Where are the safest skies

[Jeroen D]

I have read a lot about which airline is supposed to be the safest, least safest etc.
Im interested to see some comparison on a more regional level. Not on a airline basis, but more country or specific aviation regime basis

E.g. Are the skies over the USA safer then say over Europe.
I've been living in India for the last year and there are some commercial pilots here that claim India has the most stringent rules and regulations and a whole host of other things that make it safer. For instance it seems they enforce shorter working hours, longer resting periods and such.

So i'm interested to understand how the various aviation regimes differ, and whether that results in safer skies. Admittedly, a very broad and diverse topic that would easily fill a whole forum by itself, but Im sure somebody must have done some research on this already.

Jeroen

[Jeroen Hoppenbrouwers]

We can start summarizing these:
http://www.skybrary.aero/index.php/Aviation_Safety_Statistics

[Jeroen D]

Thanks, Interesting sites. Tonnes of data, difficult to find anything specific on India.

Jeroen

[Sese]

Here is another cool site.

India:
http://aviation-safety.net/database/country/country.php?id=VT

[Jeroen D]

Thanks, very usefull!

Jeroen

[IefCooreman]

Ever heard of the swiss cheese safety model?

http://csel.eng.ohio-state.edu/productions/pexis/images/swisscheese/swisscheeselatent.gif

Claiming an overall high safety level on the basis of one slice of cheese (if I may express myself like this) is pretty exagerated. It helps, but there is much more stuff influencing the "final safety level".

First thing that came to my mind was the "asian" culture. The further you go east (from my home), the more the captain-first officer work environment becomes more and more "captain dominant". Even with a good set of rules, fatigue is a pain in the ass to manage, and there are plenty of situations where it's better to have two equally accepted pilot minds up front, than one telling the other to shut up because... he had a bad night (which is far from the fatigue problem, it is only bad sleep...).

There are other issues in India pilots have to deal with: a lot less radar coverage, sometimes even loss of radio contact for short periods ('if you don't catch the next frequency now, try again in 20 miles"... and there you are, on your own at FL360), interesting weather situations (monsoon and the general "hazy" Asia), and the political situation with Bangladesh where you end up listening to 3 frequencies at the same time.

I suppose there is some "nationalism" involved. There is nothing wrong with that, all nationalities in aviation have their pride. But every region also has its problems. Europe has problems as well, but from all the areas I have flown in, Europe seems to be by far the most safe one. This is only my feeling. But there are a zillion options you have flying over Europe. All it takes is a mayday and you do what you want, and you are two backing eachother up. Fatigue? We know it's a problem. We know the difficult flights. When I end up tired after 6 hours inflight, I will tell my captain and vice versa. That allows you to deal with it and find a solution. Even before we takeoff we might have already had the discussion.

Any other regions you fly to, something changes in the equation. Adequate airports, ATC, ground facilities (hospitals,...), ATC,...

Some might say USA, but the general aviation overthere has almost ultimate freedom and I've already had that one encounter too many (I guess retired farmer loving class B airspace and taking his C172 for a cheat peak into our 777 flightdeck at 5000ft...   )

[Phil Bunch]

A thought about the country-specific fatality statistics in the aviation databases:  isn't the death rates from aviation too low to meaningfully compare most countries?  

With only a very small number of fatal events per year, it's statistically unreliable to compare countries IMO.  Now if they were many hundreds of crashes per country in each year, these comparisons would have statistical significance.  

An example:  country A has 2 crashes per year and country B had 1.  While it sounds like country A is twice as unsafe as country B, these are probably just random fluctuations of small numbers and there aren't enough events to reliably compare them.

Does this make sense?

[torrence]

Hi Phil,

The problem of statistics and their interpretation is the big difficulty in most of these discussions and I agree you probably can't define a 'safest country' this way.  As noted earlier, there are studies of safety by airline, which are tricky to interpret also.  Interestingly enough, I read a very interesting book back in the 70's about the 1974 DC-10 crash of Turkish Airlines flight 981.  It was written, I think, by some London Times reporters who had done a series of in-depth stories about the background and factors leading to the accident.  The reference is:

Destination Disaster, by Paul Eddy et al., Quadrangle, The New York Times Book Company, 1976. ISBN 0-8129-0619-5.

There were lots of things that led to the crash, involving several countries - including the US and the then fierce battle between Lockheed (L1011) and MacDonald Douglas (DC-10) to dominate the jumbo market (eclipsed eventually by the 747), which led to corner-cutting design choices on the DC-10 that were a factor in the crash.  

Relevant to this discussion was that the authors in the course of their research compiled and analyzed statistics of fatal accidents by airline, not country per se.  They made a pretty good case that there were statistically significant groupings of 'safe', 'less safe' and 'not so safe' airlines after trying to normalize by numbers of flights, passengers, etc.  Some of the big airlines in US and Europe got pretty good ratings (probably related to strong safety standards, good technology etc) but there were some surprises - e.g. TAP airlines has still, I believe, only 1 fatal accident in it's entire history.  I think Turkish airlines at the time was in one of the less safe groups and drew a lot of the public blame at the time, but corporate greed in the US was identified as one of the underlying factors which set the stage for an air safety 'perfect storm' in the authors' opinion.

I haven't come across any updated version of this type of analyses - probably because it's very sensitive on numerous counts and also, of course, the airline industry is now unrecognizably different from the early 70's records analyzed by the book, with deregulation, mergers, new airlines, old airlines out of business, etc. etc. etc.  

Cheers,
Torrence

[Jeroen D]

I agree that the statistics dont make much sense anymore. Especially if you look at fatalities and hull loses. For some reason certain, institutes keep measuring it that way and even over long periods, e.g. 10 - 20 - 30 years, which gives a very skewered view as the last decade has seen huge improvement, so why drag those historical numbers forward all the time. If you google 'worst" or best airline you typically get such a result. It also makes Air India show up as one of the worst airline due to a few incident with heavy loss of life many years ago.

Maybe looking at not just the big accidents, but at all incidents would present more usefull data, not sure.

So I was thinking more along the line how do you rate or quantify the total system which encompasses aviation safety.

The link shared by Sese is actually such an example from the FAA. Although it doesn't look at all aspects of aviation safety it has eight distinct areas which by the look of it, the can measure and quantify and thus come up with a rating and a relative position between countries compared to average.

Unfortunately the data in the the graph is a bit dated and I cant find a more recent one I do think it makes more sense to talk about aviation safety in those terms (sort of
look at leading indicators, process, implementation, adherence to rules etc) then to look at the end results in terms of crashes/fatalities  and such. Its one of these case where the old "the proof of the pudding is in the eating" doesn't apply. The taste so to speak is great, but you need to look into the recipe, the ingredients, the kitchen, the cooks resume etc. to form an opinion on how good it really is.

Are there any such evalutation sort of systems in place in the aviation world?

Jeroen

[John Golin]

I can tell you where they are NOT - Chipping Norton during the first week of November!

[Jeroen D]

So what's happening in Chipping Norton in November?
Jeroen

[Jeroen Hoppenbrouwers]

See logo.  :-)

[Jeroen D]

Right, got it!

[Jeroen Hoppenbrouwers]

Be especially careful for people staggering out of the yellow building somewhere during midday, just having completed a flight to inner Siberia in pitch black darkness, being both worn out and hung over, forgot their sunglasses, and don't remember to look right when crossing the road.

What? That was ten years ago? Oh my.

[Jeroen D]

Just a little update. I found this: http://www.icao.int/safety/Pages/USOAP-Results.aspx

You can build your own country specific graph comparing to the Industry average.

Conclussion when it coms to the Indian skies being the safest: Myth Busted.

I built the attached graph showing India, France, Germany, the Netherlands, the UK and the USA. Pay attention to the dates of the respective survey as well.

During the last four weeks there have been a lot of articles in the press about India not meeting ICAO requirements and or not implementing the findings of previous audits.

I played around with this graph a bit. Belgium did not do very well. Also, our favourite holiday destination as well as my wife's home country Barbados did not fare well at all.

I believe, to my original question, this USOAP status is probably the best way of looking at "safe skies". It doesn't look at accident statistics, but eveluates across the board the various processes, effectiveness, implementation and oversight of these in a particular country. Good stuff!

[/URL]
Jeroen

[Jeroen Hoppenbrouwers]

ICAO 2013 report just added.

http://www.skybrary.aero/index.php/Aviation_Safety_Statistics

[Jeroen D]

Thanks, the graphs I showed are against the overall USAOP status mentioned in this overall 2013 report.

The ICAO 2013 report doesn't say anything on particular countries or airlines. Its's all reported on regional level. The only one that gets singled out is Africa, which is getting additional attention! Still, it makes interesting reading. (Well, I think so, my wife probably wouldn't)

Jeroen