000 controllers have been ordered to take a two-hour refresher course after a year collision over a busy airport. The Federal Aviation Administration said all controllers who handled take-offs and landings had to complete the course by the end of June. The action was prompted by a close call in April between two planes at New York's LaGuardia Airport. The NewsHour has the report.
PHIL PONCE: Last April there was a near midair collision between two passenger planes at New York's LaGuardia Airport. That prompted the Federal Aviation Administration to order 10,000 air traffic controllers nationwide to undergo retraining on handling takeoffs and landings. In a memo about the near collision the FAA said, "Nationwide increases of 19 percent in operational errors and 49 percent in surface errors clearly indicate a need for increased emphasis in this area." For more on this order we're joined by Randy Schwitz, the executive vice president of the National Air Traffic Controllers Association, and Michael Goldfarb, former chief of staff at the FAA. He now runs an aviation consulting firm. And, gentlemen, welcome. Mr. Schwitz, I understand you've been speaking with air traffic controllers at LaGuardia. What do they tell you happened?
RANDY SCHWITZ, National Air Traffic Controllers Association: Well, as far as the April incident is concerned, there was a departing aircraft, Air Canada, Airbus, departing one runway, a USAir aircraft cleared for approach at an intersecting runway when the controller made the clearances to both aircraft. He was anticipating separation, using visual separation. By the time the Air Canada began its roll, he noticed that the aircraft was slow rolling down the runway, anticipating it was going to go a little faster, seeing that, looking back, seeing that the USAir was now about ½ mile final, he decided to abort the landing for the USAir aircraft, told him to go around. The Air Canada aircraft at that time was airborne, climbing at a rate a little faster than was anticipated. The USAir continued down the runway to go around. The Air Canada climbing out, climbed above the USAir and approximately 20 foot of clearance was between the two aircraft as they passed one below the other.
PHIL PONCE: Mr. Goldfarb, how serious of an incident is that?
MICHAEL GOLDFARB, Former FAA Chief of Staff: Well, it's serious, Phil, but on the other hand, here's an example where the FAA-it's a new FAA. The FAA is getting out ahead of this problem. The ordering of the retraining of 10,000 controllers does not imply we have this problem in every control tower around the country. But we've had a 19 percent rise in operational errors this year, and although that's down from the 80's significantly, it's an indicator and a warning that maybe perhaps we need to redouble our efforts and make sure that controllers understand better when go-arounds should be called, when procedures should be put in place. But on the other side we've got to stop playing catch-up in this country on the air traffic control. Every time we have something that happens we have sound bite politics where we talk about needing more controllers, needing more technology. When the lights go out and the FAA every day has to carry on its mission, it's faced with lack of funds to put new technologies in air traffic control, lack of money to really do the kind of modernization and upgrade to handle the traffic growth, which is fairly significant, over the next five years in this country.
PHIL PONCE: Staying with this specific incident just for a minute, Mr. Schwitz, who was at fault?
RANDY SCHWITZ: I think the system was at fault. It's gotten to the point that the demands on the airports in this country are exceeding the capacities that they were designed for. This was a case that a controller, applying separation standards, which were completely legal, and the pilots doing what they do, also looking out the windows, those two combinations actually prevented a near miss-well, actually prevented a mishap.
PHIL PONCE: Although the FAA in their memo issued last month did indicate that while the air traffic controller-while what the air traffic controller did was right, he maybe waited too long to do it.
RANDY SCHWITZ: That's a matter of opinion. It's Monday morning quarterbacking, and it's very easy for an air traffic controller-for me to look back at what the air traffic controllers-what they may have done.
PHIL PONCE: --quoting what the FAA memorandum said-and by the way, we invited the FAA, and they declined our invitation. You were going to say.
MICHAEL GOLDFARB: It's under investigation, and while it's being investigated, the FAA probably wouldn't talk about the details, but I think there is a problem here. While the FAA is playing catch-up, it's not appropriate-it's not-it shouldn't be tolerated to have such a delay in the reporting of the information, both from the pilot, who thought we should have reported it to the LaGuardia Tower, and as well as the air traffic supervisor, who didn't think it's significant enough to report it up the chain. We can't have these kind of things not being reported immediately to senior managers. So that delay was too long.
PHIL PONCE: Why the delay? This-the incident-this memo came out in May, this memo ordering this retraining. How come it's taken so long, Mr. Schwitz?
RANDY SCHWITZ: As far as the FAA reporting there was really nothing to report. There was no incident. The controller followed the rules of separation that he was supposed to in this case. The supervisor has even said that, well, there was nothing to report. There was no incident.
PHIL PONCE: You're saying that when two planes come within 40 feet of each other, there's nothing to report?
RANDY SCHWITZ: As far as the rules of separation that were applied, there was no incident the controller was involved in, because technically there was nothing to report.
PHIL PONCE: Is that right?
MICHAEL GOLDFARB: Well, I think from a technical standpoint, Randy's right, but I think that two planes coming 20 feet-imagine had those planes hit-we wouldn't be talking about whether it was appropriate to retrain air traffic controllers today. We'd be having a different kind of conversation. It was a serious incident. The FAA took the right action. They were aggressive on this case, as they have been in the last several months on the wiring of the 727's, and it's a new FAA that's not going to let these problems fester, and we need to look into why it did take so long--I believe they will-why it took so long to report these things. We have to know right away. We're having errors in the system. The system really has to be upgraded, and I think Randy would agree that the controllers, who have the outdated tools, you can get a better computer at Radio Shack than you have in some of the terminal approach facilities today.
PHIL PONCE: Well, as far as the public is concerned, how common are these near collisions, these near misses? How common? How often does it happen?
MICHAEL GOLDFARB: Well, actually I think that the FAA is encouraging pilots to report pilot deviations whenever separation standards are violated. So technically if you violate four miles on trail, that's considered an unsafe separation. The FAA needs to know that. They need to know about pilots performing in the system, about air traffic controllers, so they can put in better procedures. They're uncommon to have that kind of close separation-I don't think we've had that close separation in, you know, in many, many years.
PHIL PONCE: The FAA is evidently keeping track. I mean, in their memo they talk about a 19 percent increase in operational errors and 49 percent in surface errors. Those kinds of errors relate to-do both those categories of errors relate to what controllers are doing or maybe not doing?
RANDY SCHWITZ: To me, to look at those errors, if you look at the FAA's record keeping, they run in cycles. Working in the system I can tell you they don't run in cycles. The error rates that are out there are pretty steady. We see increases when there's a heavy push on reporting, when there's a heavy push on getting the errors down, you see the errors go down. And what's the easiest ways to get errors to go down is you don't report 'em. I see-there is-that has gone on, and I can tell you, it continues to go on today. When they try to get the errors down, there's pressure on the managers and facilities to get the numbers down. So the easiest way to make them look good is you don't report errors.
PHIL PONCE: Mr. Schwitz, let's talk about the training that's been ordered. Is this a prudent thing for the FAA to do, in your opinion?
RANDY SCHWITZ: No, I don't believe it is, especially the training that's going to be provided to controllers and is being provided to controllers in this instance. The stuff that they have in the memo that they're providing the controllers with is, in my opinion, is silly.
PHIL PONCE: What kinds of things-with real things-asking for?
RANDY SCHWITZ: These are real basic training asking them to review portions of the handbook they operate with on a day to day basis. They're viewing a videotape of past incidents that have resulted in crashes at airports. The best thing that happened to this controller, as far as retraining, is when he sees a situation present itself similar in the future, he knows exactly what he would do different now.
PHIL PONCE: Mr. Goldfarb, do you think it's a prudent step by the FAA, or unnecessary or redundant?
MICHAEL GOLDFARB: Well, I would say to the FAA overreact, overreact. That's the kind of response we want from the FAA, and I understand from a controller's standpoint that it may look like you're placing blame on the controller work force. Nothing could be further from the truth, Phil. Here's an example where we just don't know. And if we don't know, let's redouble our efforts. Let's do proficiency training, and I would guess that other things are going to come out of this as well in terms of how to tighten procedures in the field, the reporting of information so that we have a safer system going forward.
PHIL PONCE: Mr. Schwitz, do you think it's conceivable that this training could conceivably help prevent incidents like this in the future, or not really?
RANDY SCHWITZ: There's really no way to tell. Based on personally, my opinion, what the controllers are telling me in the field that are going through this training, they think it's a waste of time.
PHIL PONCE: Does the FAA have enough air traffic controllers, gentlemen? Mr. Goldfarb.
MICHAEL GOLDFARB: Once again, the numbers become somewhat of a shell game where we say how many-how many is enough. It's really staffing at the facilities. Are they staffed? Is there enough bench strength? Are the controllers there to support other controllers when they have to be taken off the scopes, or when there's a significant amount of traffic in weather, bad weather? So it has to be looked at more individually. We've got to get away from these indicators of how many controllers, or how many errors. It's very deceiving, you know. Whenever there's been a crash, people who've lost people in crashes don't like the statistics that it's safe to fly. It's the margins of safety. Do have in place the kind of system where the pilots, the controllers, and the people feel encouraged to report information back without sanction, so that better steps can be put in place? And I think we need to encourage-and, in fact, if we saw a rise in errors, that might not necessarily indicate an unsafe system; it may be just a better ability to prevent future errors.
PHIL PONCE: Mr. Schwitz, is it a relevant question to ask, if there are enough air traffic controllers?
RANDY SCHWITZ: Well, I think it is. As you initially reported at the beginning of the show, that there's 18,000 controllers. Evidently, the FAA told you that. But now there's a little over 14,000 controllers in this country. They lump in the numbers of staff personnel they have that work approximately less than 1 percent of the time during the year on air traffic control positions, supervisors who also spend less than 1 percent of their time working air traffic, and they include those in the air traffic control numbers and report those to Congress.
PHIL PONCE: Quickly, how many air traffic controllers do you think are needed?
RANDY SCHWITZ: In 1981, we had over 16,000 controllers. Today we're working 35 percent more traffic and have about 3,000 less controllers than we had then.
PHIL PONCE: Gentlemen, I'm afraid we're out of time. That's where we'll have to leave it. Thank you both.
RANDY SCHWITZ: Thank you very much.
MICHAEL GOLDFARB: Thank you.