Why not have real-time black box reporting from aircraft?

probe

The only hard information they seem to have on the Air France crash is the ten "text messages" sent by the A330 to the Air France real-time monitoring system in Paris.

Airlines can provide internet access to passengers. Why not use the same infrastructure to provide a real-time feed of black box data back to the airline HQ or the aircraft manufacturer's servers? In addition to, rather than in substitution for the physical black box.

While there may be difficulties keeping the data link up and running during certain types of emergencies, it would provide useful information in a more timely manner. And probably provide information on the cause of accidents where the black box can't be located or is damaged beyond repair.

It would also facilitate proactive measures to prevent accidents by using real-time data mining on incoming signals.

Tenger

I am of the opinion that this incident (whatever the final outcome) may act as spur on the aviation inductry to create some sort of global monitoring system like this. This is the first incident of this magnitude by an A330, and I believe, from an Air Accident Investigators point of view the ACARS system has already greatly advanced the investgation. The automated ACARS message as you say are the only info so far on the possible cause.

The capability certainly exists on current and planned longhaul aircraft (A330/340, A350, A380, B747-800, B777, B787) so we may well see this technology/procedure implemented on all long overwater flights.

Another possiblity is the development of a 3rd 'black box' device but this one would be designed to detach automaticall from the aircraft in a castatrophic situation. I don't see this as being probable, aircraft manufacturers would prefer to try to prevent future accidents rather than ensure we get full info on any them situation.

Dartz

It'd be too expensive.

It'd be a logistic nightmare maintaining servers to record hundreds of thousands of aircraft, each with hundreds of paramaters, coming from all across the world. Good in theory, but a bloody expensive pain in practice it seems.

Who's responsible?

The aircraft operator?
The aircraft owner?
The aircraft manufacturer?

Who pays the charges to send this infor via satellite?

Will it be retrofitted to existing craft?

It's the Ford Pinto equation again.

partyatmygaff

A private company can be subcontracted by airlines to offer this perhaps with wifi and phone access as an extra to help pay for the system.

One private company starts it up, they manage the servers and sat. space and they relay the info back to KLM/AF headquarters and ATC etc...


Just needs alot of starting capital though.

probe

Dartz wrote: »

It'd be too expensive.

It'd be a logistic nightmare maintaining servers to record hundreds of thousands of aircraft, each with hundreds of paramaters, coming from all across the world. Good in theory, but a bloody expensive pain in practice it seems.

Who's responsible?

The aircraft operator?
The aircraft owner?
The aircraft manufacturer?

Who pays the charges to send this infor via satellite?

Will it be retrofitted to existing craft?

It's the Ford Pinto equation again.

Just like google.com. A logistical nightmare tracking website changes in near real-time, news feeds, providing email, social networking, google maps, mapping people walking around with their mobile phones, satellite imagery, google apps..... etc providing service to billions of users all over the planet. Can't be done....

In terms of legals, it needs to be done on a "best efforts" basis, with laws implemented (or carriage conditions modified to make the PAX agree to best efforts).

Who pays the data transmission charges? The same person who pays for the fuel. The data transmission charges would be infinitesimal in the scheme of things.

probe

No country is better placed to fix this issue on a global basis compared with France. Aside from Airbus being based in Toulouse. And thalesgroup.com who provide much of the worldwide ATC kit, and the suspect Pitot tube on the A330*.

France is the only country on the planet with “counties” on virtually every continent.... If something requires a global infrastructure deployment.... France can do it. http://upload.wikimedia.org/wikipedia/commons/1/16/Outre-mer_en.png. The French city of Cayenne is within a few hundred km of the Air France crash site. It has a French postcode (97300) just like Paris or Lyon, is in the EU, the Euro is the local currency, and they vote in the French and EU elections. Unlike their neighbours across the river in the third world Dutch colony of Suriname. The yellow La Poste van delivers letters every working day, right up to the border between France and Brazil. And Électricité de France is the local ESB**. And they have regular 10 digit French phone numbers, even though they are 7,000 km from Paris!

Not forgetting http://www.arianespace.com that has the satellite capability to launch the satellite kit to facilitate real-time data logging.

Pourquoi sommes-nous attendre, SVP?! La France n'a pas d'excuse.

*http://www.bloomberg.com/apps/news?pid=20601087&sid=atCz81jNNuUU

**http://sei.edf.fr/accueil/accueil/guyane-particuliers-offres-et-services/informations-et-tarifs-1260673.html

Foggy43

As far as I am aware there are only a handful off airlines that currently have this facility to send maintenance messages. Boeing, in the past, have said they would like all airlines to carry QAR's. AIr France and BA have them. Do not know about Aer Lingus or Ryanair.

Since the ending of the cold war some airlines wanted to use Satcom as a way of ATC communication and ending HF. Again, like QAR's, cost put an end to the idea with the majority of airlines.

BrianD

It appears that this aircraft was sending real time data before it vanished. This is pretty close to what the OP is asking for.

If every aircraft was sending real time flight data i.e. the type that is stored on flight data recorders would the existing communications systems be able to handle it all? I don't believe that there would be any benefit to anybody in having all this data available and transmitting it around the world. The number of aircraft disasters is minute compared with the number of flights every day so most of the data collected would be mundane and with no real use.

Most aircraft incidents seem to happen in a matter of minutes so the opportunities for third party intervention is slim. Furthermore, most incidents involve a combination of factors which come together e.g. weather, training that are external to the technical performance of the aircraft. The other issue is that would pilots have the "time" to liase with third parties? It seems that in many disasters the pilots were struggling to understand the behaviour of their aircraft believing that instrument readings were incorrect. They would also likely to be precoccupied with phsyically flying the aircraft. I recall the conversations between the US Air flight crew that ditched in NY and ATC were extremely brief.

In most cases, the black boxes are recovered. So the combination of warnings transmitted in real time and the recovery of the black boxes is still the most effective. The authorities have indicated that they appear to have enough information to work out what happened. Accepted that this woud be definitive if the FDR was recovered. I woulc imagine that there is a strong chance that the Air France box will be recovered once the actual crash site is located.

Celtic Mech

I have seen the messages from the Air France that were transmitted via ACARS. My guess is Airfrance may use the Airbus system called AIRMAN. This is a real time Maintenance programme that reports back maintenace details back to the airlines Maint/engineering dept. You can see the warnings the pilots see, the source of the problem and what computer has identified the problem. you also see the time of when it happened and also the phase of flight it happened. Its very handy and allows you to prepare for aircraft incoming with problems, but also helps via preventitive Maintence...i.e. following up on very minor faults before they may escalate and possibly cause larger faults.

probe

BrianD wrote: »

It appears that this aircraft was sending real time data before it vanished. This is pretty close to what the OP is asking for.

If they got everything online - why are they bothering to search for the "black box"?

We really need an IT upgrade for the entire air travel network.

1) The high resolution weather forecast model - the best features of the Swiss / German COSMO meteo system and HIRLAM, covering all of the main air routes, end to end. Many/most countries don't have precise weather forecasting infrastructure.

2) Blow by blow real-time reporting of black box data. In the same way as one has real-time backup for computer systems in retail, financial services and other industries. You can't rely on one or two copies of data - especially in an air accident - aside from the benefits of timely reporting. If you went to your local bank to withdraw EUR 200, and they told you sorry we have lost our data, your account no longer exists on our books. You wouldn't be too pleased. Which is a minor issue in comparison with having your life terminated in an air accident....

3) The availability of a realtime firehose of data would facilitate data mining to help identify problems that involve environment, meteo, aircraft trajectories, to provide enhanced ATC/flight management and greater safety. There is far more data out there than a single aircraft system has access to. These data need to be consolidated in real-time from all available sources.

4) It could provide part of the infrastructure to allow direct flight routing between origin and destination airports. One can't but help getting a bit pissed off watching the route of a commercial flight on one's GPS device - (back seat driver mode in row 7!) usually way off the mark following a point to point marker route, rather than travelling in a straight line to the destination airport! Wasted air-kms, wasted fuel, wasted time.....! This is 2009, not 1909....

partyatmygaff

probe wrote: »

If they got everything online - why are they bothering to search for the "black box"?

We really need an IT upgrade for the entire air travel network.

1) The high resolution weather forecast model - the best features of the Swiss / German COSMO meteo system and HIRLAM, covering all of the main air routes, end to end. Many/most countries don't have precise weather forecasting infrastructure.

2) Blow by blow real-time reporting of black box data. In the same way as one has real-time backup for computer systems in retail, financial services and other industries. You can't rely on one or two copies of data - especially in an air accident - aside from the benefits of timely reporting. If you went to your local bank to withdraw EUR 200, and they told you sorry we have lost our data, your account no longer exists on our books. You wouldn't be too pleased. Which is a minor issue in comparison with having your life terminated in an air accident....

3) The availability of a realtime firehose of data would facilitate data mining to help identify problems that involve environment, meteo, aircraft trajectories, to provide enhanced ATC/flight management and greater safety. There is far more data out there than a single aircraft system has access to. These data need to be consolidated in real-time from all available sources.

4) It could provide part of the infrastructure to allow direct flight routing between origin and destination airports. One can't but help getting a bit pissed off watching the route of a commercial flight on one's GPS device - (back seat driver mode in row 7!) usually way off the mark following a point to point marker route, rather than travelling in a straight line to the destination airport! Wasted air-kms, wasted fuel, wasted time.....! This is 2009, not 1909....

There is always a reason, airlines don't just do this to waste time and fuel. It could be restricted airspace, bad weather, busy air traffic in the area or whatever really.

wylo

partyatmygaff wrote: »

There is always a reason, airlines don't just do this to waste time and fuel. It could be restricted airspace, bad weather, busy air traffic in the area or whatever really.

not to mention the fact that they must be within certain ranges for emergency landings to airports afaik

BrianD

probe wrote: »

If they got everything online - why are they bothering to search for the "black box"?

We really need an IT upgrade for the entire air travel network.

1) The high resolution weather forecast model - the best features of the Swiss / German COSMO meteo system and HIRLAM, covering all of the main air routes, end to end. Many/most countries don't have precise weather forecasting infrastructure.

2) Blow by blow real-time reporting of black box data. In the same way as one has real-time backup for computer systems in retail, financial services and other industries. You can't rely on one or two copies of data - especially in an air accident - aside from the benefits of timely reporting. If you went to your local bank to withdraw EUR 200, and they told you sorry we have lost our data, your account no longer exists on our books. You wouldn't be too pleased. Which is a minor issue in comparison with having your life terminated in an air accident....

3) The availability of a realtime firehose of data would facilitate data mining to help identify problems that involve environment, meteo, aircraft trajectories, to provide enhanced ATC/flight management and greater safety. There is far more data out there than a single aircraft system has access to. These data need to be consolidated in real-time from all available sources.

4) It could provide part of the infrastructure to allow direct flight routing between origin and destination airports. One can't but help getting a bit pissed off watching the route of a commercial flight on one's GPS device - (back seat driver mode in row 7!) usually way off the mark following a point to point marker route, rather than travelling in a straight line to the destination airport! Wasted air-kms, wasted fuel, wasted time.....! This is 2009, not 1909....

Because obviously the black boxes will provide more definitive information and allow a complete recreation of the events including what the pilots were saying or doing.

What I am saying that there is no economic or human value in 100% real time "black box reporting" for all aircraft flights. We would have an over supply of largely redundent data being transmitted around the world. We are gathering data for the sake of it with no actual benefits whatsoever. However, it seems that this ACARS system already has much of the functionality you refer to by transmitting summary data.

In any case, many air crashes are a result of a very specific sequence of events that are unique to that incident. They may never be replicated again. In the case of a design flaw or mechanical failure, quite often these will begin to manifest themselves without causing an aircraft to fail. This is where this simple in flight data transmission comes in useful. I am guessing, that probably a lot of this data is probably recorded somewhere on the aircraft anyway and can be "downloaded" during maintinance. The problem is that sometimes it takes time for a clear pattern to emerge and an engineering fix or update to be implemented. Real time data while flying is of limited use.

What's unusual about this incident is that the crash occured outside of radar coverage which just adds to the mystery. I am guessing that if it was within radar coverage, it may not be such a mystery. Also this aircraft is reported to have been transmitting 'alerts' for 4 minutes before the last transmission. I am unsure if these messages include a GPS location but if they did it certainly hasn't helped in locating the crash site.

Would real time black box data transmission have helped at all? i doubt it. It's transmissions would have ended as soon as the electrical systems failed. All it is saving is the cost and time of recovering the black boxes and knowing a bit quicker what actually happened. Would it improve safety? Probably not that much unless a plane was to follow the exact sequence of mechanical, human and weather circumstances that befell the unfortuante passengers and crew of that Air France flight.

pclancy

Maintenance info or warning messages are one thing but how the hell could you transmit the voice recordings of the cockpit and 88+ different flight data measurements for so many flights at the same time anywhere on the planet, its just not really possible right now and the exisiting system of black box recording makes much more sense considering how rarely aircraft crash.

Also...how can you pick up GPS from within the cabin? I thought you had to have line of sight with a few sats to get a fix?

*Kol*

pclancy wrote: »

Maintenance info or warning messages are one thing but how the hell could you transmit the voice recordings of the cockpit and 88+ different flight data measurements for so many flights at the same time anywhere on the planet, its just not really possible right now and the exisiting system of black box recording makes much more sense considering how rarely aircraft crash.

Also...how can you pick up GPS from within the cabin? I thought you had to have line of sight with a few sats to get a fix?

If you are sitting at the window its no problem to pick up the satellites with your GPS.

Biro

BrianD wrote: »

What I am saying that there is no economic or human value in 100% real time "black box reporting" for all aircraft flights. We would have an over supply of largely redundent data being transmitted around the world. We are gathering data for the sake of it with no actual benefits whatsoever.

I wouldn't worry about the over-supply of redundent data - we have that already in shows like big brother. Anyway, the data sent would be fairly small. Voice would be the biggest, and they can compress that significantly. You could over-write the data every flight or every second flight etc.
In a lot of cases the 30 minutes of data the recorders keep isn't enough anyway. In one case the thing was malfunctioning and recorded the 30 minutes voice before take-off instead of the last 30 mins, but as it turned out that was more helpful than had it just done it's job as they got more info from the pilot's conversation.
I'm all for the OP's suggestion, despite the nay-sayers arguements against!
In any case they could look at what the FDR's record at the moment. Storage is cheap now a days. THey could record the info from the last week of flights if they wanted to.

Owen

I'm for it in principle, and in theory, but in practicality, I can't see it happening for a few years yet for reasons other users have stated here.

Bearing in mind this will be mission-critical stuff, they will need military levels of communication reduncancy, both in terms of signal transmission/reception, and in data backup and interpretation. You'd need a web-farm the size of Google to store all the information, not to mention an indestructable infrastructure to transmit all this data. Worse again, you'll have the various nations squabbling over intellectual property rights and data retention, not to mention any national security issues (If the president of france were to fly, what's to stop terrorists hacking in and seeing the flightplan as the pilots enter it from the cockpit?). There are just too many what ifs for this to be implemented any time soon, which is a pity.

BrianD

But where is the advantage to air passengers and aviation in general?

Bear in mind that aircraft manufacturers spend billions designing and testing their aircraft and they have built in redundencies across a range of systems. They are designed to be safe and survive a vast range of scenarios

In service, these aircraft store vast amounts of data both in the official flight data recorders and whatever other service black boxes they have, These are downloaded and used by both the operator and I presume shared with the manufacturer allowing issues to be tracked and dealt with.

The existing data communications system seems to be more than adequate to deal with communicating real time information in both directions with aircraft.

Then you have the "after the fact" recovery of the data recorders. In nearly all cases these recorders are recovered.

There is zero benefit in building vast networks for transmitting the complex data from aircraft in real time. 99.999% of data would be of no use to anyone. I doubt if the recent tragedy could have been averted had their been real time data transmission from the aircraft.

The speculation now is that pilots were attempting to deal with inaccurate or conflicting flight data. This is the same data that would be transmitted to the monitoring but without the context of what else is happening in and around the aircraft. This seems to be a recurring issue with recent disasters where there is a disconnect between reality and perception. It seems to me, based on the outcome of a number of these disasters, that there is an over reliance on sophisticated avionics and flight data and it is more difficult for the pilot to intervene and fly the aircraft in the "old fashioned" sense.

probe

Biro wrote: »

I wouldn't worry about the over-supply of redundent data - we have that already in shows like big brother. Anyway, the data sent would be fairly small. Voice would be the biggest, and they can compress that significantly. You could over-write the data every flight or every second flight etc.
In a lot of cases the 30 minutes of data the recorders keep isn't enough anyway. In one case the thing was malfunctioning and recorded the 30 minutes voice before take-off instead of the last 30 mins, but as it turned out that was more helpful than had it just done it's job as they got more info from the pilot's conversation.
I'm all for the OP's suggestion, despite the nay-sayers arguements against!
In any case they could look at what the FDR's record at the moment. Storage is cheap now a days. THey could record the info from the last week of flights if they wanted to.

They can always make the data anonymous by removing fields which would identify a specific flight or crew, a short period after the flight has landed safely. The remaining data would be available for data mining in relation to aircraft type, flight routing, and the airframe identity for maintenance and longer term research.

Biro

BrianD wrote: »

But where is the advantage to air passengers and aviation in general?

Bear in mind that aircraft manufacturers spend billions designing and testing their aircraft and they have built in redundencies across a range of systems. They are designed to be safe and survive a vast range of scenarios

In service, these aircraft store vast amounts of data both in the official flight data recorders and whatever other service black boxes they have, These are downloaded and used by both the operator and I presume shared with the manufacturer allowing issues to be tracked and dealt with.

The existing data communications system seems to be more than adequate to deal with communicating real time information in both directions with aircraft.

Then you have the "after the fact" recovery of the data recorders. In nearly all cases these recorders are recovered.

There is zero benefit in building vast networks for transmitting the complex data from aircraft in real time. 99.999% of data would be of no use to anyone. I doubt if the recent tragedy could have been averted had their been real time data transmission from the aircraft.

The speculation now is that pilots were attempting to deal with inaccurate or conflicting flight data. This is the same data that would be transmitted to the monitoring but without the context of what else is happening in and around the aircraft. This seems to be a recurring issue with recent disasters where there is a disconnect between reality and perception. It seems to me, based on the outcome of a number of these disasters, that there is an over reliance on sophisticated avionics and flight data and it is more difficult for the pilot to intervene and fly the aircraft in the "old fashioned" sense.

The recent tragedy wouldn't have been averted, but they're not likely to find the "black box" from it, therefore real time info from the plane would help no end in the investigation. There are a lot of A330's up there - the problem needs to be identified. You don't have to have the data transmitted from millitary or private aircraft, just commercial.
What exactly do the flight data recorders record? Raido communications, control inputs from the pilots and actual responses from the plane, am I right? So basically all the inputs sent to the various controls from the pilots or autopilot, and any movement the controls made, either as a result of an input or at random. Any thing else? All that data could be also relayed real time. You don't have to go mad protecting it, it would just be used along side the FDR info, and in the case where the FDR can't be found it would provide insight that wouldn't otherwise be available.

probe

It seems to me that there are at least three applications for real-time data reporting from aircraft – (a) an immediately accessible, backup on the ground for the black boxes, (b) a systematic reporting infrastructure for defects in critical aircraft parts and (c) real-time accident prevention by data mining the feed for aircraft and ground radar systems.

Data elements which might be collected on the ground – say every 10 seconds:

1. the GPS computed latitude and longitude, “pitot tube speed”, GPS speed, GPS and barometric altimetry (including standby), the degrees of heading, rate of climb or descent...., a turbulence indicator ranging from perhaps 0 to 100 where 0 is totally smooth and as one gets close to 100 (we won’t dwell on that....), logged every 10 seconds back to base with the origin timestamp – shared by airline, ATC, aircraft manufacturer and maintenance service providers – and perhaps others in the safety context.

2. cockpit voices might be continuously recorded and compressed into say 16k bitrate files – but not included in the feed sent to the ground until aircraft systems detect one or more critical events. At that point send the last, say, 10 minutes of cockpit chatter in burst mode back to the ground station. This would reduce the bandwidth required for the overall application, storage and processing, and 10 minutes of voice could be sent in a second or two over a high speed connection. Eg in the Air France case, when the A330 started sending those 24 messages including cabin pressure warnings, dump the datafile with the last 10 or 20 minutes of chat back to ground, and continue to feed real-time cockpit voice to the ground logging system, and the ground systems start “screaming” to get people focused on that flight and flights in the immediate area. By “burst” I mean that the aircraft with the problem gets maximum bandwidth and priority to transfer the data at the highest bitrates. The burst could involve not only voice – but all available data that wouldn’t routinely be fed to the ground system for a normally functioning flight. This would help speed up finding the site(s) of a crash, among other things.

A real-time accident prevention system could be developed over time to data-mine the feed from aircraft, radar systems, meteo satellites and other ground meteo info. The feed could be two-way with observations being sent back from the data mining systems to the cockpit. The feedback could cover collision avoidance with other aircraft and ground obstacles, warnings on speed, altitude and other data errors based on comparison with radar and other systems, up to the minute data on bad weather experienced by other aircraft ahead of you.....

The system could present a real-time turbulence map showing the aircraft ahead of you, and the turbulence (0 to 100) they are experiencing at their location/altitude etc – perhaps with other weather info. The turbulence trend for that zone could be shown with + / ++ / +++ or - / -- / --- indicators, based on reports received in the immediate past by traffic in that zone, to indicate increasing or reducing weather problems. There are zillions of possibilities ... it is a matter of prioritising them and not causing information overload. Information like this would be impossible to deliver without real-time reporting and data mining on a consolidated basis. In an emergency scenario, this system would be working in the background. In normal flying conditions, it would provide up to the minute data to help minimise the need to route an aircraft through problem weather areas.

Storms usually have a “bell shaped curve” where they start off weak, get more violent until they reach a maximum point, and then reduce in intensity. Where you have a continuous flow of aircraft, a consolidated view of each aircraft’s individual experience can help build up the bell shaped curve view of how bad things are at any location and point.

[If you think real-time data analysis is a big job for all aircraft flying (a few thousand of them) Visa and MC are doing real-time data mining on all card transaction authorizations – eg buy fuel more than once over a short period in a gas station in the US (where they don’t have EMV smart payment cards and PINs –ie you just insert your card into the pump and remove it and tank up....), and then go to a Nike shop or similar and buy a pair of shoes and the card you used will be stopped! Because they know from experience when a guy steals a credit card in the hood, he gasses up his car, and his friend’s car, and goes to buy new runners! This involves comparing hundreds of thousands of transactions per second from all over the planet.... probably millions per second at peak periods. ]

Similar concepts could be applied to the air traffic data analysis on a pro-active basis.

Concerns about “spy in the cab” could be addressed by giving the crew – specifically the captain - the power to delete identifying information after the flight has landed safely. The data could be stored in an Oracle database (or similar) with secure* web access. The pilot could go into the secure website, and mark the flight date and cockpit chat fields (if any - because normal flights would not give rise to cockpit chat logging in real-time) for example for deletion. These field deletions could be signed-off by the person in the airline to whom the captain reports.... and they would be zapped. Statistical information would remain for aircraft maintenance purposes. If a captain marked stuff for deletion and it was not counter-signed for deletion, the un-deleted stuff could be visible to her/him the next time they logged in. Everything could remain transparent.

PS: *By secure web access I mean https with userID, password, and multi-factor authentication (a code that changes every minute or so generated by a token**) – to prevent keystroke loggers capturing login credentials and reporting it to third parties. Internet services provided in hotels and internet cafes are notorious for keystroke logging, mainly to steal payment card numbers and email logins. **http://www.rsa.com/node.aspx?id=1156 You don’t want some Chinese or Russian hacker getting into the real-time air safety monitoring system!

BrianD

probe wrote:

It seems to me that there are at least three applications for real-time data reporting from aircraft – (a) an immediately accessible, backup on the ground for the black boxes, (b) a systematic reporting infrastructure for defects in critical aircraft parts and (c) real-time accident prevention by data mining the feed for aircraft and ground radar systems.

I really think that really important point is being missed. Why would we clog the airwaves with streams of redundent data from thousands of aircraft on a daily basis when it can be done in a better way?

Every aircraft probably has it's flight data downloaded on a regular basis and sent through various systems for analysis whether owned by the operator or manufacturer. So this function is being fulfilled. Thus point B, above, is achieved under existing systems.

There is also the existing real time data system which is providing essential data flows in both directions.

In reality, the only flights where there is a demand for real time flight data are the ones that unfortunately crash. The data for these flights is in the black boxes which in 99% of times are recovered. Therefore data is available.

The black boxes only give flight data, voice recordings and data from external sensors from the aircraft. Most aircraft losses result from a unique chain of events of external factors, aircraft specific factors and pilot behaviour. Usually this chain of events happens rapidly. The opportunity to intervene is minimal. Therefore I don't believe that point C above, while desirable, will ever be achievable.

Furthermore, the flight data is only one - albeit vital - component of a crash investigation. In some cases it may be of no use. If the plane is destroyed as the result of a terrorist bomb or a missile then the data wouldn't be the core of the investigations. So having a second back up in point C wouldn't be of much benefit.

I don't think it is reasonably to compare the multitude of credit card transactions transmitted wordwide and real time flight data. The one hige difference is that the banking network transmits USEFUL information. Every data transmission is used to modify somebodys account. Not so with real time flight data where much of the data is useless when the plane lands safely. Also banking networks have access to high speed fixed line communications instead of over the air. Finally, we all know the amount of credit card fraud that exists, look at the delay between transactions and postings on your statement and that there is a lot of time where retail outlets are not connected to the banking network and will happily authorise the transaction anyway.

Also. just on the point of weather and turbulence. There are sophisticated weather forecasting systems for the aviation industry that are in use. They are not just taking off blindly into unknown weather conditions. They have their onboard radar, information from the ground and chatter from other aircraft on the route. Bad weather can be overcome by the aircraft systems and the pilot judgement.

probe

BrianD wrote: »

I really think that really important point is being missed. Why would we clog the airwaves with streams of redundent data from thousands of aircraft on a daily basis when it can be done in a better way?

One would not be "clogging the airwaves" with systematic reporting every 10 seconds or so. The position data (altitude, co-ordinates, speed etc) is no longer than many website URLs. A passenger watching a youtube video on their PC using inflight internet access would be using the same bandwidth a probably 500 aircraft would use reporting their data!

Every aircraft probably has it's flight data downloaded on a regular basis and sent through various systems for analysis whether owned by the operator or manufacturer. So this function is being fulfilled. Thus point B, above, is achieved under existing systems.

Obviously not frequently enough and it is not being consolidated in real-time to provide useful timely information back to the cockpit and on the ground

In reality, the only flights where there is a demand for real time flight data are the ones that unfortunately crash. The data for these flights is in the black boxes which in 99% of times are recovered. Therefore data is available.

It took them several days to find the Air France remains because they had to search a wide area and had no recent position info before the crash. If the infrastructure was there to burst dump the blackbox data to the ground (after sending the last 10 minutes stuff) they wouldn't need to be sending submarines and divers down to -5,000m ocean depths to find a needle in a haystack.

The black boxes only give flight data, voice recordings and data from external sensors from the aircraft. Most aircraft losses result from a unique chain of events of external factors, aircraft specific factors and pilot behaviour. Usually this chain of events happens rapidly. The opportunity to intervene is minimal. Therefore I don't believe that point C above, while desirable, will ever be achievable.

There seems to be a problem determining precise speed and location of the aircraft in remote locations (eg over oceans) in certain instances. If the available data is cross checked with multiple sources, chances are material errors will be picked up in a timely way.

Furthermore, the flight data is only one - albeit vital - component of a crash investigation. In some cases it may be of no use. If the plane is destroyed as the result of a terrorist bomb or a missile then the data wouldn't be the core of the investigations. So having a second back up in point C wouldn't be of much benefit.

That is the equivalent of saying that there is no point in a company keeping offsite real-time backup of its computer data, because if the company goes out of business, it won't need the data anyway!

Also. just on the point of weather and turbulence. There are sophisticated weather forecasting systems for the aviation industry that are in use. They are not just taking off blindly into unknown weather conditions. They have their onboard radar, information from the ground and chatter from other aircraft on the route. Bad weather can be overcome by the aircraft systems and the pilot judgement.

They don't have high res weather forecasting for many areas of the world - ie over oceans. No matter how sophisticated the weather forecasting system is anyway, it is impossible for them to predict the level of turbulence with accuracy. If you get an automated report from an aircraft that has past through the airspace in front of you ten minutes ago, that will be far more accurate compared with something from a ground weather forecaster. By consolidating these reports in near real time the flight crew would get more accurate information - particularly if the trend is shown in terms of increasing or reducing levels of a storm condition. This would be far more systematic and useful data compared with chatter with other pilots. The data could also be fed into the meteo models used by forecasting agencies greatly enhancing the information they have to work on.

They already have consolidated near-real time tracking and weather data collection for ships at sea http://www.sailwx.info/shiptrack/shiplocations.phtml - ships travel slowly so the reporting interval can be longer. The aircraft travels perhaps 50 times as fast as a ship. Its reporting frequency needs to be higher too, by at least the same ratio.

probe

BrianD wrote: »

But where is the advantage to air passengers and aviation in general?

Bear in mind that aircraft manufacturers spend billions designing and testing their aircraft and they have built in redundencies across a range of systems. They are designed to be safe and survive a vast range of scenarios

In service, these aircraft store vast amounts of data both in the official flight data recorders and whatever other service black boxes they have, These are downloaded and used by both the operator and I presume shared with the manufacturer allowing issues to be tracked and dealt with.

The existing data communications system seems to be more than adequate to deal with communicating real time information in both directions with aircraft.

Then you have the "after the fact" recovery of the data recorders. In nearly all cases these recorders are recovered.

There is zero benefit in building vast networks for transmitting the complex data from aircraft in real time. 99.999% of data would be of no use to anyone. I doubt if the recent tragedy could have been averted had their been real time data transmission from the aircraft.

I'm not suggesting that they build "vast networks" for this. Use the global IP based networks - i.e. the internet to transfer the data. Sooner or later every airline will have to offer internet access to their customers. It is a simple matter to leverage this platform to carry aircraft data reporting. At the other end of this internet connection, the kit for the servers, database software, secure access control, data mining, etc is available off the shelf.

e.g. http://www.oracle.com/solutions/business_intelligence/data-mining.html

http://www.the-data-mine.com/bin/view/Software/DataMiningSoftware

probe

Cloud Computing above, inside and below the clouds

http://www.ibm.com/ibm/ideasfromibm/us/smartplanet/topics/virtualserver/20090618/index.shtml?sa_campaign=message/leaf1/smarterplanet/cloud

BrianD

Guys this data for the sake of data collection. An absolutely useless purpose that has no real benefits for passengers or the airline industry.

I would compare it to the vast amount of data and metrics that I have available in Google analytics. Vast amounts of info on the 0.11% on average that will click on an advertising banner. Wow!

The only benefit of real time data collection is a speedy investigation and that the television networks will be able to use nice graphics to tell us quickly what went wrong.

The reality is that the current data systems told us that there were issues with these speed sensors, the current systems told us that there was weather issues on the day (and didn't adversely affect any other flight) and the current systems told us that the plane was malfunctioning as the chain of events took place. I challenge you to tell me how real time black box data transmission would have changed the course of the events. Could real time data change the course of events in the 4 to 20 minute window that resulted in the loss of the flight?

Look at ths statistics. In 2008, there was 32 airliner acidents that resulted in 577 fatalities worldwide. 2,948 people died on the roads in the UK. There are an estimated 93,000 flights per day with 8 to 13,000 flights in the air!!! The fact of the matter is that the industry and passengers are better served by R&D and after the fact reporting in the event of the industry.

The fact of the matter that the current recovery attempt is probably cheaper than building a global network to report 99.99% redundent and utterly useless data transmitted from the worlds airline fleet as they fly. Bearing in mind, that the useful data is being captured through service routines and in the majority of cases the flight data can be recovered after the fact.

Also bear in mind that if we have this data capture in place that there is going to have to be a huge human or some sophisticated automated resource on 24 hour standby to be able to interpret and react to the data. Not forgetting the fact that these remote staff will only ever see part of the picture of what's happening at any point in time.

probe

BrianD wrote: »

Guys this data for the sake of data collection. An absolutely useless purpose that has no real benefits for passengers or the airline industry.

I would compare it to the vast amount of data and metrics that I have available in Google analytics. Vast amounts of info on the 0.11% on average that will click on an advertising banner. Wow!

The only benefit of real time data collection is a speedy investigation and that the television networks will be able to use nice graphics to tell us quickly what went wrong.

The reality is that the current data systems told us that there were issues with these speed sensors, the current systems told us that there was weather issues on the day (and didn't adversely affect any other flight) and the current systems told us that the plane was malfunctioning as the chain of events took place. I challenge you to tell me how real time black box data transmission would have changed the course of the events. Could real time data change the course of events in the 4 to 20 minute window that resulted in the loss of the flight?

Look at ths statistics. In 2008, there was 32 airliner acidents that resulted in 577 fatalities worldwide. 2,948 people died on the roads in the UK. There are an estimated 93,000 flights per day with 8 to 13,000 flights in the air!!! The fact of the matter is that the industry and passengers are better served by R&D and after the fact reporting in the event of the industry.

The fact of the matter that the current recovery attempt is probably cheaper than building a global network to report 99.99% redundent and utterly useless data transmitted from the worlds airline fleet as they fly. Bearing in mind, that the useful data is being captured through service routines and in the majority of cases the flight data can be recovered after the fact.

Also bear in mind that if we have this data capture in place that there is going to have to be a huge human or some sophisticated automated resource on 24 hour standby to be able to interpret and react to the data. Not forgetting the fact that these remote staff will only ever see part of the picture of what's happening at any point in time.

Please take your head out of the sand! I suspect you have problems about real-time spying on what flight crew might do.... no thinking person wants that, aside from you know who.... Which is why I have suggested that the Captain gets access to self-service deletion and proof of deletion after the flight - of any data that might identify the crew involved or the specific flight in question.

Most of the data is irrelevant after a successful safe flight - aside from maintenance stuff.

During the flight it could provide far more useful and timely meteo info that would make flights more comfortable and safer, and in an accident situation where all is not lost, could assist in speeding up recovery and rescue before it is too late.

It could also be part of the infrastructure to modernize the air navigation system to enable aircraft to fly more direct routes from origin to destination. Most flights go way off course from the straight line, consuming thousands of tonnes of additional fuel every year. Dumb. Antiquated..... While a new navigation infrastructure will require a lot more infrastructure than my simple proposal in this thread, there needs to be a plan.

"A European Commission report last year said the existing route from Lyon to Frankfurt was 40.7 percent longer than necessary, while the route from Amsterdam to Milan in Italy was 23 percent too long."

http://www.oneclimate.net/2009/03/26/eu-moves-to-straighten-air-routes-cut-fuel-burn/

And I won't delve into the EFB - http://en.wikipedia.org/wiki/Electronic_flight_bag

The industry is relying on some antiquated "technologies".

BrianD

probe wrote: »

Please take your head out of the sand!

With respect I am not the one who is advocating a global network for the transmission of useless flight data from thousands of aircraft worldwide because humans have a natural anxiety about airline travel. The facts speak for themselves. We are talking about less than 1000 deaths worldwide and less than 50 aircraft. I fail to see how any of these lives could have been saved if the type of information stored on both black boxes was transmitted in real time. In fact nobody advocating this approach has stated how it could help. The only benefit that I can see is that it saves looking for the black boxes themselves.

I suspect you have problems about real-time spying on what flight crew might do.... no thinking person wants that, aside from you know who.... Which is why I have suggested that the Captain gets access to self-service deletion and proof of deletion after the flight - of any data that might identify the crew involved or the specific flight in question.

I am not in the least bit concerned about the privacy of flight crew conversations and it doesn't even enter my arguement. Though it has been raised by another poster in this thread and it is another layer of redundent and useless data that would need to transmitted around this global network.

In reality, if you were to have this self deletion approach, you talking about a huge data management system that would just become ridiculous to manage. Does ever pilot get a log in? Do you have an office full of people who have to maintain this system checking that stuff is deleted.

In any case, the flight crew should be the last people allowed to delete flight information.

Most of the data is irrelevant after a successful safe flight - aside from maintenance stuff.

Agreed so why not stick with the current situation where it is collected and stored securely on the aircraft and can be downloaded post flight

During the flight it could provide far more useful and timely meteo info that would make flights more comfortable and safer, and in an accident situation where all is not lost, could assist in speeding up recovery and rescue before it is too late.

As far as I am aware there are no real issues with the quality and availability of weather information. Pilots have access to this information pre-flight and aircraft have their own weather radar to deal with the weather situation as it evolves. A large percentage of flights operate within areas that are covered by ground radar and would be able to be advised of any weather issues on route. It is my understanding is that the current communication system does allow flight plans to be updated en route to take account of weather conditions.

It could also be part of the infrastructure to modernize the air navigation system to enable aircraft to fly more direct routes from origin to destination. Most flights go way off course from the straight line, consuming thousands of tonnes of additional fuel every year. Dumb. Antiquated..... While a new navigation infrastructure will require a lot more infrastructure than my simple proposal in this thread, there needs to be a plan.

Says who? An aircraft has sophisticated navigation and on board computers. If they are flying that much of course then we have other serious issues on our hands!

"A European Commission report last year said the existing route from Lyon to Frankfurt was 40.7 percent longer than necessary, while the route from Amsterdam to Milan in Italy was 23 percent too long."

http://www.oneclimate.net/2009/03/26/eu-moves-to-straighten-air-routes-cut-fuel-burn/

And I won't delve into the EFB - http://en.wikipedia.org/wiki/Electronic_flight_bag

The industry is relying on some antiquated "technologies".

I have read this article. There is nothing to state that " real time flight data transmission" in both directions would improve this situation. It seems that everyone knows that the route is too long without this technology and it seems to me that flights are religiously following an established route that may have been set for reasons that are now out of date.

What I would do is take the notional budget for the global real time flight data transmission system and plough it into radar cover for areas that currently has none e.g. the south atlantic where the air france aircraft was lost. It seems to me that if radar coverage was available in that region that answers would be more forthcoming. The track, speed and altitude could have been recorded along with any changes. At the very least, the point at where the search and rescue/recovery should start would be more definitive.

Biro

BrianD wrote: »

With respect I am not the one who is advocating a global network for the transmission of useless flight data from thousands of aircraft worldwide because humans have a natural anxiety about airline travel. The facts speak for themselves. We are talking about less than 1000 deaths worldwide and less than 50 aircraft. I fail to see how any of these lives could have been saved if the type of information stored on both black boxes was transmitted in real time. In fact nobody advocating this approach has stated how it could help. The only benefit that I can see is that it saves looking for the black boxes themselves.



I am not in the least bit concerned about the privacy of flight crew conversations and it doesn't even enter my arguement. Though it has been raised by another poster in this thread and it is another layer of redundent and useless data that would need to transmitted around this global network.

In reality, if you were to have this self deletion approach, you talking about a huge data management system that would just become ridiculous to manage. Does ever pilot get a log in? Do you have an office full of people who have to maintain this system checking that stuff is deleted.

In any case, the flight crew should be the last people allowed to delete flight information.


Agreed so why not stick with the current situation where it is collected and stored securely on the aircraft and can be downloaded post flight



As far as I am aware there are no real issues with the quality and availability of weather information. Pilots have access to this information pre-flight and aircraft have their own weather radar to deal with the weather situation as it evolves. A large percentage of flights operate within areas that are covered by ground radar and would be able to be advised of any weather issues on route. It is my understanding is that the current communication system does allow flight plans to be updated en route to take account of weather conditions.


Says who? An aircraft has sophisticated navigation and on board computers. If they are flying that much of course then we have other serious issues on our hands!



I have read this article. There is nothing to state that " real time flight data transmission" in both directions would improve this situation. It seems that everyone knows that the route is too long without this technology and it seems to me that flights are religiously following an established route that may have been set for reasons that are now out of date.

What I would do is take the notional budget for the global real time flight data transmission system and plough it into radar cover for areas that currently has none e.g. the south atlantic where the air france aircraft was lost. It seems to me that if radar coverage was available in that region that answers would be more forthcoming. The track, speed and altitude could have been recorded along with any changes. At the very least, the point at where the search and rescue/recovery should start would be more definitive.

With due respect, you seem obsessed with "useless data clogging the airwaves".
85% of the internet is useless data. Possibly more. This data would only take up a tiny fraction of internet traffic, and it might be useless 99% of the time, but that 1% would be so invaluable as to make it all worth while.
If they don't find the FDR's from this flight (looking likely), then our theoretical system would prove invaluable. You're stating it's useless and stupid.

BrianD

Biro wrote: »

With due respect, you seem obsessed with "useless data clogging the airwaves".
85% of the internet is useless data. Possibly more. This data would only take up a tiny fraction of internet traffic, and it might be useless 99% of the time, but that 1% would be so invaluable as to make it all worth while.
If they don't find the FDR's from this flight (looking likely), then our theoretical system would prove invaluable. You're stating it's useless and stupid.

Where's the value to the flying public? If they never find the black boxes or find the exact cause of the loss this flight it won't make the slightest bit of difference to pasenger safety. I wouldn't be surprised if the same type of aircraft flew the same route today.

Why? Aircraft manufacturers have spent vast amounts of money on R&D designing safe and durable aircraft. The stats support this. In the event of a disaster, the black boxes are recovered in 99% of time allowing investigators to establish what happened.

History also shows that nearly all (non-criminal) air craft losses and fatalities are caused by a unique chain of events that is often impossible repeat. Not only that, each of the events would not be sufficient to destroy the flight - a plane can fly safely with these impaired pitot tubes, other planes flew through the turbulence safely ... etc.

So I am stating that the use of real time data transmission is a pretty useless idea and does nothing for anybody. It even defies good management. Why would we collect, store and manage this data in real time when it can be more efficiently collected retrospectively.

To me, it all comes down to a general fear or anxiety of flying that humans have. Why else would we invest in some much for so little gain?

Biro

BrianD wrote: »

Where's the value to the flying public? If they never find the black boxes or find the exact cause of the loss this flight it won't make the slightest bit of difference to pasenger safety. I wouldn't be surprised if the same type of aircraft flew the same route today.

Why? Aircraft manufacturers have spent vast amounts of money on R&D designing safe and durable aircraft. The stats support this. In the event of a disaster, the black boxes are recovered in 99% of time allowing investigators to establish what happened.

History also shows that nearly all (non-criminal) air craft losses and fatalities are caused by a unique chain of events that is often impossible repeat. Not only that, each of the events would not be sufficient to destroy the flight - a plane can fly safely with these impaired pitot tubes, other planes flew through the turbulence safely ... etc.

So I am stating that the use of real time data transmission is a pretty useless idea and does nothing for anybody. It even defies good management. Why would we collect, store and manage this data in real time when it can be more efficiently collected retrospectively.

To me, it all comes down to a general fear or anxiety of flying that humans have. Why else would we invest in some much for so little gain?

I don't consider 300 lives saved as "little gain". If it were to happen only once more in the future of the A330 then it's once more too often. Every accident teaches lessons about how to avoid a similar one in the future.
I agree with the chain of events thing, but in that chain if the pilots did one thing differently they may have prevented the crash, even though you can't blame them. The only way to learn what they could have done differently is to recover all the information they can about the relevant events.
For example how did the Pilots react to the info they were given by the displays. A simple change in procedure for all A330 operations in the future when situation A, B and C happen simultaneously, can potentially save 300 lives.

BrianD

Biro wrote: »

I don't consider 300 lives saved as "little gain". If it were to happen only once more in the future of the A330 then it's once more too often. Every accident teaches lessons about how to avoid a similar one in the future.
I agree with the chain of events thing, but in that chain if the pilots did one thing differently they may have prevented the crash, even though you can't blame them. The only way to learn what they could have done differently is to recover all the information they can about the relevant events.
For example how did the Pilots react to the info they were given by the displays. A simple change in procedure for all A330 operations in the future when situation A, B and C happen simultaneously, can potentially save 300 lives.

Unfortunately, those lives were lost irrespective of the current system or the suggested real time system. It's an unfortunate fact of life that an average of 500 lives will be lost each year. The fact of the matter, is that an analysis of this accident and the 30-50 worldwide that happen each year won't contribute much to air line safety given that tens of thousands of flights take place every year safely in all sorts of conditions. It's such a small "loss" in the grand scheme of things. Let's face it there are many other things that this type of technology (or the cost of it) could be applied to that would save more lives. Each of these accidents is so unique an occurence that I doubt if they can be repeated and it seems to me that the massive investments in R&D is what drives airline and passenger safety.

Already, there is a stong theory as to what did happen with the available evidence. Airlines can start training for a situations with faulty pitot tubes and turbulent weather.

Biro

BrianD wrote: »

Unfortunately, those lives were lost irrespective of the current system or the suggested real time system. It's an unfortunate fact of life that an average of 500 lives will be lost each year. The fact of the matter, is that an analysis of this accident and the 30-50 worldwide that happen each year won't contribute much to air line safety given that tens of thousands of flights take place every year safely in all sorts of conditions. It's such a small "loss" in the grand scheme of things. Let's face it there are many other things that this type of technology (or the cost of it) could be applied to that would save more lives. Each of these accidents is so unique an occurence that I doubt if they can be repeated and it seems to me that the massive investments in R&D is what drives airline and passenger safety.

Already, there is a stong theory as to what did happen with the available evidence. Airlines can start training for a situations with faulty pitot tubes and turbulent weather.

There's no point in clogging up this thread any more with this debate, it's going around in circles. I disagree with you, you disagree with me. I just think that there's already too many useless applications of this technology for useless info. This would be useful, even if only once every 5 years that would make it 100% more useful than the majority of information traffic currently.