Real time tracking, FDR transmission needs to happen now

As time passes following the Dec. 28 crash of AirAsia flight 8501, it has become even more apparent that the need for real time tracking of the position of a commercial flight and real time transmission of flight data recorder information is necessary.

The issues aren’t new. Airplanes have gone missing since the dawn of aviation. Real impetus began to pick up after the 2009 disappearance of Air France flight 447. This flight went down over the South Atlantic in proximity to major thunderstorms. Radar coverage indicated the flight track but not the final position. ACARS, the system monitoring the health of the aircraft’s systems, transmitted enough information so that the airline, investigators and Airbus had a reasonable picture of what happened but a picture that was incomplete. It took five days to locate the first debris and bodies in the water and two years to locate and retrieve the black boxes, which filled in the picture of a flight crew in complete befuddlement over what was happening to their airplane. The accident was entirely avoidable, and the circumstances incorporated into future training may well save lives.

Despite the delay in finding the airplane and black boxes, the airline industry and regulators did nothing to require real time tracking and data transmission.

Then came Malaysia Airlines Flight 370. The flight, in clear weather, disappeared. The transponder and all radios were shut down. Had it not been for a system attempting a “handshake” with satellites, authorities would have had no idea where to look. The flight is still missing nine months later. Real time tracking and a fool-proof FDR transmitting system would not have prevent what is presumed by most to be a criminal act taking over the airplane, but the aviation industry would have a much clearer picture of what happened from FDR information and it would know where the airplane is.

But, predictably, the industry did nothing again.

Now there is AirAsia 8501. It disappeared in heavy thunderstorms, in an area where radar and radio coverage exists and nine days later authorities still haven’t definitively found the airplane, let alone its black boxes. This hasn’t stopped media and so-called experts, and even an Indonesian government agency, from declaring the cause of the accident and elements of the incident. Supposedly, any of the following happened:

  • There was a high altitude stall.
  • The engines iced over.
  • There was a safe water landing (in waves 3-5 meters!) before the airplane was swamped.
  • The airplane broke up in mid-air.
  • The airplane impacted the water in one piece.
  • The pilot raised the nose of the airplane too high and stalled (this one ignores the Airbus A320 envelope protection). CNN raised this prospect and should be embarrassed for its ignorance of the A320.
  • The airplane descended at between 11,000 to 24,000 ft per minute.
  • The airplane ascended up to 9,000 ft per minute.
  • The use of the metric system might be responsible. (Fox News on this one.)
  • The pilot opened his cockpit window and waved at Santa Claus as he returned to the North Pole. (For avoidance of doubt, this last one is fiction, as are probably most of the above.)

It’s one thing to mention many of these, and other theories, as possibilities to explore (as we have done); it’s quite another to make declarations that this-or-that is what happened.

According to a news report, real time tracking is something AirAsia was already, on its own, beginning to install on its fleet. The airplane in question hadn’t been modified yet. Had it been, the location of the airplane would have been know right away. Had there been a safe water landing, as with US Airways 1549 in New York’s Hudson River years ago, getting to the site rapidly could have meant the difference between life and death. Had there been real time FDR information transmission, all these wild theories wouldn’t be floating out there. That they are is to the great disservice and most likely distress of the families and friends of the victims on the flight.

For the airline industry, installing real time transmission isn’t, on a per-plane basis, all that expensive. Canada’s First Air has locator transmission on its small fleet of Boeing 737s because it serves far north Canada, outside of radar coverage. The equipment is C$110,000 per plane. For an American Airlines or a Delta Air Lines, we’re talking real money with these large fleets.

For the industry, it all comes down to costs and in this context, dead people don’t matter, only cost matters. It’s the infamous tombstone mentality that enough people have to die before there is enough of an outcry to force regulators to do the right thing and force the airlines to follow. First Air and even AirAsia are to be commended to being leaders in real time location transmission.

Far more needs to be done.

49 Comments on “Real time tracking, FDR transmission needs to happen now

  1. I do not catch the rationale behind the statement that tracking equipment is too much money for large operators. If American Airlines and Delta Air Lines generate as little money as First Air, that is not the cost of tracking equipment they should be worried about.

      • Is it necessary to equip ALL aircraft in the fleet with such tracking? I’m of the opinion that it is most important for those aircraft that are used on over water flights.

        • Or to be more precise, over areas where aircraft tracking is inadequate.

        • Over the jungles of South America, Asia, African deserts and mountains, etc., also qualify. Remember that MD-80 that crashed last year (weather) in Africa wasn’t found right away, either.

          • if you look into the supplier of the technology (Flyht Aerospace) to First Air, there is actually a less than 1 year payback on the AFIRS 228 equipment, so while there is a cost to installing the equipment there is actually an economic payback on the investment. It’s too bad that so many airlines are too closed minded to not even consider the big picture.

          • Remember that MD-80 that crashed last year (weather) in Africa wasn’t found right away, either.

            Air Algérie 5017 crashed on 24 July 2014.
            Yes, there were initially conflicting reports on where wreckage had been found. Still: Within two weeks (by August 7th), the wreckage was found, data recorders retrieved and analysed and a small press conference held that outlined the first results of the analysis.
            A preliminary report from the BEA was released less than two months after the crash.

            As with QZ8051, I’m not sure that Air Algérie flight is a good example for the need for realtime tracking.

            Mind you, I’m not against realtime tracking per se – but I think we should think twice about why it should be introduced and what we want to achieve by it.
            To be honest, avoiding bad reporting in the media shouldn’t be that high on the agenda, because more data will just give them more material to spin theories off from.
            Even if you have all the telemetry data you could wish for, it won’t prevent the UK Daily Mail writing an article saying the plane was struck by lightning (or any other unlikely scenario you can think of), and some other blog, newspaper, etc. will copy and multiply the nonsense. You won’t be able to keep up with that, and I don’t think aviation safety equipment should be designed/introduced to do that, either.

          • I think one of issues of fixed FDRs and CVRs is that it is not entirely damage proof. The Air Algerie crash lost its CVR data but fortunately, at least the FDR data were recoverable.

          • I think one of issues of fixed FDRs and CVRs is that it is not entirely damage proof. The Air Algerie crash lost its CVR data
            True, because it was an old style magnetic one which also didn’t erase old recordings before recording on top of them.
            These days, recorders are Solid State Disk-based, which is a much more reliable technology – AF447 had this type on board and I would expect the same for QZ8051.

      • I do not argue on the magnitude of the investment, just saying that if First Air can afford it, other airlines should be able to. Unless their operation is in terrible state, profit wise.

        • Yeah, that didn’t make sense to me either. Obviously the total costs get larger for a big fleet, but if anything a larger airline should be *better able* to install equipment like this as standard…

      • ONE seat on a long range airliner could generate gross revenue per year of over ONE million dollars

        • $1m revenue … and 999,999$ cost for $1 profit 😉

          Revenue is nice but note you can “die” on too much revenue.

          Then I don’t think that the “duh obvious and so cheap” solutions offered would actually achieve their objective.

          It is suprisingly difficult to write up rules to catch these special conditions. My experience is that they tend to work perfectly for hindsight but flounder when mother nature has another LSD dream original.

      • With a large fleet you would have big discounts.

        That also discounts if the Northern Canada operation is the right solution.

        I am certainly open on the issue, maybe something is better than nothing.

        Position tracking would seem to be the first immediate need. Keep in mind they have not picked up any pings which again brings up questions if that system needs a review as well.

        Frankly if we can find the wreck the FDR streaming is less important.

        And I do not see this as saving the lives of those on board. It might save lives in the future but only if something is implemented as a result of the findings. While there is movement on the lack of pilot skills its still moving slowly.

        Boeing refuses to do anything about its finagally Auto Throttle system.

        I had missed the point on the A320 and Airbus control logic assuming it was in flight envelope protection, i.e. if kicked into alternate control or direct.

        • Correct me if I’m wrong. I think First Air uses exactly what you are describing – “triggered transmission”.

  2. ” Had there been real time FDR information transmission, all these wild theories wouldn’t be floating out there.”

    Media appear to be so used to inventing out of whole cloth that early FDR information wouldn’t make a difference.
    ( You’d still find jokers that have a much more sexy story to tell _and_ that will be printed first. )
    What we do need is “media sanitation/reeducation/.. ” what ever you want to call it : A return to reporting facts.

    Is there a recent case around where early information could have saved lifes?

  3. It is a lot of money, but the fact is, governments are picking up the cost after the mishaps, when they expend millions in fruitless searches. It is entirely appropriate for regulators to require a basic data bundle be generated to ensure support of these search and rescue missions. These systems are needed over water, and over remote (e.g., Arctic) areas. The data bundle can be as simple as just three basic parameters: position (lat/long), altitude, and indicated airspeed, transmitted once per second. The transmission can be triggered by odd parameters (e.g., if the system notes excessive pitch angle, and/or bank angle, or substantial heading/speed/altitude changes within the previous 15-seconds), in which case two or three other parameters could be easily added to the bundle. If the transmission is triggered during a flight, the shutoff/override would NOT be in the aircraft, but by the ground dispatch/monitor personnel, who would need to communicate via ACARS etc. to ensure the transmission is an anomaly, not a real emergency.

    All things considered, it seems this type of system should cost a lot less than $100K to install, and should cost virtually nil to operate … yet, when it is triggered, would save millions in overall costs. As noted in the Post, the present system is a painful and gross disservice to passengers and families, but it also undermines aviation in general. This does need to change.

    • the manufacturing cost of satellite smart phone for once per 5 seconds data reporting (GPS update rate is 5 second intervals), capable of independently monitoring GPS position/velocity/vector and receiving a minimal diagnostic data input feed (and power) with battery backup is on the order of $10.

      add the external RF transparent blister and all mounting hardware, we are talking under $500 parts cost (and likely it could share an existing RF blister reducing costs).

      the big cost is initial integration with any given airframe design, but once that has been done per unit costs should be very low.

      data requirements for ~1K of data every 5 seconds 24/7 are roughly 0.5 GB/Mo, which is in the couple hundred a month range. presumably bulk buys by airlines would drive costs down.

      there should be no off switch, this should not be controllable/disableable and should be a go/nogo requirement for all overwater flights

  4. Pingback: Real time tracking, FDR transmission needs to happen now | Aviation Impact Reform

  5. Lorries and delivery vans are GPS tracked by their owners to make sure the drivers are not buggering around .Its not unheard of for an airliner to go off in the wrong direction or land at the wrong airport. I think this is the key to it,getting more than just the position of crashed planes.Obviously it’s important that someone is actually looking if it’s to be useful.Just the fact that someone has turned off the signal is very interesting.I don’t really understand the technicalities of it,why is it so expensive when they are already transmitting maintenance data?

    • You can also track your kid by the cell phone GPS—but not an airliner….

      • Yes I did know that. I was just comparing the situation.with the mh370 situation it appears that there was information from ground based radar tracking,but no one did anything.same thing with mh17,the information was available to whoever shot it down they just didn’t bother to use it.I prefer the principle of spending money on preventing crashes and it would be nice if all this technology had a wider use than finding aircraft that have already crashed.

      • You can track your kid by the cellphone GPS – as long as it is within range of a cell tower. It certainly isn’t sending data via Iridium or Inmarsat. I don’t know if the trucks are relying on the cell phone system these days or not. Ships, sure, inmarsat or whatever it happens to be called.

        • Afaik Thurya is the only service that has precise position reporting ( via GPS) in the default configuration. All user nodes handshake with position data regularly.
          For Iridium phones it is an option available on some enduser equipment.
          for the terrestric systems ( GSM, … ) you have basic localisation via the system infrastructure.
          for all types reporting GPS position over a dedicated third party data service is a valid option.
          ( Like for the kiddie and pet trackers that use SMS messaging as one possible comms path .)

  6. The simple answer would be to get the insurance industry to fit the equipment and add it to the premium charged for the hull insurance.

    • Interesting proposition. But I’d have thought that the insurance companies should charge lower premiums to those who equip their planes than those who do not. Sort of like driving insurance monitoring devices.

  7. Yes, Andrew, you’re spot on… It will have taken three disasters to have ICAO and the industry seriously beginning to address the problem. QZ8501 hopefully will be the last
    aeroplane to be lost and untraceable just because the airline industry feels it’s too expensive to install a few hundred thousand dollars worth of the right hardware. I understand that AirAsia had already begun to do this BEFORE losing its A320. Congratulations to this very, very safe airline. Just my few cents. Jerome, Capt 332 Rtd.

    • QZ8501 hopefully will be the last aeroplane to be lost and untraceable

      Except it wasn’t really. Debris and bodies were found already, within 2 days of the crash – which is remarkable considering the weather in the region.
      (Yes, that’s the same weather that may have played a role in the crash but that everybody seems to ignore in their demand for more answers right now or better even: yesterday!)

    • It doesn’t bring any increase in savety. Not a single life is saved by FDR streaming ( just like nobody is saved by video surveilance in public spaces. What you might get is better material for post incident litigation. )

  8. With all due respect, Scott, I don’t think QZ8051 is a flight that shows how great it would be to have real-time streaming of FDR data.
    SAR knew quite soon where to look for the plane (first pieces of debris were found only 10km away from the plane’s last known position) – the biggest problem for them was and continues to be the weather. The area isn’t easily accessible at the moment.

    But: They’ll get to the black boxes, and they have a pretty good idea where they are, but it takes longer to get them because the weather is so bad in the region at the moment.

    From all we’ve seen so far, real time data and voice streaming would not have helped save the plane. Bodies were found still strapped to their seats and the debris we’ve seen so far was quite small, indicating that the plane broke up either in flight or upon impact.

    If you assume the pilot was able to make a successful water landing (the only context in which realtime data streaming might have made a difference) you also have to assume that the pilot was not only aware of a problem, but also in full control of the plane. In which case an emergency call would likely have been possible.

    But as it stands, all that realtime data streaming would have done in this case is giving the media more information to take out of context and speculate upon.

    • Imagine we had know exactly where the plane hit the sea – we’d still be where we are now. With no access to the wreckage because the conditions aren’t safe for man or machine. Granted, we could’ve possibly got there a day or two earlier. But then – maybe not, because some debris floats, some doesn’t. Both types of debris will move in the sea even in good weather, but not necessarily at the same rate or in the same direction. All of this becomes more complicated if the water is deeper (think AF447) and the sinking wreckage travels through multiple layers of water and currents.

      Again – I’m not against realtime tracking per se, but we should be realistic about what we can expect to gain from it.

    • “Bodies were found still strapped to their seats and the debris we’ve seen so far was quite small, indicating that the plane broke up either in flight or upon impact.”

      – I’m not sure we can come to this conclusion this easily. We’ve seen a few debris, some seats, life rafts, some luggage, a couple of doors and sidewall panels. But it somehow seems very little for a such a scenario you describe.

      When AF447 which impacted the water at a very high vertical velocity, among the mangled floating mess recovered, you had the tail fin, cabin fittings, belly fairing, wing parts, a galley, and a whole lot of other pieces of debris (all showing a high degree of deformation). So far, I’ve not heard much in the same way of the Air Asia aircraft in terms of structural debris recovered apart from the doors.

      http://www.bea.aero/docspa/2009/f-cp090601e2.en/pdf/f-cp090601e2.en.pdf

      IF the reports of the 4-5 large objects (one of them 18m long reportedly) detected near each other indeed turn out to be the Air Asia aircraft, those are some fairly huge pieces considering that the A320 is only 37m in length.

      • Ethiopia flight 961 proves it is possible for at least some passengers to survive an airliner breaking up whilst ditching.Would the waves in this case have definitely prevented it?

          • I read somewhere today that a rescue pilot claims to have seen 2 survivors in the water holding hands. Can’t find it again to provide a link. I’m not sure I believe it.even if true could rescuers have got there in time? It’s not impossible, 2 of the pan am 103 passengers are believed to have survived the desent. I know its unlikely, but it’s not a nice thought that people are waiting for rescue and no one knows that it’s crashed or where to look.

          • I read somewhere today that a rescue pilot claims to have seen 2 survivors in the water holding hands.

            That has long been debunked and untrue, unfortunately.

  9. I’m going to play contrarian for a moment or three. When it comes to the matter of improving safety for future travel, that it took two years to get the black boxes for Air France 447 would only “really” matter if there had been another accident like it in that time. A similar argument could be made for Air Asia 8501 – unless a similar accident happens before the black boxes are located, whether it takes two hours, two weeks, two months or two years doesn’t really matter.

    Sure, not KNOWING sucks, and we are all increasingly accustomed to instance gratification which means we increasingly want to KNOW NOW, but thus far has there been an accident in the last 30 years where we didn’t get data from the black boxes until after the next similar/identical accident took place?

    • AF447 black boxes may have taken two years to recover, but a likely cause was already pinpointed from the ACARS data it sent. This was critical since it hastened the implementation of the pitot tube fix. The sooner you know the cause……

      • The pitot tube changes weren’t really a full fix.

        One finding in the BEA report is that the use case was beyond the expected environmental conditions and thus the certification scope. Synopsis: further research to scope out recent changes in atmospheric conditions. Afaik Airbus sent an A340 down under for further data acquisition. Haven’t followed this any further, though.

    • Yes. There is a strong voyeuristic aspect to wanting “InstaCrash Information”.
      People seem to have lost patience and the will to handle unpleasant information.

  10. Sadly they know about that stuff already, took the wreck to get them in serious high gear.

    Did they really have them on Radar that far out? I had not thought so.

  11. Last August, Barry Eccleston, President of Airbus have appointed director of the board of Flyht Aerospace…About FLYHT: «provides proprietary technological products and services designed to reduce costs and improve efficiencies in the airline industry. The Company has patented and commercialized three products and associated services currently marketed to airlines, manufacturers and maintenance organizations around the world. Its premier technology, AFIRS™ UpTime™, allows airlines to monitor and manage aircraft operations anywhere, anytime, in real time. If an aircraft encounters an emergency, FLYHT’s triggered data streaming mode, FLYHTStream™, automatically streams vital data, normally secured in the black box, to designated sites on the ground in real-time. The Dragon is FLYHT’s latest product, a revolutionary light weight portable satellite communications device that blends existing FLYHT technology with that of the iPad.»

  12. I’m not sure were really addressing the problems, which are:
    1. Timely tasking of SAR assets (in the case of an accident where there have been survivors)
    2. Cost effective recovery of the CVR/FDR
    The fact is, 406mhz ELT’s can help with both of the above, the problem is the only automated ones are the fixed ELT’s which are as much use as a chocolate teapot. Various military and civilian aircraft use deployable ELT’s which would address this problem (providing the ejected airfoil was of the type containing the FDR/CVR data), and would provide a proven solution that the NTSB have wanted since 1999 (and Airbus are moving unilaterally to offer them on the A350 and A380 lines). We could be for ever and a day waiting for agreement on streaming services with suitable reliability, when we have a mature 80/20 solution right now.

  13. I second the motion
    there can be no excuse in this day and age why something as valuable as an airliner, not to mention it’s passengers, is not at least tracked remotely (and maybe there should be possible overrides as well)

    remember the initial satellite track gave two circular paths MH370 could have taken, and one went over Afghanistan and/or Ukraine and terrorist activity was considered?
    this (http://en.wikipedia.org/wiki/Satellite_Internet_access#Portable_satellite_Internet) doesn’t look expensive in airline dollars…

  14. Leeham: Your simplistic argument, that the aviation industry should spend any amount to add real time telemetry immediately, is weak. You can and should strengthen your argument by identifying the alternative number of lives at stake, cost, and other safety and maintenance and design issues. For example, exactly which safety items do you intend to postpone or cancel, to attack this one first? How many lives do you save? At what cost?

  15. Interesting article, people need to know that they are safe in the air and that if something was to happens their loved ones would be able to find them! The cost might be high to put tracking on every plane but if you are flying over an area with bad tracking the plane should require tracking!

  16. It is true that the cost of real time tracking is extremely costly but what is the cost of safety? It doesn’t have to be on every plane immediately but it should be looked at more in depth.

Leave a Reply

Your email address will not be published. Required fields are marked *