Hmmmm... From Canada, joined May 1999, 2114 posts, RR: 5 Posted (15 years 4 months 5 days ago) and read 2023 times:
This is an interesting point about statistics. Often we read forum members quoting statistics to support a claim that one kind of airliner is safer than another. I don't like statistics for that type of argument. Here's why:
With One Crash, Concorde Ranks Last in Safety
By Don Phillips
The crash of the Air France Concorde near Paris on Tuesday transformed the supersonic aircraft from the safest plane on earth to the most dangerous, statistically speaking.
This quirk shows the dangers of measuring aviation safety by numbers alone in an era when flight is so safe that just one accident can skew the ratings dramatically. But statistics can also tell useful stories.
For more than three decades, the Concorde flew the earth's airways with no crashes, no deaths and no injuries more serious than bumps and bruises from occasional evacuations after nonfatal incidents. That means that on the industry's standard safety measure, "hull losses" per million flights, it scored a perfect zero. A hull loss is counted when an airplane is damaged so badly that it will never fly again.
And because the Concorde has been in service far longer than other aircraft that now have zero hull-loss ratings – the Airbus A330 and 340 and the Boeing 777, 737NG (for "new generation") and 717 – many people considered its record to be the best.
However, because there are so few Concordes and because each flies fewer than 1,000 hours a year, the Tuesday crash boosted the hull loss per million flights figure to 11.64, according to statistics developed by Boeing Co.
This is by far the worst record among jetliners flying today. It is exceeded only by those of the first generation of jets, which have long since been phased out – the Comet, the Caravelle, the Trident and the VC-10. Together they racked up an average figure of 15.51.
"I am bothered by continual references in all the media about the safety record of the Concorde," said John Purvis, Boeing's former safety chief who is now a consultant and partner with Safety Services International of Seattle. "The entire fleet was good up to now, but in fact, it had less than 80,000 takeoffs and landings.
"I think people have to realize that safety is a numbers game," Purvis said. "Concorde had a good record partly because there weren't many of them, they were underutilized and had special maintenance handling and crews that were the cream of the crop."
The Boeing 737, by contrast, has had 77 crashes but still has an excellent safety record, ranging from 1.25 per million to .43 to zero hull losses for different versions of the plane, because it is the world's most common airliner. The huge 737 fleet flies more hours in one week than the Concordes have flown in their entire existence. The 737 has great exposure to potential crash situations but still avoids accidents.
At first glance, the wide-body McDonnell Douglas MD-11, with five crashes and a hull-loss rate of 6.54 per million departures, appears to be the least safe subsonic aircraft now flying. This is worse than the first-generation U.S. airliners – the Boeing 707 has 116 hull losses and a rate of 6.51, while the McDonnell Douglas DC-8 has 72 losses and a rate of 5.91.
However, a more careful examination tells a different story. First, the MD-11 has not yet flown a million departures. This makes it susceptible to statistical quirks. And several of the crashed MD-11s were cargo planes, meaning there was no danger to the flying public.
The 707 and DC-8, on the other hand, suffered from the growing pains of the early jet age. And now most of them are flying in the Third World, where crash rates tend to be higher for all types of planes.
Safety statistics can tell stories that defy conventional wisdom.
For example, John Lauber, vice president for safety and technical affairs at Airbus Industrie, conducted a study comparing the hull-loss rates year by year for each of the four generations of passenger jets that are still in the air.
In each case, the hull-loss rate was very high in the first few years of flight for each generation. It then dropped to a very low rate for each generation in all subsequent years.
Airbus concluded that pilots, maintenance crews and engineers using new planes must go through a significant learning curve before they can establish optimum safety.
F4N From , joined Dec 1969, posts, RR:
Reply 1, posted (15 years 4 months 4 days 10 hours ago) and read 1949 times:
Hmmmm: I have to agree with you 100% here. As someone who has to use statistics virtually every day in the course of my employment(assessing failure rates, performance analysis, risk, ect), I can assure everyone that statistics can easily be manipulated by interested parties. In the case of commercial aviation, we often see claims by the various manufacturer propaganda machines that they are the "world leaders" or that one of their planes (let's call it the A377 since no individual party is blameless here) is preferred by more travellers than any other a/c. Or, that the B327NG is the most popular, fastest-selling, most fuel-efficient, most comfortable, farthest flying and lowest cost a/c " in its' class".
Ever wonder how you can have TWO most preferred a/c (by people who fly over 6 kabillion miles a year, of course)?
Or, how you can have TWO most fuel efficient, lowest cost per seat mile planes(in their class)?
Simple! You manipulate statistics. Crunch things like fixed costs, variable costs, miles flown, where flown, by how many a/c, blah-blah-blah and presto...you can make a claim! You aren't necessarily doing anything wrong, but....
Of course, this is a gross oversimplification, but I'm sure everyone gets the point:
Be wary of any "biggest, best, most" claims unless the data to support it is supplied as well.
Still wondering what happened to all the airline orders that were supposed to be at FIA...
Na From Germany, joined Dec 1999, 11381 posts, RR: 9
Reply 2, posted (15 years 4 months 4 days 10 hours ago) and read 1940 times:
I think rarities like the Concorde should be excluded from statistics. Otherwise: Why is Howard Hughes "Spruce Goose" not included then?
Statistics make only sense, if the subjects are comparable. The Concorde is not comparable with any other civil transport. Not in numbers, not in performance, not at all.
As well its not very clever to include brandnew aircraft types younger than 5 years. The 777 is still too young, the A340 not, beeing around 8 years. By far the most losses happen after more than 10 years of service.
To include Concorde in statistics is like what a german magazine wrote recently: Rolls-Royce is the most successful car manufacturer in Germany this year. Means, compared to 1999 they sold 3 times more cars here. 10 vs. 30.
Ab.400 From , joined Dec 1969, posts, RR:
Reply 5, posted (15 years 4 months 2 days 5 hours ago) and read 1889 times:
Air-travel claims to be the safest way of traveling, but this is only true in a statistic which puts the number of miles in favour. Compared to the number of single trips it is way behind travelling by car or train.
Surf From , joined Dec 1969, posts, RR:
Reply 6, posted (15 years 4 months 2 days 4 hours ago) and read 1881 times:
Exactly Ab. 400: Everyone says the statistics are skewed when it comes to the concorde's safety and yet, the statistics that point out that flying is the safest way to travel are flawless....Me thinks aviation enthusiasts believe what they want to believe.
Aviation may have fewer accidents per so many millions of miles but flying is really the most DANGEROUS way to travel, in terms of survivability. Being hurled through the air in a giant metal tube filled with tons of flammable fuel where any number of human, weather, or mechanical problems can bring it down in a huge fireball. They dont find your body. They find PIECES of bodies....