# What Are the Odds You Survive an Airplane Crash?

George Bibel has written a fascinating book entitled Beyond the Black Box: The Forensics of Airplane Crashes.

I suspect this is one book that you are never going to find in the airport bookstores.

Bibel tells you when planes crash (focusing in particular on DC-10s). Forty-five percent of the crashes happen on landing, but remarkably these crashes account for only 2 percent of all the fatalities. The worst crashes are those when you are climbing or cruising (14 percent of crashes, but 37 percent of fatalities).

He can tell you why each crash occurred, describing the forces at work on an icy runway, the relevant coefficients of friction, and the impact of thrust reversers.

His chapters have uplifting titles like “In-flight Breakup,” “Pressure, Explosive Decompression,” “Burst Balloons,” and “Metal Fatigue: Bending 777s and Paper Clips.”

In spite of all this, it turns out that for most people in most crashes there is a surprisingly happy ending. Take, for instance, crashes that result in “total hull loss,” which means that the crash damages the airplane beyond economic repair. Of the 446 DC-10s ever delivered, 27 of them were involved in crashes that led to “total hull loss.” Overall in these crashes, 69 percent of all passengers and crew members survived. If you throw out the three worst crashes, the survival rate is nearly 90 percent!

#### Dr. Gary

This is an economics column, right?

Please break out the survival stats by the ticket prices paid by passengers.

Something tells me there is an inverse correlation here between price and value.

Plus, does that universally loathed middle seat offer an unanticipated benefit in a crash?

#### Cliff

Whether you'd be better off jumping out before the crash depends on the circumstances. In most cases, probably not, but if you have little downward velocity and a modest (say 150mph) forward velocity and the plane is going to crash into a giant tree in a field, you're probably better off jumping out. Few people die from intersecting a field with a 150mph forward velocity, as long as they don't run into something. Superbike racers fall off their bikes at upwards of 150 mph somewhat routinely and apparently it doesn't feel too bad as long as you don't hit anything.

#### W K

"There were recently two plane crashes in Brazil, both with no survivors. The first one happened with a Boeing 737, which crashed with an Embraer Legacy. The Legacy managed to land, but the Boeing fell right in the middle of the Amazon jungle. The other crash happened on landing when an Airbus A-320 couldn't brake and hit some buildings after the runway."

(Posted by Anselmo)

Both accidents were human issues. Since in Brazil we say that a nice child has many fathers, and a bad one does not has fathers, about both accidents, there are some people fighting on not to discover their mismanagements:

a) at Gol crash (new B737 x new Embraer Legacy) over Amazon, there was obviously someone flying on the wrong altitude.

b) at TAM A320 crash at São Paulo, there is a computer programming defect - made by EADS: one reversal component was put out-of-work, and the onboard computer misinterpreted this signal that the airplane was on air, not on ground, and so it did not freed the brakes to the pilot.

I notice that Patrick Smith has already made an appearance, but for the answer to a lot of your questions, you should go back through the "Ask the Pilot" archives at Salon.com where he specifically addresses a lot of these topics.

Or find his book at the library or bookstore, since he's got one of those, too.. shockingly it's called "Ask the Pilot."

For example

1. Actually, a LOT of crashes involve both survivors and fatalities. Pay attention to where your nearest exit is when you get on the plane, people! And don't wear high heels on the plane, ladies (or gents)!

2. No, you don't automatically die during water landings. There were quite a few survivors of one of the most famous water landing incidents off the coast of Africa (Nigeria, I think it was...) Ironically a lot of the deaths were because people wearing the underseat life vests ignored the safety instructions, and chose to inflate their vests before the crash, making it difficult for them to swim out of the fractured hull.

#### Sam C

"How far do you think only one engine can take us!?"

"Just about to the scene of the crash. And that's pretty handy, because that's right where we're headed! I bet we beat the paramedics there by a good fifteen minutes!"

}Ron White{

#### Henry Franconia

Re: Comment #22 Patrick Smith -- Excellent excerpt at the link you post, thank you. I encourage others to check it out, and Dubner and Levitt should prevail on you to post a different segment here.

#### jk

why buy the book? just google arnold barnett (MIT), clinton oster (indiana), and nancy rose (MIT). there are a lot of research available on the internet for free.

#### Jeffrey

I'm guessing that one of the "three worst crashes" was Air New Zealand Flight 901 (28 November 1979) which crashed on a sightseeing flight to Antarctica killing all 237 passengers and 20 crew.

#### corinne

If Herbert-age-7 were right, nobody would need a parachute to go skydiving. They could just take a board, jump out of the plane, get the board positioned underneath their feet, then when they are a few feet from the ground just jump off the board.

Don't try this. It won't work and you will die if you try it.

#### Patrick Smith

>> I would guess that most plane crashes have either a 100% death rate or a 0% death rate, while few have both fatalities and survivors. >

That's a common question, but there is no good answer. As a practical rule, *all* commercial airliners are equally safe. With crashes so rare, to cite one model as "safer" than another is to engage in a tedious statistical hairsplit that effectively proves nothing. That most accidents share human error as a direct or contributing cause further muddles any plane-specific data.

- PS

#### Gabe

The whole survivablity issue all depends on what you consider a crash. It's pretty clear cut when a plane does an uncontrolled dive into the Atlantic Ocean, but what about when a landing gear collapses and all the passengers feel is a big thud. It's like counting a slow-speed fender bender in rush-hour trafic as a car crash.

The Aviation Safety Network Web site (http://aviation-safety.net) is an excellent source of data. Perhaps what's most alarming is the frequency of crashes occuring in other (mostly poor) countries that you never hear about in the media here.

#### Tom

I would guess that most plane crashes have either a 100% death rate or a 0% death rate, while few have both fatalities and survivors. If the crash involves something like a high-speed collision or explosion then it will be deadly (probably to all aboard), but if not (like in a relatively controlled landing where the plane does not use its landing gear as it should) then it won't be all that serious for those on the plane (though the plane itself may be unsalvageable). It seems strange to lump these two kinds of "crashes" together. Perhaps a separate look at fatal crashes and nonfatal crashes would be more informative.

#### Anselmo

There were recently two plane crashes in Brazil, both with no survivors. The first one happened with a Boeing 737, which crashed with an Embraer Legacy. The Legacy managed to land, but the Boeing fell right in the middle of the Amazon jungle. The other crash happened on landing when an Airbus A-320 couldn't brake and hit some buildings after the runway.

#### jamie

Come on guys....Herbert is obviously refering to the Loony Tunes method of escaping a crash! jeez!

#### Mitchell

Jonathan K, unfortunately you are wrong. It works for Bugs Bunny.

#### Jonathan

#8. Matt:

The Swissair Flight 111 crash occurred off the coast of Nova Scotia (specifically in the area of Peggy's Cove) not Labrador.

#### stretchy54

I picked up Professor Bibel's book at his recent book signing at the Smithsonian Air & Space Museum in DC. What a great book to help motivate high school students understand the relevance of science and engineering. - stretchy54

#### Matt

So, Herbert, I guess you're one of those people who thinks jumping right before the plummeting elevator hits the bottom of the shaft is the secret to survival...

I think you're probably joking, but just in case: you can't forget about momentum!

#### Aaron V

FYI the most common source of crashes are know as CFIT, controlled flight into terrain. Mid air collisions are relatively rare. And regardless were the accident happens, everything lands at some point.

#### SDC

Actually if the board were heavy enough, and your legs strong and (especially) fast enough, the jumping off the board thing might work via Newton's 3rd law maybe.