U.S. Air Traffic Control: The Moral Equivalent of Semaphores and Signal Fires
Adventure is back in air travel as 60,000 passengers on 600 cancelled flights found out, when the Federal Aviation Administration's (FAA) medieval air traffic control (ATC) system crashed earlier this week.
According to Reason Foundation transportation maven Robert Poole, the problem is that the U.S. ATC system is a step up from semaphores and signal fires, but not by all that much. Poole explains:
Today's ATC system is built on a 1950s paradigm or concept of operations. Because it is so imprecise, it must create huge buffer space around each plane, wasting valuable airspace. All communications between ATC and planes go by voice—on frequencies that are jammed and relaying numbers that can be mis-heard. It is hugely labor-intensive, when software could accomplish many routine tasks in keeping planes safely separated.
Go here for some excellent Reason Foundation proposals on how to finally bring America's ATC system into the 21st century.
Editor's Note: As of February 29, 2024, commenting privileges on reason.com posts are limited to Reason Plus subscribers. Past commenters are grandfathered in for a temporary period. Subscribe here to preserve your ability to comment. Your Reason Plus subscription also gives you an ad-free version of reason.com, along with full access to the digital edition and archives of Reason magazine. We request that comments be civil and on-topic. We do not moderate or assume any responsibility for comments, which are owned by the readers who post them. Comments do not represent the views of reason.com or Reason Foundation. We reserve the right to delete any comment and ban commenters for any reason at any time. Comments may only be edited within 5 minutes of posting. Report abuses.
Please
to post comments
And Reason gets burned by the copyright monkeys!
And everybody knows that software is free!
And on the 3rd attempt, we get a cartoon that's legally appropriated, legible, and somewhat funny.
If only Friday Funnies could be this way!!!
LMNOP: I took down the first cartoon because I hadn't initially seen the permission required notice. It's polite to ask permission in such cases, but that would take time and there are other cartoons to illustrate the problem after all.
But a cumbersome civil-service bureaucracy, funded in dribs and drabs by annual congressional appropriations, is more likely to impede that transition than to facilitate it.
So the Reason Foundation is actually calling for massive increases in expenditures. aka MORE government.
Beam me up, Scotty. Reason has become a socialist think tank.
LMNOP: I like the third one better anyway. Always aim to amuse H&R commenters. 🙂
LMNOP: I took down the first cartoon because I hadn't initially seen the permission required notice. It's polite to ask permission in such cases, but that would take time and there are other cartoons to illustrate the problem after all.
I'm just joshin' ya. Misdirected anger from the convention, I'm sure.
Oh, for certain, the third one was the better of the two that I could see.
IF: Had you looked at some of the Reason Foundation research to which I so helpfully linked you would have found that Poole favors essentially privatizing ATC.
It ishugely labor-intensive, when software could accomplish many routine tasks in keeping planes safely separated.
Well, there's your problem.
Nobody in the current system has any incentive to reduce head count. Just as in the private sector, no chief wants to have fewer indians. Being the supervisor of a bank of blinking lights doesn't pay as well as being the supervisor of a hundred human controller. However, unlike in the private sector, there is no inherent pressure to improve efficiency. No one can create a competing system and force an upgrade.
It's just another systematic defect of a government run system. I suspect we'll just stagger along until we have a horrible catastrophe that will create the political will to clear out the deadwood.
What, Ron, you couldn't find an Airplane screenshot? I guess I picked the wrong day to stop taking amphetamines.
Always aim to amuse H&R commenters. 🙂
You know, I'm always a bit stunned that you guys even put up with us idiots. I would abuse the shit out of the banhammer if I were in your shoes.
Episiarch: Thought that a screenshot from Airplane would be just a tad too obvious.
Isn't every day the wrong day to stop taking amphetamines? 😉
You know, I'm ok with there being larger buffers between planes than a computer simulation designed for maximum efficiency would recommend. I'm funny that way.
Isn't every day the wrong day to stop taking amphetamines? 😉
Ron, you are a man of insight.
joe,
Wuss. 🙂
robc,
Ma! Ma! In here! Wave to gradma, kids!
You know, I'm ok with there being larger buffers between planes than a computer simulation designed for maximum efficiency would recommend. I'm funny that way.
Well, as in most engineering exercises, you would build in a lot more slack than than the formula says you need.
The current system is incredibly wasteful and inefficient. The former is annoying, and probably contributing to the airlines economic problems. The latter is dangerous.
I'm wondering if the "major organizational and funding reform, as has already been done in Australia, Canada, and most of Europe" to which Mr. Poole alluded has made the presumably government-run air traffic any better in those areas?
You know, I'm ok with there being larger buffers between planes than a computer simulation designed for maximum efficiency would recommend. I'm funny that way.
"Dang burn Postal Service! If I wanted my letter there sooner than 8 weeks, I'd 'taint have sent it by pony!"
I'm guessing that there might be a design tolerance that's somewhere in between the UNIVAC-level system we have today and maximum efficiency.
My understanding is that the real challenge is actually implementing the upgrade. You have to pretty much replace the whole system at once, and you're not allowed any downtime - ever.
This will take some special doin', as it's already been tried a couple of times, then abandoned after spending millions of dollars.
Privatization=a Weld Scam.
So read the bumper stickers found on so many cars of union members and big government types in Massachusetts back in the day when gov. Weld floated the idea of privatizing certain state activities. Although I scoffed at such assertions at the time, I have grown to realize that privatization is a scam. It is inherently antithetical to libertarian principles and the Air Traffic Control System is no exception.
Free Minds and Free Markets necessarily doe not embrace the endorsement of privatization schemes that only serve to prolong the involvement of government in transportation activities. Last I read, the Commerce clause does not provide the government with the authority to establish an Air Traffic Control system or that the government has a right to regulate air traffic.
Let's face it. Privatization ensures that some rent seeker is going to get rich. Not by free and unfettered competition-but by government fiat. Not very good. Not very efficient.
The real solution to the problem is eliminating the state from the equation and to leave it to the market to solve. No subsidies. No mandates. No regulations.
libertymike:
I believe navigable waters doctrine extends to the air. Certainly most flights qualify as interstate commerce. If there is a need to regulate air traffic, I don't see how that could really end up anywhere but in federal hands.
According to Reason Foundation transportation maven Robert Poole,
Sheesh, Bailey. You might have mentioned also that Poole is one of the founders of Reason Magazine. I'm sure not everyone on these boards is aware of that fact. 😉
Except as this very story illustrates, sometimes software as glitches. As a professional software engineer I can assure you that there is no such thing as perfect software*. The core problem with the current ATC system is that it can't handle failure mode. Failure is going to happen and a new system will have to be able to cope with a complete breakdown of the software.
*) There does exist bug-free software, but it deals with highly deterministic domains, and air traffic control ain't one of them.
"semaphores and signal fires"
Tell me, was that about the time runways were invented too ?
We could argue all day about privatization, the role of the State and a bunch of other things. Without some additional runways, none of it matters.
What's the latest slot restriction at JKF ? 81 an hour ? ATL (Atlanta) handles around 120 an hour. Has anyone asked themselves why ATL can handle that many and JFK can't ?
ATL has five parallel runways. JFK has two. Put five parallels at JFK and it can handle 120 operations an hour also -- without any new gee-whiz technology (purchased from Mr. Reason's supporters) and without any useless debate about whether the government or private industry is better.
It's the runways. Always has been -- always will be. Any air traffic controller knows it but I'm guessing no one here is one and I *know* Mr. Reason isn't.
Don Brown
The problem with most real-life privatization schemes is that they aren't privatization schemes; they're management contracts or--at best--long-term leases. Often there're strings attached: you can't split it, you can't liquidate its assets, you can't change your prices, you have to submit your management to government directives, you can't fire anyone or change their compensation, etc. These factors discourage bids, so sometimes governments will sweeten the deal by awarding the buyer a legal monopoly over that sector of the economy.
Criticism of privatization schemes is better directed not at the concept of privatization, but at the myriad ways in which such schemes deviate from the definition of privatization.
LOL, Software, saved man hours, more efficient, safer, economical?? Nah, that'll never happen. Common sense is a thing of the past.
Whisperer
http://www.anoweb.alturl.com/
To Don Brown:
Care to share where you would like the 3 new runways located?
Alan:
Hint!
One antonym of "privatization" is "condemn through eminent domain".
New World Dan-
Thecommerce clasue does not say that congress has the right to regulate air travel. That is undeniable. Second, I do not and have never bought expansionist arguments urging that if congress can regulate interstate commerce over land and water, why shouldn't it apply to airspace. Choplogic. That is what it is. To argue otherwise means that the federal gvt. is not a gvt. of limited powers but is a gvt. that has unlimited implied powers.
Alan,
Looks like 5 to me
Some Fed-
1/2 and 1/2.
In and of itself, privatization is bad if it means the continuation of some compelled activity or scheme. Just because the administration or management of the compelled activity or scheme is in "private" hands does not mean that it is good.
Of course, you are right that the myriad ways of implementing privatization plans are often fraught with all sorts of mischief.
How did Hong Kong do it?
No offense to Mr. (Dr?) Poole and all, but I don't trust the current administration *at all* in this. Give me a real deregulator like Jimmy Carter any day over what currently passes for 'privatization.'
Auction those slots, and see what happens to congestion and efficiency.
I had no idea there even was such a creature as a "Bush transportation policy".
Howard Beach is an armpit - they can have that.
Otherwise the real answer is here.
"You know, I'm ok with there being larger buffers between planes than a computer simulation designed for maximum efficiency would recommend."
Any pilots in this crowd? I'm just wondering how many people know what sort of dedication and hard work it takes to have a mid-air collision.
"It's the runways. Always has been -- always will be. Any air traffic controller knows it..."
Uhm, my cousin with over twenty years at Chicago ARTCC says shut up.
"Care to share where you would like the 3 new runways located?"
The same place ATL will put the 6th, 7th, 8th or 15th.
That's the point. There are limits. You aren't going to build any new runways at LaGuardia. A new ATC won't increase the capacity. We need to update the ATC system but we always need to update it. And we always have. Running airplanes closer together doesn't do you much good if they don't have a place to land.
Don Brown
The commerce clause doesn't say a lot of things. It's vague and open ended and has, of course, led to a great deal of abuse. But I don't dispute that the feds have the power to regulate flights. I don't agree that they need to or should actaully operate air traffic control directly. Nor do I know enough about air traffic control to tell you how it should be done. In the grand scheme of things, the FAA is one of the smallest abuses of the commerce clause that I've seen, if it is indeed an abuse at all.
LIT: Yes, Atlanta has 5. I was saying that JFK, while it only has 2, is completely surrounded by water, highways, and houses. It would be the mother of all Eminent Domain projects to expand the airport in the way suggested by Don Brown.
P. Brooks: Easy - It was relocated from Kai Tak to Chek Lap Kok - which was conveniently barren at the time.
JFK has four runways: two sets each in parallel.
"Uhm, my cousin with over twenty years at Chicago ARTCC says shut up"
Billy,
Tell your cousin that the guy with 25 years at Atlanta ARTCC says, "Retire". 🙂
Don Brown
http://gettheflick.blogspot.com/
Not likely. I know him. I don't know you at all except that I might have spoken with you here & there some years back.
Billy - you are correct. I only meant that only 2 are in use at any time. All 5 of ATL's runways are parallel, which appears to be a cleverer design, but a bit late for JFK. I guess it was built before pilots learned to how to land in cross-wind.
"JFK has four runways: two sets each in parallel."
Too bad using all four at the same time (or even three of the four really) is such a bad idea.
http://www.youtube.com/watch?v=fi9GQ8lYg7Y
Don Brown
http://gettheflick.blogspot.com/
I guess it was built before pilots learned to how to land in cross-wind.
ATL was built when that area was in the middle of nowhere (it's still 17 mi from downtown) while Idlewild is in Queens 10 miles from the center of what has been the top ten most densely populated places on the planet for a century.
Whoops hartfeld is only about 8 and half miles from the centential park, must have been thinking about the round trip
We have examples of goverment privitization now. The FSS (Flight Service System) system has been privitaized for years now and has been a total disaster. This is the system that pilots are supposed to use to obtain weather. Not one experienced pilot I have talked with likes the system. But no one seems to care because the only people dying are a few GA (small) types.
Usally when some one whats to privitize a goverment function, its because there is a belief that private sector can do a better job, or that companys can be trusted to govern themselves. But what happens when connected political types stack the deck against the public to make it seems is if government can't do the job. Like forcing out older and qualified workers to be replaced by younger trainees unfamilar with the job. How about removing qualified Inspectors replacing them with "auditors" that only inspect by telephone or from thier desks. (Southwest Airlines, American Airlines ect...) So when the enevitable happens, the death of passengers on an airliner, the powers that be can scream, "see government can't do the job!!! " And the public not knowing the back story eats up the spoon fed argument and demands that new companys begin doing the work that the goverment can't obviously do any more. Just by happenstance there are companies in place ready and able to take over where the FAA and government "failed". Oddly enough staffed by former goverment managers with the training and know how to do the job.
Unfortunalty the public right now only cares about cheap fares and arriving somewhat on time. Its the goverment workers that are being set up to fail that know whats happening.
"The FSS (Flight Service System) system has been privitaized for years now and has been a total disaster."
What are you talking about? I've used FSS and loved it.
Mr Brown, if you are the former columnist from Avweb, I always enjoyed your column. Cool that you lurk around here.
I'm guessing the title of this blog wasn't written by a pilot. I won't dignify it by explaining why it's stupid.
"Any pilots in this crowd? "
Yup, I'm a pilot. (Say that like the drunk dude in Independence Day.)
Mid-airs? Well, they tend to happen where it gets crowded, like around an airport. Someone mis-tunes the traffic freq., or mis-hears a tower command, coupled with a lack of vigilance using the old eyeballs, and boom, shit happens. It's very, very rare, however. Even more so when in positive control airspace (class C or class B). Your typical pilot, when not on a jihad, just don't like dying.
Solutions to the current capacity problem? This is a pretty complicated subject. ADS-B could be part of the solution. This technology uses GPS to encode a given aircrafts position in a digital transmission to ATC. All aircraft within 100 miles equipped with ADS-B will then receive this transmission and know where the other traffic is. Everyone gets to see the real time big picture, not just the radar jockeys on the ground. In theory, this will allow more precise spacing and faster traffic flow.
What ADS-B won't solve:
Slots at the terminal. There are only so many berths available at a given time.
Airlines preference for the current hub and spoke system. It's cheaper for them, but less efficient for passengers. Pick one, fast or cheap.
Additionally, most airports have superb radar coverage and controllers. Having everyone see the big picture is not likely to double the landing rate.
"Additionally, most airports have superb radar coverage and controllers."
GA could still use a lot of help. There was flap six or seven years ago about TARDIS (Tornado Advanced Radar/Map Display Information System) installations that FAA would not authorize. I was flying around LZU (Gwinnett County, the third busiest airport in Georgia) then, head on a swivel like a fighter pilot in Class D airspace with no radar. A couple of higher-profile crashes around the country loosened their hand a bit but that program is still going nowhere serious.
There was no reason on earth why we shouldn't have been able to take up a collection on the field and have a TARDIS installed on our own without their leave.
"What are you talking about? I've used FSS and loved it."
FSS privatization has been neither horrible nor wonderful. The first year was pretty rocky.
But remember, the contract to Lockheed Martin was for far less than Uncle Sam was spending on it. Lower funds plus lack of local area expertise produces the predictable result. Want good weather info? Sign up for XM satellite weather, get the proper gear on the panel, and get real time (sort of) data.
"Mid-airs? Well, they tend to happen where it gets crowded, like around an airport. Someone mis-tunes the traffic freq., or mis-hears a tower command, coupled with a lack of vigilance using the old eyeballs, and boom, shit happens."
Here's one around LZU that I wrote up, once. Nobody swapped paint, but it was pretty sporty for a minute.
you would have found that Poole favors essentially privatizing ATC.
Real privatization means competing systems are A-OK. Poole is not calling for that, so there's nothing "essentially" privatized about his proposal at all. It's merely a privatization veneer over a government-run-in-all-but-name bureaucracy.
Poole's suggestion praises Norm Mineta - a man with such high regard for the co-op paradigm that he dismantled the security co-op system in favor of the TSA.
On a related note, isn't the subject of capacity rather moot? The major problem with US airlines has been too much capacity. Seems like this is a solution in search of a problem. Why would any sirline want to voluntarily pony up to modernize the system when their lax attitude towards security resulted in offloading that problem on the taxpayers. The proactive pioneers are the one with the arrows in their backs.
My understanding is that the real challenge is actually implementing the upgrade. You have to pretty much replace the whole system at once, and you're not allowed any downtime - ever.
That's why a better idea is to let private plane operators and hobbyists work on the problem. You're going to have to run parallel systems, and I can certainly see the willingness in those people for inexpensive experimentation. If a few hobbyists crash their planes and die, it's sad news but it's not a national crisis and the experiments don't end. One foulup with a passenger plane and progress grinds to a halt. With Poole's bureaucracy-led idea you'll also wind up with a system requiring reasonable expense rather than a system requiring massive capital expenditure designed to keep startup airlines from entering the business.
Pick one, fast or cheap.
LOL! In order to do that, you have to acknowledge the law of supply and demand. Too few are willing to.
With Poole's bureaucracy-led idea ...
Should've read "without..."
"That's why a better idea is to let private plane operators and hobbyists work on the problem."
Yes. A good example is the TARDIS thing I mentioned. There was no reason in the world why we shouldn't have been left to do that on our own.
"Mr Brown, if you are the former columnist from Avweb, I always enjoyed your column. Cool that you lurk around here."
Jasa,
I are and lurkin' is what retired guys do. I just go wherever Google sends me.
Thanks for the kind words.
BTW, "Mr. Brown" is still alive and well. I'm just Don.
Don't feel bad. I called Robert Poole "Mr. Reason". We all make mistakes.
I agree that ADS-B is cool. But it won't increase the capacity of a runway. Slight improvements are possible. Runways will always be the big limiting factor.
Don Brown
http://gettheflick.blogspot.com/