UAL Traveler: There are definitely improvements that can be made to the air traffic system. One of them in the works is to give aircraft an arrival time and how they make that time is up to them - whether they want to hold, slow down, take a tour, etc. Utilizing en-route airspace better, whether by reducing separation or more direct routes (don't forget the military blocks a tremendous amount of airspace), it still comes down to the fact that SFO, for example, can land between 28 and 60 aircraft an hour depending on weather. And more aircraft than that want to use it.

As far as wheels-up time, that basically describes the current ground-delay program, which doesn't work that great either. However, it does limit the amount of arrivals at an airport such as San Francisco during weather delays, which are frequent. For some insight into what fliers face in getting more runways built, you may want to do a search on the UA forum for a Chicago Tribune series that ran a few months ago relating to flight delays and expansion at ORD. It was very well done.

You can also think a minute about the San Francisco Bay Area in the last 10 years. Navy Alameda closed, Moffet Air Field closed and Hamilton Army Air Field closed. All very good usable runways nearly as close to San Francisco as SFO is and not being utilized.

The airlines need to figure something out also. Even flying out of a relatively small hub, such as St. Louis, look out your window at the line of 40-50 TW/AA jets, all trying to depart at the same time.

It all comes down to what happens on the ground. No matter what happens in the air, you simply cannot land all the airplanes when they want to.

letiole, free flight is no panacea, and I certainly don't debate the need for more and better terminal (in the broad sense of the word) facilities. Runways/taxiways/ramps/gates, or lack thereof are the ultimate choke points. As not a lot of options exist in that arena, I certainly support facility expansion.

However, when it comes to the rest of the system, there exists tremendous opportunities to save fuel, time, and headaches by making use of proven technologies to safely and effectively break us out of the highway-in-the-sky paradigm. There really is no excuse for en-route weather to have to paralyzing effect across the system that it often seems to have.

When it comes right down to it, if I may make so bold, I'd guess that you and I'd probably agree on quite a few more points than not.

Yes, I'm sure we do agree on many things. I do think, however, that it's difficult for the traveler to fully understand what weather does to ATC. The weather moves so there's no way to completely flight plan around it and to have 35 airplanes in your sector all going different directions around thunderstorms makes for an exciting afternoon for the controllers - not what we travelers want. At some point it does become a capacity issue for individuals.

They've spent millions so far on research and development and still have not come up with a program or a computer that can keep up with the decision-making requirements for rapidly changing air traffic situations.

I'm not sure what proven techniques you refer to, but remember that other countries simply don't have the traffic levels we have. Euro control is extremely restrictive compared to us and Canada and Australia have no airplanes.

letiole, I will take you points in reverse sequence.

First, when I spoke of making use of proven technologies to safely and effectively break us out of the highway-in-the-sky paradigm, I was referring to both hardware and software technologies that the FAA considers proven, if for no other reason than they approve the use of it for flight-critical operations. For example, we can use the (mainly) hardware technology of GPS for RNAV, where air-traffic density is high, MDA’s are close, and the margin for error is small. It seems logical that GPS can be deployed and fully relied upon in other areas of the national airspace where separations and distances from the nearest solid objects can be huge, without the restrictions of pre-determined routes. Even forgetting about the merits of free-flight vs the present system, I simply believe that greater use of GPS will lead to more accurate fixes, with less room for human error. GPS is intrinsically much more precise, and much less error-prone (system wise) than cobbling up a three-D fix based on a combination of ground-based and aircraft-based systems. The whole procedure of squawking transponder codes with an occasional ident, followed by some controller saying ‘radar contact…’ has never instilled much confidence in me. However, a direct broadcast of 3-D GPS coordinates (or 3-D plus the three velocity components) for all to collaboratively receive and use, both in the air and on the ground, as part of the support for a free-flight type of system does give me a warm feeling.

Your second point is well taken, if there is an insistence on maintaining centralized control and routing of air traffic. However, software technologies that can dynamically determine reasonably optimal routes, capable of accounting for moving weather systems and maneuvering aircraft, have been around and deployed on mobile platforms for quite some time. The difference here is that each router is working independently, counting on the probability of minimal interference from other vehicles. When that does occur, a conservative multi-tiered warning system kicks in. Thus far, it seems like progress is indeed being made, with the Free Flight Phase 1 tools apparently being on track to be in place by the end of next year.

Your first point is what its all about. Let the controllers efficiently grapple with the junk in the terminal airspace, and not worry about what’s coming down the pike. Sure, there will be times when thunderstorms blow up and arrivals/departures will be temporarily halted, but at least that will occur in a limited domain, and not needlessly foul up the global system, as is the current situation.

UAL Traveler: My husband, an ATC manager, agreed to address your post. My apologies for this all being so long. I don't know if anyone else is hanging in here ... but if you want to continue the discussion beyond this, perhaps we should email as perhaps we're tying up too much bandwidth??? Anyway, here's his response:

Ninety-nine percent of all air carriers rely on RNAV-GPS-INS for their navigation. There's no question about the equipment capabilities. It's a question of aircraft separation, which we'll come back to. There's already a program in place called the National Route Program, which allows the airlines to fly any routes they want between feeder fixes, which are about 150 miles from the departure and arrival airports. A built-in constraint to this is they have to fly around military airspace. So they end up flying airways on the West Coast until they're out of Nevada. For the controllers' part, we know exactly where the military airspace is and the altitudes it's in use at (some changes many times a day; some never changes) and we will send aircraft as directly as possible. There's nothing we want more than to get the aircraft out of our sector fast.

The main job of route flight planning falls to the aircraft company dispatchers. They take into account winds aloft, whether or not it's better to be in winds and take a shorter route vs. be out of them and take a longer route, known areas of forcast turbulance, do they want to avoid the best route where all the other aircraft is going to be, etc. Anyway, very complicated, done manually with the help of computers, but works pretty well.

Beginning last year, the FAA and industry decided to collaborate on routings during bad weather and peak demand periods. As a controller, I don't think this works very well. I have a tough time routing aircraft around weather five hours in front of them. My understanding is that no matter what the airlines publicly say they want to go back to letting their dispatchers pick their own routes and avoid the weather on a real-time basis.

The airways system is outdated and to a large extent is not used as anything more than a guideline by pilots and controllers, if that.

As far as giving everyone real-time 3-D data, that's a fine idea, but I don't see what good it does. A. The equipment is very expensive and airlines don't want to pay for it. B. Someone's got to be in charge and everyone doesn't need the same information. For example, try driving in Rome or Athens and remove all stop lights, signs, lanes, intersecting streets, etc. Look around and you have all the information a pilot would have with the new on-board equipment. Now who gives way? It doesn't work in these cities at 15 mph and it certainly wouldn't work in the sky with 1100-mph closure rates.

As far as equipment making the decisions, current state-of-the-art would be TCAS, Oceanic Conflict Probe and En-Route Error Detection Program (conflict alert). For TCAS, it's basically a simple program that hums around and says, "All these airplanes are around and none of them are bothering me. Then it's, Uh-oh, one's coming at me, it's time for evasive action." It's exactly that, evasive action, last minute, abrupt and error prone. Usually it's telling pilots to take evasive action for a climbing or descending aircraft that ATC has stopped 1000 feet above or below, but TCAS doesn't know that.

For the Oceanic Conflict Probe, the FAA promised a BMW and wound up with a Hyundai. The system took years of development simply to be able to compare times given to us by pilots and their times over a fix and is anyone else going to be there. After hundreds of patches to repair the original software, it works, kind of. There's a lot of caveats with it and it's not guaranteed. In fact, we're not allowed to use it for separation.

For the radar detection package, it's been around for about 15 yearsand still is unreliable and 99 percent of the time it provides false alerts. It has saved everyone at least one error in their career, so we do like it, however it's constantly flashing when it should not be. I would hate to think of it trying to make actual separation decision rather than the TCAS type warning that "these guys are going to hit."

As for what's on the horizon, my facility is near NASA/Ames and our controllers sometimes are paid to help develop and test experimental software. So far, no software is even close to being able to project courses, take into account winds that change by the minute, pilots changing their airspeeds, more than a single confliction (as in, move one and what does it do to the airplane next to him), not impact spacing requirements, allow for handling pilot requests in a rapid manner, separate random routes and pilots diverting around weather ... and these are probably less than half the constraints the computer would have to take into account.

Time for breakfast .... more to come.

letiole, and Mr. letiole, thank you very much for your thoughtful feedback. Yes, this discussion is eating bandwidth, but others might find it interesting. If anyone is still listening perhaps they can indicate whether this discourse is of value or not. If we get no positive responses, then lets take this off the boards and go to email.

I will keep this short, since you say there is more to come, and I am looking forward to hearing your thoughts on whether an en-route free-flight system will substantially materialize in the near future.

The only point I'd like to raise is relative to your driving analogy. You use a dense city as an example of why unsupervised control would be problematic. I fully agree that attempting to implement a free-flight type of system in dense, terminal airspace faces unquestionable challenges. (Aside: several years ago a candidate for Bangkok Governor proposed a solution to the city's traffic woes. He felt that cars were being held up at red lights, so he suggested turning all the lights green, all the time... seriously! He didn't win.) However, back to the traffic analogy, we see that cars driving on the open highways between cities, have no external controls, hurtle towards each other at rather impressive closure speeds, and yet rarely collide. Now, one could say that free-flight eliminates such highways, but in effect it doesn't for the purpose of collision avoidance. Aircraft involved in the imminent threat situation could (would) be capable of establishing the 'lanes' for themselves and for the oncoming traffic, and then decide what action to take. Based on the little I understand about FFP1 (free-flight phase 1), a two-tiered inner/outer zone would exist around each aircraft. It is the coincidence of the outer zones that would trigger such virtual lane creation and decision-making. As far as collision avoidance goes, I thought that the rules were well established, at least as far as FAR 91.113 goes. I guess I don't see much problem in having an en-route free-flight system utilize such rules. Transmitting GPS with radar backup would effectively render the sky more transparent than the best CAVU VFR day.

Anyway, I'll end this response by noting that I'll be heading to HKG/mainland China tomorrow (over the pole on UA 821), back next Tuesday, and might be just a bit delayed in answering posts during my trip. Again, I really appreciate the discussion.

Mr. letiole continues ...

For now, the software is not yet developed, but they are working on it. A few of the obstacles are that it has to work 100 percent of the time with no errors with an almost infinite number of possible permutations. Windows is having quite a difficult time just making a computer run reliably ... and they're trying to do it for a much more important corporate reason - money.

With the computer, it has to operate 100 percent accurately and perfectly all the time. It can never freeze up. A computer-based system will have no human back-up as once the controllers have lost their proficiency or left their posts, it's all electronic. The current system allows for mistakes by controllers, pilots and computer and relies on controllers to rectify their mistakes in a timely manner. I'm confident that software will be available at some point, maybe not even that far distant, but then you'll be looking at the same issues that pilots are looking at now, which is what happens if ... What I mean by this is, modern aircraft can take off, land, taxi, fly themselves, they don't need pilots. In fact, Airbus prefers pilots keep their hands off. Problem is how safe do you feel in the back without a pilot? And, all these perfect systems on aircraft with five levels of redundancies fail at what would be a frightening rate without a pilot in the cockpit. As an aside, I'm not too worried about my job because I retire relatively soon (and automation would probably bring with it a nice buyout package), but I do see major obstacles to a fully automated system.

As far as en-route congestion goes, the highest paid controllers in the country are at the 20 en-route centers, plus Chicago, New York and Bay approach. There's a reason for this, the skies are actually incredibly crowded en-route. For example, the airspace bounded by Yosemite on the West, Reno on the North, Mina and Coaldale, Nev. on the East and Bishop on the South handles about 40 aircraft at a time on a routine basis. This doesn't sound like a lot, however, they're going all directions plus up and down and the closure rates are more than 1000 mph. This area is about 100 x 100 miles. The typical aircraft transiting this airspace takes about eight minutes and routinely would have three to four conflictions. Multiply that by 40 aircraft at a time and you're getting into a complex computer program or, under the protected bubble theory, you're still getting an awful lot that has to be taken into account with projected courses and aircraft that your plane is trying to miss moving from where they currently are to avoid additional traffic.

I agree that there is a lot that can be done and that the software will sometime be available. Technology will, I hope, do a lot to help in the future. In the meantime, as in tomorrow, we could reduce separation standards, which would get aircraft to their destinations and to their requested altitudes quicker thus saving fuel and time. For oceanic flights, we could use the GPS satellite data you mentioned to give the controllers a kind of radar for the ocean, instead of the pieces of paper that we currently use. There are also obvious airline fixes such as staggered scheduling, moving hubs (why would you hub at ORD?), flying larger planes less frequently, etc.

There's a lot to be done, but I don't think we're within even five or 10 years of true en-route automation.

If you are interested in seeing a center or a Tracon work, check your government pages for DOT. The facilities are generally happy to give a tour. But make sure they let you see a busy sector. If not, you'll be convinced we don't need automation or controllers.

As an aside, about 18 months ago two Fed-Ex DC-10s - one West Coast to East; one Northwest to East - both lost radio contact with ATC. About an hour later they crossed in Denver Center's airspace at a relatively low-density area. It's estimated they missed by less than 100 feet. I'm not sure if they had TCAS, although I believe they did. At any rate, there are a lot of system users who don't have and don't want to put in a lot of expensive equipment and the FAA has been reluctant to mandate for other than passenger carriers. ... and it just takes one guy running a stop light (or having equipment fail) to make a mess.

Well, I don't think that anyone is proposing any sort of fully autonomous en-route traffic control. Just a decentralization of the decision-making process from controllers to pilots. The stakeholders will still be in the loop. As such, the requirements on bulletproof software drop dramatically.

With more freedom to choose direction of travel, the congestion in the areas you speak of would naturally decrease due to normal diffusive processes. Pilots would naturally fly where there were less problems with traffic, and decisions to avoid potentially congested areas would be made well in advance, so that total flight path-length would only increase by a tiny fraction.

It has been a number of years since I was last up in a civilian tower, and I've not yet been to a TRACON, so I appreciate your suggestion.

Finally and BTW, on additional reflection on your traffic analogy, if all the lights, signs, lanes, etc., were removed in Rome, nothing would change http://www.flyertalk.com/forum/wink.gif