Worlds Safest Airlines for 2012

[amaroo]

I would imagine that QF32 is recent and serious enough that it has affected the Qantas rating, but there must be some other incidents that have happened to contribute to that low rank.

Maybe the A330 issues? Cookies must be enabled. | The Australian

 

[mannej]

Maybe the A330 issues? Cookies must be enabled. | The Australian

And the 747 over Manila - however, they all ended with the same outcome did they not?

 

[markis10]

And the 747 over Manila - however, they all ended with the same outcome did they not?

Not quite the same when it comes to injuries, and lets not forget the Bangkok golfer, and the recent multiple 717 hard landings.

 

[amaroo]

And the 747 over Manila - however, they all ended with the same outcome did they not?

If they are being marked down for these incidents IMO it's showing a skewed result.

Rolls Royce, Airbus & Oxygen tank - AFAIK these problems are not of QF making. However, at the end of the day the colours were QF & I suppose they have to deal with it.

 

[NM]

If they are being marked down for these incidents IMO it's showing a skewed result.

Rolls Royce, Airbus & Oxygen tank - AFAIK these problems are not of QF making. However, at the end of the day the colours were QF & I suppose they have to deal with it.

But what the raw stats don't show is how they dealt with it. What changes were implemented to maintenance procedures/schedules, to pilot/crew training, to operating procedures? In many cases, the end result is a safer airline as a result of the changes implemented, not a less safe airline as a pure historical review of the statistics would imply. For example, it is my understanding that the BKK golfing expedition resulted in some changes to standard operating procedures and crew training aimed at minimising the probability of it happening again. Without the changes being implemented, safety is compomised. But following the changes that were implemented as a direct result of the incident, the probability of it happening again is reduced. Logically this should mean an improved future safety rating, not a reduced one.

Without consideration for how each airline responds to incidents (their own and those experienced by others), any historical safety statistic review is just that - historic. It does not imply likelihood of future issues unless no changes have been implemented as a result of past failings - and I doubt there are many airlines that a guilty of that and if there are any, those are the ones I don't want to fly (ok, a few may come to mind). In my opinion, airlines that bury their heads in the sand and refuse to acknowledge any responsibility for mistakes are the ones that are most likely to have repeat issues. But its all statistics and in the case of the referenced study, I believe there are several factors that have a significant impact on the statistical validity of the results, especially for predicting future results, that appear to have been ignored. Then again, its very hard to find any details about the mechanism used to calculate the safety factor rating in that study.

 

[anat0l]

But what the raw stats don't show is how they dealt with it. What changes were implemented to maintenance procedures/schedules, to pilot/crew training, to operating procedures? In many cases, the end result is a safer airline as a result of the changes implemented, not a less safe airline as a pure historical review of the statistics would imply. For example, it is my understanding that the BKK golfing expedition resulted in some changes to standard operating procedures and crew training aimed at minimising the probability of it happening again. Without the changes being implemented, safety is compomised. But following the changes that were implemented as a direct result of the incident, the probability of it happening again is reduced. Logically this should mean an improved future safety rating, not a reduced one.

Without consideration for how each airline responds to incidents (their own and those experienced by others), any historical safety statistic review is just that - historic. It does not imply likelihood of future issues unless no changes have been implemented as a result of past failings - and I doubt there are many airlines that a guilty of that and if there are any, those are the ones I don't want to fly (ok, a few may come to mind). In my opinion, airlines that bury their heads in the sand and refuse to acknowledge any responsibility for mistakes are the ones that are most likely to have repeat issues. But its all statistics and in the case of the referenced study, I believe there are several factors that have a significant impact on the statistical validity of the results, especially for predicting future results, that appear to have been ignored. Then again, its very hard to find any details about the mechanism used to calculate the safety factor rating in that study.

With factors like this, I'd expect that this kind of information is not taken into account at all.

It is much the same as many safety ratings for various industries out there. Take mining safety for instance. There are many metrics and measures - e.g. annual fatalities, lost work time, etc. - that can be used to rank and judge these companies, but it doesn't really measure or examine if and/or how companies responded to such safety incidences, and thus set a vector for the overall safety position of the company.

Not to say that the rating is complete rubbish; it all depends on how you interpret the metric. The academic quoted in markis10's referenced article probably is the only one who has come close to examining the context around the safety metric. Everyone else has just taken it at prima facie value without much consideration, from what I can see.

Metrics are metrics - on their own they are frequently useless. They must be interpreted and understood to actually be of any use.

That said, certainly less incidents in history is better, so airlines clearly without any incidents have some right to boast accordingly. Notwithstanding the attribution (or misattribution) of primary responsibility of an incident, the increase or decrease in the safety metric would also serve some useful preliminary insight (but there's been no historical JACDEC results presented with the current results to make those kinds of judgements).

 

[dajop]

The other thing this study does not seem to consider is the outcome following such incidents. Has the airline (or the equipment manufacturer as the case may be) implemented changes to minimise the risk of a repeat incident as a result of learning the root cause of the incident. If so, then the end result is a safer airline, not a less safe one. Historic statistics does not necessarily predict future probability if changes have been implemented to minimise the probability of a repeat.

Totally agree. A oft quoted case study is Korean Air. If you look at the record during the 90's it is truly scary stuff. The 00's are much less scary, after significant efforts to change the culture at the airline.

 

[NM]

That said, certainly less incidents in history is better, so airlines clearly without any incidents have some right to boast accordingly. Notwithstanding the attribution (or misattribution) of primary responsibility of an incident, the increase or decrease in the safety metric would also serve some useful preliminary insight (but there's been no historical JACDEC results presented with the current results to make those kinds of judgements).

indeed a lack of incidents means either they have been very lucky or they have a culture and procedures in place where mistakes are found and corrected before they result in an incident.

 

[RooFlyer]

Has anyone managed to see or get hold of the actual report?

From what I gather from its internet site, the 'report' is in some German fly magazine. I'm going to look for it when I'm there in a couple of weeks, but any links in the meantime ?(to the report, not the summary table).

Cheers

 

[jb747]

indeed a lack of incidents means either they have been very lucky or they have a culture and procedures in place where mistakes are found and corrected before they result in an incident.

Or that they don't report things...

 

[JohnK]

Does anyone take these seriously?

I mean are you going to drop all your travel on QF and travel NZ and VA? Highly unlikely....

 

[dajop]

Does anyone take these seriously?

I mean are you going to drop all your travel on QF and travel NZ and VA? Highly unlikely....

Some people would. Probably not QF vs NZ/VA - same order of magnitude, and I don't think you can attach any sort of significance or have the granularity for carriers rated roughly rated same order of magnitude.

Many would think twice about taking carriers that are an order of magnitude or two different. (eg China Airlines, TAM & Air India vs QF, VS, NZ, BA etc). Of course, as reminded by several above past performance (good or bad) does not necessarily correlate with future performance.

 

[anat0l]

I mean are you going to drop all your travel on QF and travel NZ and VA? Highly unlikely....

Well... more likely than you think, but I doubt the substantial reason would be safety concerns...

 

[TheEmu]

Or that they don't report things...

Which is probably the scariest scenario of all…thanks jb! :shock: