AAIB investigation into serious fatigue-related incident near Birmingham, UK, 2004

[shoey1976]

Dear all

Firstly the usual introduction: my name is Ian Shoesmith and I am a journalist working for BBC News in the UK.

For the past three months or so I've been investigating pilot fatigue: its causes as well as its consequences.

As part of my research I would very much like to speak to anybody directly involved in the following incident, as reported by the AAIB in its April 2006 bulletin:

Avro 146-RJ100, registered G-CFAF (http://www.aaib.dft.gov.uk/publications/bulletins/april_2006/avro_146_rj100__g_cfaf.cfm)

Plane was an Avro 146-RJ100, registered G-CFAF.

Not sure who operated this flight at the time, or whether that airline is still operating under its current name.

Were you either the captain or FO on that flight? Do you know who was? If you could let me know, in confidence, I'd be very grateful. Either email me [email protected] or ring me on +44 (0)7769 977665 or pm me.

Best wishes
Ian

[flybmi]

http://www.aaib.dft.gov.uk/publications/bulletins/april_2006/avro_146_rj100__g_cfaf.cfm ;)

Looks like it was operated by BACX at the time.

[rhythm method]

Ian, I would doubt it was a fatigue related issue, as the RJ100 fleet were not particularly overworked, especially in comparison to low-cost carriers. Sorry can't be more informative as I no longer work for them.

[compressor stall]

From the analysis....

The co-pilot commented that he had had a particularly
busy roster during the month leading up to the incident,

[shoey1976]

Just to quote the key bits, in my opinion, from that AAIB report

Analysis
Investigation of this incident has focussed on trying to determine why the co-pilot mistakenly selected the flap lever when attempting to raise the gear. In doing so several significant areas were identified which may have been contributory.

a Fatigue
The co-pilot commented that he had had a particularlybusy roster during the month leading up to the incident, flying 87 hours 50 minutes in the previous 28 days. The legal absolute maximum number of flying hours quoted in CAP 371 over the same period is 100 hours. In the same period he had been rostered nine days off, two of which were single days off. In all but one case these days off were preceded by duty periods finishing between 2035 hrs and 2110 hrs (local) and all were followed by duty periods starting between 0600 hrs and 0925 hrs (local). He stated that the week leading up to the incident had been particularly busy with six duty periods, half of which involved early starts.
Whilst his roster conformed to the required legal limitations the co-pilot believed that it had left him generally fatigued. This, in his opinion, was the major cause of the incident and he stated that he had failed to recognise in himself "a level of fatigue that would facilitate such an error".

and yet, in the Conclusions and Safety Recommendations section, there is no mention of fatigue.
I'm intrigued to know why not - is it just because the crew were within legal FTLs, and it's not the AAIB's role to criticise CAP371?

[dontdoit]

And therein lies the problem. Airlines increasingly see the CAP371 figures (900 hours, 100 hours etc) not as limits but planning targets. Swiss cheese, anyone?

[bermondseya]

and yet, in the Conclusions and Safety Recommendations section, there is no mention of fatigue.
I'm intrigued to know why notOh Yes There Is!

You simply haven't read the report properly.

[shoey1976]

Quoting: Bermondseya: oh yes there is...

Ian Shoesmith writes: You're right: apologies for that. The Conclusion indeed states: "The reasons for this are likely to have been the result of a combination of fatigue, distraction and inappropriate task prioritisation."

but, and this is what I believe is important here, there were no (explicit) recommendations made regarding fatigue.

Hope this clarifies things a little.

Ian

[Magplug]

Ian,
I don't recall fatigue EVER being cited as a causal factor, it is only ever cited as a contributory factor. The fact that two relatively bright guys end up doing something rather dumb on a dark night after 13 hours of duty is popularly called human error. The fact that those guys were acting on only 25% of their normal capacity due fatigue was only a contributory cause.
We are not yet sufficiently enlightened to appreciate the folly of pilots working these kind of durations. For the AAIB to make a recommendation for change to CAP371 would take an accident of such size and controversy where fatigue played a primary role.

We have seen prosecutions recently of drivers working professionally in unregulated situations where they were found culpable after accidents. The law has already decided that being tired amounts to recklessness and has delivered several judgements accordingly. Perhaps you might ponder the wisdom of flying an aircraft of infinitely greater complexity towards the end of a duty period of 12/13/14 hours, after all the most demanding phase is always at the end of a flight!

CAP 371 was designed as a sensible maximum in the age when rostering officers used pencil and paper. Today trip sequences are constructed by computers to achieve the greatest possible productivity within the regulatory confines of the maximum legal day. Working to the absolute limit has become the industry norm.

Do not make the mistake of thinking our regulatory body, the CAA, is the guardian of air safety in the United Kingdom. De-Facto... the CAA exists to protect the commercial well-being of UK airlines. For the CAA to legislate in a manner that might place a UK airline at a commercial disadvantage to a European competitor would be utterly unheard of.

[CaptainFillosan]

And therein lies the problem. Airlines increasingly see the CAP371 figures (900 hours, 100 hours etc) not as limits but planning targets. Swiss cheese, anyone?

I will treat your naivety gently. But.........one particular airline, in my certain knowledge, have been doing that for years. It was not unknown that some guys had 3 months free after flying 900hours in nine months..........then flew the haj! A big money earner that.

[ZeBedie]

Shoey, are you familiar with the Air Algerie 737 crash near Coventry, about 10 years ago? I think the AAIB report suggested fatigue as a causal factor. Things might have been different if it hadn't just missed the housing estate and/or it had been carrying passengers.

[shoey1976]

I vaguely remember that, being from Coventry myself. I'll certainly have a look into it.
To the best of my knowledge, the above incident was the only one involving a CAA-regulated operator in the last five years - I spent most of an afternoon going through AAIB bulletins since April 2002.
Again, if anybody knows anything to the contrary I'm all ears.
Scientific consensus suggests fatigue to have been a factor in around 15-20 per cent of incidents and accidents -- way, way higher than the figure suggested by those AAIB stats.
Why the discrepancy??? Have the scientists got it wrong, or are there reasons why fatigue is (a) not always reported or (b) not always looked for?
You may think I'm being very naive in asking that question, but again I stress I'm a journalist, not a pilot, and that's why I'm asking.

[BOAC]

Something you may wish to focus on, Ian, is the distinction between 'very tired' and 'fatigued'. The CAA say that to feel 'very tired' at the end of a duty close to maximum is 'normal'. To fly when 'fatigued' is contrary to the Air Navigation Order. You will note that the F/O in your highlighted report did not realise he was 'fatigued'. Define the difference?

I know from my time in BA shorthaul at Gatwick that 3 or 4 days of getting up at 0400'ish (UK clock time) somewhere in Europe (packing the suitcase) to fly 2 or 4 sectors to a different hotel somewhere in Europe (unpacking the suitcase) was extremely 'tiring' by the end of the duty. Whether or not I would have been 'fatigued' at the end of day 3 or 4 is a difficult call. Now build in the usual 'life-factors' - children, illness etc, and start the tour with a bad night's sleep. Now call the company and explain? They would be delighted to hear that you do not consider yourself fit to operate. OK - you reckon you are just 'very tired' (keep going!). Now have a major emergency on landing on the last trip.

A lot of the time the job is 'easy'. A 'nice' roster day, good rest before, no problems........wonderful. Now throw in the oddball. Judging by other threads on PPrune about working patterns elsewhere in aviation we are lucky in the UK. That's a thought?

[Jaun Huw Nose]

Sadly the CAP 371 rules assume you can sleep to order and have no life...
so if you have to work at say 1600 local you can do 12.25 hrs duty, ie finish at 04.15 local, but the chances are you woke up at 7 or 8am when the rest of the family did and then did other stuff until going to work. Assuming after ending that duty you get home at 0600 you ethieir have to go to bed and be woken up in a hour or so or stay awake because you are then going to work at 2100 that night... do this a few times with local time changes and you get VERY tired!

[GK430]

Shoey, if it is fatigue that you are looking at as a causal factor, did you read the final report on the accident of N90AG at Birmingham on Jan 4th 2002. Whilst not directly fatigue related, the whole issue of time zones, medicinal sleep aids and other factors led to a poor decision.

On another note, how do middle east based crews where hubs (i.e DXB) come alive with flights departing around 02:30 cope on a regular basis? As someone else has said, assuming a good quality sleep pattern prior to getting up to fly at those times is virtually impossible!

[alf5071h]

One of the problems of incident investigation and perhaps a particular problem of investigative journalism is hindsight bias (both for the investigator and participant). Thus, if you are looking for a link with fatigue in an incident you may easily find it. Often in this type of incident, the individual does not know why they did something, and subsequently (unconsciously) seek a closing/explanatory reason for their actions.

The AAIB has provided an excellent and well balanced report into an incident involving human error – not fatigue. Human error can be influenced by fatigue, but also and more likely, by many other factors. Incidents rarely have an identifiable cause; they consist of a series of contributory factors where all are necessary, but on their own would not lead to an incident, i.e. fatigue may contribute to an accident, but on its own does not cause an accident.

The AAIB report considered the relevant human factors, which are often subject to supposition or hindsight bias, in a manner which enables the industry to learn from this incident. The report considers the organizational aspects and wider issues that might have influenced the crew. It is also pleasing to see a report that addresses wider safety issues such as RNAV, which could contribute to serious events.
The first paragraph of the conclusion captures the critical issues – human error (no blame), and the cognitive contributions originating from the situation and organizational issues. (I would have liked to see comment on the pilots experience; did the action originate from a previous type – reverting to basics in the stress of the situation?)

Fatigue is a problem for the industry, but it has to be put into perspective.
It is difficult for an individual to identify his/her fatigue, more often we see the resultant first. Whereas tiredness usually can be detected and appropriate defensive measures taken ([b]BOAC[/] above).

I wish Ian good luck with his investigation, but remain skeptical that the effects of fatigue in aviation can be adequate portrayed without considering the wider aspects of human error together with the many other contributions to accident and incidents, and of course the problems of hindsight bias.
My concerns are further heightened as to the manner in which pilot fatigue might be presented by the use of an emotional (and inaccurate) thread title. This plays on the potential for error; we are easily distracted by the big, bright, or loud item in a situation (the title) – a human weakness of inadequate control of attention and failure to evaluate a more realistic view of the event, but that’s all part of human error.

[Yarpy]

That's all very nice. Analyse the whole situation and reduce 'fatigue' to one small contsituant part of a more complex problem.

Is it not possible to take a common sense view of things and say that working people flat out is likely to cause mistakes leading to accidents?

[flaps40]

ba advised that the word fatigue is most probably inappropriate in this case, as in most cases where it is used. the pilot may have been tired in which case a few days off would have fixed the tiredness. fatique is a longer-trem condition which cannot be solved by a quick fix of any sort.