Buried in the pile of pretty incendiary documents on the Colgan Air Q400 crash that will start to become public today, I understand there is one that will cause particular angst in the pilot community. That's a suggestion from Colgan that cockpit voice recorder (CVR) extracts should become part of the flight operations quality assurance (FOQA) process.
If you're not familiar with FOQA, it's a programme in widespread
(though not widespread enough) use through which disidentified flight
data is downloaded on disk from every flight and monitored partly for
emerging patterns that may give an early indication of an underlying
problem and partly for gross errors or exceedances that need
immediately looking at. By and large the first part is
non-controversial whereas the second, which requires union agreement to
identify the crew involved and talk to them, can be.
Typically a crew in such an incident is taken aside by flt ops and union representatives to discuss what happened and why. More training may result, or other measures. But all in all, the whole idea of FOQA is to stop accidents happening. And on the whole it works - it's one of the most important safety developments of modern times.
It also depends crucially on having a legislative framework that protects all involved from the system being abused to punish or sue crews. And ensures that the data is not subject to freedom of information type requests.
What Colgan are suggesting however is a bit trickier. As I understand it, the idea is that the CVR extract covering the period of an incident - so that would be the cockpit conversation, ATC dialogue, and ambient noise - could be accessed in the event of an incident.
Expect union opposition. And you can see why. In order for this to work you'd need to download and retain for at least some period all the CVR data. Every single word spoken in the cockpit by every single crew. And then you'd need rules that said it could only ever be accessed in the event of an incident.
The thing about the flight data is that it is immensely useful in an aggregated form, even when a significant incident doesn't occur. And it's truly disidentified unless very deliberate steps are taken to identify it.
I don't think CVR data is like that. It's no use in aggregated form - in fact by definition it doesn't really exist in aggregated form - and it's instantly susceptible to identification.
On the other hand, would it be useful in the FOQA context. Well, yes, clearly it would.
No doubt Colgan will say that appropriate safeguards can be put in place, as they were with flight data, but I don't think they'll get agreement anytime soon.
And of course the timing is critical. It seems clear that Colgan has some justifying to do over the Buffalo crash and will need to show that it is taking action. I rather doubt that it would be making this suggestion in any other circumstances.
Remarkable the attention this accident is getting - from the BBC here and the Wall Street Journal here.
Comments???
Typically a crew in such an incident is taken aside by flt ops and union representatives to discuss what happened and why. More training may result, or other measures. But all in all, the whole idea of FOQA is to stop accidents happening. And on the whole it works - it's one of the most important safety developments of modern times.
It also depends crucially on having a legislative framework that protects all involved from the system being abused to punish or sue crews. And ensures that the data is not subject to freedom of information type requests.
What Colgan are suggesting however is a bit trickier. As I understand it, the idea is that the CVR extract covering the period of an incident - so that would be the cockpit conversation, ATC dialogue, and ambient noise - could be accessed in the event of an incident.
Expect union opposition. And you can see why. In order for this to work you'd need to download and retain for at least some period all the CVR data. Every single word spoken in the cockpit by every single crew. And then you'd need rules that said it could only ever be accessed in the event of an incident.
The thing about the flight data is that it is immensely useful in an aggregated form, even when a significant incident doesn't occur. And it's truly disidentified unless very deliberate steps are taken to identify it.
I don't think CVR data is like that. It's no use in aggregated form - in fact by definition it doesn't really exist in aggregated form - and it's instantly susceptible to identification.
On the other hand, would it be useful in the FOQA context. Well, yes, clearly it would.
No doubt Colgan will say that appropriate safeguards can be put in place, as they were with flight data, but I don't think they'll get agreement anytime soon.
And of course the timing is critical. It seems clear that Colgan has some justifying to do over the Buffalo crash and will need to show that it is taking action. I rather doubt that it would be making this suggestion in any other circumstances.
Remarkable the attention this accident is getting - from the BBC here and the Wall Street Journal here.
Comments???
Leave a comment
Want a user picture? Get a Gravatar!