Originally Posted by
AirRabbit
JammedStab:
You cited 4 terrible accidents dating back to 1974. Statistics (if you believe in such mundane things) say that there are approximately 100,000 airline flights per day – and if you throw in charter and cargo flights the number doubles. And I am fully aware that there were many more accidents than those you cited – although, as I think you would agree, not ALL of those additional accidents could be attributable to design/manufacturing/structural defects. So, given these numbers during the 40 years of time since the Turkish DC-10 accident, there were more than 14.6 billion flights, and during this time there were 497 airline accidents - not an estimate. With these numbers, it is easy to see that the ratio of accidents to flights is astonishing low … about 1 accident for each 29.4 million flights – a rather low average by anyone’s estimation. Of course ONE accident is one too many. But, again, we’re dealing with humans. Humans are fallible. Accidents DO happen. I’m not happy about it – nor is anyone I know happy about it. But to offer some kind of comparison outside of the aviation community, you might be interested to know that in the US – a reasonably modern country with reasonably modern safety standards – there is 1 automobile accident for every 30 automobile trips every year. I would have provided a similar data comparison for Europe or the world had I been able to find the relevant data. But, basically, the US data for automobile accidents should show the relative safety ratio to that ratio in aviation, even after using the actual number (497) of airplane accidents over that 40 year history, is a robust 29.4 million to 1 … and that is with a ratio of 40 years to 1 year.
You are of course, absolutely correct. The overall certification system has resulted in an extremely safe system and I have never denied that. And yes, there are other types that have been certified incorrectly and crashed such as the Comet and ATR-72. And in order to respond to the "that was 20 or 40 years ago argument", it sure was fortunate that the 787 battery fire happened on the ground. The result, a decertification of a recently certified aircraft where the authorities who really not particularly aware of how the whole system worked just decided to trust Boeing's good word.
http://aviationweek.com/awin/faa-boe...-certification (strongly suggested reading).
Originally Posted by
AirRabbit
As I said, accidents do happen – sadly – but they happen, regardless of the reason. However, to say that aviation accidents (particularly with the substantially reduced number that could be even remotely attributed to a structural or mechanical problem) are a result of “cavalier” decisions made by the certification portion of the industry is so far out of bounds as to logically be classified as “non-playable,” and certainly should not be cited as any logical justification for making individually preferred decisions by flight crew members.
As stated above, the overall system is safe. How often does an engine failure occur at V1 which happens to e very close to VMCG. Not often. But it is going to happen some day when there is a strong crosswind and at minimum, there will be an excursion, perhaps worse as the crew attempts to do what you and the certification authorities say is the only safe thing to do.
Think of all the expensive hoops that the certification authorities make companies do to get certain things done. Try getting an EFB installation approved. They even do decompression tests on the darn things. Yet, everyday, hundreds of airliners are taking off in strong crosswinds and exposed to what is in fact a significantly higher VMCG than what their performance figures tell them. Why don't the authorities just mandate a higher V1 based on the crosswind component? It would just be another entry on the newly updated OPT or paper graph chart to get new V-speeds based on some engineering analysis. I believe engineering analysis based on technical information from actual tests done in the '60's is all that is done for slippery runway numbers. While that is not certified information but it proves that it can be done. But maybe this would frequenly cut into payload and cost the industry money which might be a cavalier attitude.