NTSB wants to monitor flight deck conversations

[MikeD]

If used like a FOQA program, the information would be deidentified and only the union "gatekeepers" would be privy to the flight specific identification information. Problems would be forwarded to the union pro stan committee, not to management. Management gets only deidentified stuff to use in developing training programs, identify trends, etc.

Which if program integrity can be assured, may not be a bad idea to at least test out, if even. In the AF, its already known how/when that info is used by the bosses, but then again that's a whole different operating environment (from a operating rules, employee rights, and management/employee perspective) from the airlines too.

I say that last sentence above because I'm not in the airlines, so don't have the "first hand" perspective of the overall opinion on the matter.

 

[TUCKnTRUCK]

You totally miss the point. Let's say that a water bottle rolls under the rudder pedal and you try to push it and it doesn't move. You run off the side of the runway, and as you bounce along, the water bottle dislodges and heads aft. Looks like pilot error according to the DFDR. You say you were pushing on that rudder? "Yah, right, sure you were, we found nothing wrong with any of the flight controls."

The video should show you actually trying to push, and maybe even that water bottle. Same could be said of hitting a switch but it not taking, or bouncing back, etc. I could go on and on.

You mean like the cal 73 in Denver? Pretty sure they figured out what went wrong there.

I'll stick by my orriginal post. Pilots would not trust it enogh, and for the rare instance that it could catch an attempted button press, you would face a tenfold increase in crm related issues as crews were afraid of being caught on camera. There is a reason why they were not installed with cvr's as initially proposed.

If you have been insafety for 20 years then you know, that for a safety program to work, the pilots must use, and Trust it. I think the dfdr in our plane pulls 250 data streams from the cockpit alone. I don't really see video adding much.

 

[MikeD]

You mean like the cal 73 in Denver? Pretty sure they figured out what went wrong there.

I'll stick by my orriginal post. Pilots would not trust it enogh, and for the rare instance that it could catch an attempted button press, you would face a tenfold increase in crm related issues as crews were afraid of being caught on camera. There is a reason why they were not installed with cvr's as initially proposed.

If you have been insafety for 20 years then you know, that for a safety program to work, the pilots must use, and Trust it. I think the dfdr in our plane pulls 250 data streams from the cockpit alone. I don't really see video adding much.

Thats why I said in my above post, any program like this would have to have some severe degree of "program integrity", which would have to be maintained. Question is, is that possible?

 

[seagull]

Which if program integrity can be assured, may not be a bad idea to at least test out, if even. In the AF, its already known how/when that info is used by the bosses, but then again that's a whole different operating environment (from a operating rules, employee rights, and management/employee perspective) from the airlines too.

I say that last sentence above because I'm not in the airlines, so don't have the "first hand" perspective of the overall opinion on the matter.

The current FOQA programs have already demonstrated this.

 

[seagull]

You mean like the cal 73 in Denver? Pretty sure they figured out what went wrong there.

I'll stick by my orriginal post. Pilots would not trust it enogh, and for the rare instance that it could catch an attempted button press, you would face a tenfold increase in crm related issues as crews were afraid of being caught on camera. There is a reason why they were not installed with cvr's as initially proposed.

If you have been insafety for 20 years then you know, that for a safety program to work, the pilots must use, and Trust it. I think the dfdr in our plane pulls 250 data streams from the cockpit alone. I don't really see video adding much.

Pilots said that about the CVR too. The bottom line is that you don't think about it once installed. If you are so paranoid that you are, you are probably doing something that you shouldn't be doing anyway, honestly. I do my best to fly standard and by the book. I have no worries about anyone watching or recording me, just as I don't get worried about an FAA line check. In fact, I have asked them to ride up front when they were just onboard going to check some ground thing, just for extra eyes to back me up. There is truly no reason to be paranoid unless you are not following procedures. If you are, the recording can only PREVENT being accused of something, not the reverse!

Have to add that I can think of several accident investigations I have done that video would have really helped, particularly with a view out the front window!

 

[MikeD]

Pilots said that about the CVR too. The bottom line is that you don't think about it once installed. If you are so paranoid that you are, you are probably doing something that you shouldn't be doing anyway, honestly. I do my best to fly standard and by the book. I have no worries about anyone watching or recording me, just as I don't get worried about an FAA line check. In fact, I have asked them to ride up front when they were just onboard going to check some ground thing, just for extra eyes to back me up. There is truly no reason to be paranoid unless you are not following procedures. If you are, the recording can only PREVENT being accused of something, not the reverse!

Have to add that I can think of several accident investigations I have done that video would have really helped, particularly with a view out the front window!

Military airlift aircraft have the standard CVR/FDR, but with the fighter HUDs, the added advantage is being able to see whats going on out front of the aircraft, in the video.

 

[Seggy]

You need to get more involved in investigations and analysis. You would see that those are FAR from "all the data" needed.

No sir I have been intimately involved with an investigation and analysis of an accident and the implementation of these programs at an airline. I've also been on the tip of the spear with the concept of monitoring of the CVRs.

I'll say it once again...

With properly run LOSA observations, a fully functioning ASAP Program, and a robust FOQA program it can produce all the data you need to enhance safety.

Plus the technical ability to use the CVRs would be extremely difficult.

Also how do you account for this use of the CVR when ICAO Annex 13 states it is only to be used for an accident investigation tool?

 

[seagull]

No sir I have been intimately involved with an investigation and analysis of an accident and the implementation of these programs at an airline. I've also been on the tip of the spear with the concept of monitoring of the CVRs.

I'll say it once again...

With properly run LOSA observations, a fully functioning ASAP Program, and a robust FOQA program it can produce all the data you need to enhance safety.

Plus the technical ability to use the CVRs would be extremely difficult.

Also how do you account for this use of the CVR when ICAO Annex 13 states it is only to be used for an accident investigation tool?

That's nice, but you are still incorrect. If you have been involved, you have to know that a LOSA cannot, and never can, capture many things. People do NOT ignore the fact that the observer is present. The ASAP and FOQA are good too, but there are MANY times that a CVR would be a great supplement when looking at FOQA data. Not all the time, but when things pop out of the other data, it would provide an additional layer.

We worked hard on how to fix some of these issues at CAST, but the paranoia is inane, and not justified, IMO.

 

[ChasenSFO]

Because we're so willing to give it away.

So true.

And boy does it tick off those of us who aren't so willing.

 

[TUCKnTRUCK]

Have to add that I can think of several accident investigations I have done that video would have really helped, particularly with a view out the front window!

video would have helped... That's fine. Were you able to come to a satisfactory conclusion without it? You have been working at a company where you can trust your safety reporting systems. Many of us do not have that luxury. Last year I sat through 3 pc's 4 line checks 3 fed rides, and 6 observations. We had people getting yanked off the line for the smallest infractions. With the introduction of the quick acess data download system and foqa, our safety department WANTED names. They wanted to go after the bad apples. So I'm sorry, but given the corporate culture at many airlines I simply can not agree with introducing a new means for pilot monitoring that doesn't have a set precedent. Especially one that doesn't really add much to the safety culture.

Processing video is not easy to do. You are stuck with a 2d perspective that can't capture everything. There are cases where video evidence may be misleading, or flat out innacurate. Sure you can watch yoke position, but you get no feedback from the aircraft itself. I can not think of a single situation where they were unable to resolve the issues in an incident without a video feed. Untill it is actually needed to breakdown event sequences, it's not needed.

 

[seagull]

video would have helped... That's fine. Were you able to come to a satisfactory conclusion without it? You have been working at a company where you can trust your safety reporting systems. Many of us do not have that luxury. Last year I sat through 3 pc's 4 line checks 3 fed rides, and 6 observations. We had people getting yanked off the line for the smallest infractions. With the introduction of the quick acess data download system and foqa, our safety department WANTED names. They wanted to go after the bad apples. So I'm sorry, but given the corporate culture at many airlines I simply can not agree with introducing a new means for pilot monitoring that doesn't have a set precedent. Especially one that doesn't really add much to the safety culture.

Processing video is not easy to do. You are stuck with a 2d perspective that can't capture everything. There are cases where video evidence may be misleading, or flat out innacurate. Sure you can watch yoke position, but you get no feedback from the aircraft itself. I can not think of a single situation where they were unable to resolve the issues in an incident without a video feed. Untill it is actually needed to breakdown event sequences, it's not needed.

And I can think of a number of accidents where we really do not know for sure what they saw, and nothing in the CVR or FDR showed any indication of what led to the loss of control or loss of situational awareness.

Our contract specifically states that no aircraft recording devices can be used for any discipline proceeding, period. That solves that issue. Why do other carriers not have that?

 

[Seggy]

The ASAP and FOQA are good too, but there are MANY times that a CVR would be a great supplement when looking at FOQA data. Not all the time, but when things pop out of the other data, it would provide an additional layer.

Then the FOQA gatekeeper calls the crew and asks what happens.

 

[Minuteman]

I think the dfdr in our plane pulls 250 data streams from the cockpit alone. I don't really see video adding much.

The Q400? It's new, good, and light years ahead of something like a BAe 146, but parameterized flight data isn't everything. For instance, it's difficult to figure out which pilot is even flying the aircraft from only FDR data (unless there's a reliable convention about using an autopilot or something similar). For another example: altitude overshoots. Often there is not enough information to determine the flight's intention (late radio call, intruding traffic, distraction).

I think if there were to be a large-scale review of audio or video recordings it would end up being a kind of fusion between flight data monitoring and ASAP submissions; the issues don't walk up to the program like ASAP (hypotheses must be investigated), but it would be still be a qualitative assessment. That requires a classification scheme and a taxonomy, like what is used by ASAP.

It would be very nice to have a bunker of data to add context to things after the fact, but that always has to be balanced against the potential of bad-faith use, legal subpoenas, or even just the manpower required to intake a bunch of recordings (which may have technological solution someday).

 

[Seggy]

That's nice, but you are still incorrect. If you have been involved, you have to know that a LOSA cannot, and never can, capture many things. People do NOT ignore the fact that the observer is present. The ASAP and FOQA are good too, but there are MANY times that a CVR would be a great supplement when looking at FOQA data. Not all the time, but when things pop out of the other data, it would provide an additional layer.

True LOSA is a great program but it will not catch everything. Just as if pilots start pulling Circuit Breakers hanging their jackets up over cameras.

Processing video is not easy to do. You are stuck with a 2d perspective that can't capture everything. There are cases where video evidence may be misleading, or flat out innacurate.

Yep also add the CVRs are not easy to process...

And I can think of a number of accidents where we really do not know for sure what they saw, and nothing in the CVR or FDR showed any indication of what led to the loss of control or loss of situational awareness.

In a recent accident even WITH video we really would not know what was going through the Captain's head on why he reacted the way he did. Video and monitoring of the CVRs can't obviously get into a pilots mind.

 

[TUCKnTRUCK]

And I can think of a number of accidents where we really do not know for sure what they saw, and nothing in the CVR or FDR showed any indication of what led to the loss of control or loss of situational awareness.

Our contract specifically states that no aircraft recording devices can be used for any discipline proceeding, period. That solves that issue. Why do other carriers not have that?

Do you really think a video looking at endless expanses of ocean, clouds, or my yoke is going to change that? What would video have shown from twa 800? What would we have seen from 3407? Nothing that would have affected the outcome of the investigations I'm betting. External cameras would show much more ( like the tip and belly cams on a gulfstream).

Trust me I'm all for safety, and I also do what I can to keep things standard. The Colgan pilots on here also know that when I see something I don't like, I speak up. I have no issues going up the chain of command untill something happens. I agree that video could be usefull, but in the current enviroment, and culture most 121 pilots face, I don't believe that watchdogging our cvr, or putting cameras in the fd are the answer.

Someday, when we have a unified safety directive across the board, run by pilots, then such a system could be trusted. Right now, I couldn't be sure that I wouldn't get a nastygram about letting my tie down after we close the door, or hanging a part of the release to block the sun so I can read my pfd. Do I want a call asking why I left my station every time I have to go take a piss? Honestly, I don't think that managment and safety should even be in the same building. For the safety dept. To work to it's fullest potential it has to be free from the oversite managment brings.

What would you contract say about a pilot getting an extra pc because they noticed a trend that they didn't like? Suddenly, what should have been a train to prof. Becomes a jeopardy ride, and I would bet failing that can lead to administrative action. That probably isn't the culture you are dealing with... We are though. If you were to allow cameras in, it would set a precedent that would give our company means to install them as well... But we don't have the protections in place that you do. Maybe this helps you understand shy I am against it. Do you think Mesa guys, tsa, go jets etc wNt it? How about the cal guys fighting for a contract? Airlines looking to furlough? Is it easier to wash them with failure to follow sop's?

 

[TUCKnTRUCK]

I'm going to add one more thought, police dash cams were supposed to be confidential, but look how they have ended up on tv, you tube, and even led to lawsuits. What would happen after a mishap if a lawyer got their hands on a video of the pilots eating lunch moments before hitting sever turbulence, resulting in a severly injured passenger? Odds are you would fond those pilots facing negligence charges and a civil lawsuit.

I really just feel that there is too little protection. What's to stop a prosecuted from going to a judge and getting a subpeana for the cockpit video? I admit I don't know the law in this area, but I'm pretty sure a inion contract would be trumped.

 

[TUCKnTRUCK]

The Q400? It's new, good, and light years ahead of something like a BAe 146, but parameterized flight data isn't everything. For instance, it's difficult to figure out which pilot is even flying the aircraft from only FDR data (unless there's a reliable convention about using an autopilot or something similar). For another example: altitude overshoots. Often there is not enough information to determine the flight's intention (late radio call, intruding traffic, distraction).

I think if there were to be a large-scale review of audio or video recordings it would end up being a kind of fusion between flight data monitoring and ASAP submissions; the issues don't walk up to the program like ASAP (hypotheses must be investigated), but it would be still be a qualitative assessment. That requires a classification scheme and a taxonomy, like what is used by ASAP.

It would be very nice to have a bunker of data to add context to things after the fact, but that always has to be balanced against the potential of bad-faith use, legal subpoenas, or even just the manpower required to intake a bunch of recordings (which may have technological solution someday).

We have a button to select which hsi/fgcp info to use. It is explicitly clear whom is flying, what the control forces are, positions etc. Beyound that with the ecaps protocol we have a standard transfer of controls that every crewmember I have met, uses. Every button, switch or otherwise is recorded. They know when a radio is keyed, an fa calls... Or heck, when you fart in your chair. They know what is being requested via controls vs. what is happening. Heck, they know the prop blade angle of each of the 12 blades, the position of the turbine bypass valve in the pacs, bleed air level selected, flight director modes, ap modes, how's flap selections. They know when alert cancel buttons are pressed, even at no alert on invalid times. They know every goes call out, every canceled gs call out, winds aloft, tad, ias, cas etc.

Our dfdr captures mountains of data because it is tied to the computers, and computers control everything in this plane.

 

[seagull]

All of these concerns were presented about the installation CVRs, and all pretty much proved unfounded. If the agreement REQUIRED that something in the contract prevented their use for anything outside of accident investigation, it would not be an issue.

Also, regarding our contract, the answer is that no, they would not be able to use anything from a CVR, FDR or a video, or any other aircraft recording device (there are dozens on any modern digital aircraft, incidentally) for any sort of action, to INCLUDE requiring extra training or anything else. Period.

 

[seagull]

We have a button to select which hsi/fgcp info to use. It is explicitly clear whom is flying, what the control forces are, positions etc. Beyound that with the ecaps protocol we have a standard transfer of controls that every crewmember I have met, uses. Every button, switch or otherwise is recorded. They know when a radio is keyed, an fa calls... Or heck, when you fart in your chair. They know what is being requested via controls vs. what is happening. Heck, they know the prop blade angle of each of the 12 blades, the position of the turbine bypass valve in the pacs, bleed air level selected, flight director modes, ap modes, how's flap selections. They know when alert cancel buttons are pressed, even at no alert on invalid times. They know every goes call out, every canceled gs call out, winds aloft, tad, ias, cas etc.

Our dfdr captures mountains of data because it is tied to the computers, and computers control everything in this plane.

Which also raises the issue of sampling rates on digital recordings. In many respects, the old analogue were better. Sad but true. Again, video would solve the issue.

 

[TUCKnTRUCK]

Again, you have those protections, we do not. Our cvr is not protected by contract, but rather the FAA/ntsb.

Cockpit video is not covered yet.