It is also important to look at the individual accidents and their respective reports. I understand what the statistics you’ve calculated say. But there is more nuance to each individual accident. It’s not that hard to look up all 11 hull losses. And there is no debate on the MD11s handling characteristics as stated by both the NTSB and FAA. But 3 of the hull losses were Korean 6316, CFIT due to confusion on meters and feet. Avent 324 which tried to takeoff without takeoff power. The captain had been flying an A340 for 6 months. The FO had not flown in over a year and the two other crew had 0 time in type. And lastly Swissair 111, which we all know about.

China eastern 583 was an inadvertent slat deployment as well causing 2 fatalities. The flap handle was then modified with that purple plastic to prevent accidental deployment.

For the bounced landing accidents:
FDX14
FDX80
LH8460

China 642- landing in tropical storm 26G38 direct x wind. The aircraft touched down on its right engine separating the wing.

FDX87.
45 knot airspeed discrepancies. And continuing the approach and landing fast.

And there was the UPS reject above V1.

Im not trying to dismiss the numbers or the inherent instability. Im just trying to provide some information that you can think about. So there is more information than just 3.125.

 

Wasn't the UPS 61 overrun at ICN a result of the center gear failing and twisting such that the airplane would never reach Vr on the runway?

 

I need to do some more in depth research on this. The report is only in Korean and I cannot find it. I’ll have to look for it. But here is a link to photos of the landing gear.

https://australianaviation.com.au/20...unway-overrun/

 

Yes, the captain was stuck in no man’s land. Over V1, but unable to accelerate.

 

Here is the report. Gonna run it through a translator https://araib.molit.go.kr/USR/airboa...ce=chatgpt.com

 

Copied from section 2.5 of the report

edit. I’m not here to question anything or say right or wrong. This is the first time I’ve ever seen this report translated.

2.5 Aircraft Performance if Takeoff Had Continued

Up to the point on the runway where engine thrust was reduced as part of the rejected takeoff action,three distinct segments of tire friction markswere observed on the runway for aircraft N277. FDR data show that there was adecrease in longitudinal accelerationin the takeoff direction up to the initial rejected takeoff actions.

Thefirst decrease in longitudinal accelerationoccurred whenTire No. 10 ruptured and the center landing gear (CLG) drag brace was damaged. Thesecond and third decreases in longitudinal accelerationlikely occurred when thedamaged CLG wheel and tire assembly moved rearward toward the aft fuselage, and when itstruck the aft fuselage and then fell back down and again made frictional contact with the runway.

At1.7 seconds after V1, asudden change in acceleration in all three axeswas recorded on the FDR. This abrupt change in accelerationrecovered approximately 0.5 seconds later.

During the approximately1-second interval between the sharp decrease in longitudinal acceleration and the initiation of the rejected takeoff, theaverage longitudinal acceleration was +0.15 G, corresponding to an acceleration of approximately2.8 knots per second. Immediately prior to the abrupt acceleration change, theaverage longitudinal acceleration was +0.17 G, corresponding to approximately3.3 knots per second.
Therefore,the longitudinal acceleration rate immediately before the rejected takeoff decreased by approximately 14%.

Performance evaluation indicated thattotal brake efficiency was reduced by approximately 52%, due tohydraulic fluid loss in Brake System No. 1 and the loss of braking capability of the CLG. According to the same performance evaluation,failure of the CLG increased the stopping distance by approximately 500 feetcompared with normal braking (maximum pressure applied to both brakes). With the CLG defect andonly 60% brake efficiency available on the remaining landing gear,more than 1,500 feet of additional stopping distancewas required compared to normal conditions.

However,had the flight crew continued the takeoff, the aircraftwould have met the takeoff performance requirements. Theshort-term effect of the failurewas limited tothe inability to retract the CLG. Climb performance calculations assumeone-engine-inoperative (OEI) conditions, which have a significantly greater impact than a CLG retraction failure. Withall engines operating normally, the reduction in climb performance due to the CLG remaining extendedwould have been negligible.

Although the aircraftmight have safely become airborne, theimpact on landing performance was not evaluated. While asimulator evaluation of continuing the flightwas conducted at the UPS training center,abnormal conditions after liftoff were not simulated. The outcome of any scenario different from the actual eventshould not be used as an indicator of what might have occurred had the crew taken different actions. Numerous unknown variables exist thatcannot be faithfully reproduced in a simulator, including theflight crew’s response to the emergencyandaircraft handling with a fractured drag brace.

Footnotes (translated)

82)Approximate decreases/increases: longitudinal acceleration −0.25 G, vertical acceleration −1.4 G, lateral acceleration approximately +0.3 G to the left.
83)The reduction in acceleration began at176 knots. The tire warning aural alert began at180 knots. If the flight crew had decided to continue the takeoff, aircraft rotation would have begun at185 knots (Vr).

 

They were also pointed right at North Korea fwiw.

 

Of course we can do this, but if the goal is to somehow explain away the MD11s poor safety record and say that they're all isolated incidents that aren't really representative, then we should do the same for the other airplanes. The statistic in question is hull losses per departure, so only the hull losses which occurred after setting takeoff power would be counted. Consider the 777...if we play the same game there we note that the triple has had a total of eight hull losses, but only 5 of these occurred after setting takeoff power, and these would be the ones counted in this statistic:

1) British Airways 38 - fuel starvation due to ice in the fuel system
2) Asiana Airlines 214 - crash on final due to unstable approach
3) Malaysia 370 - unknown...lots of speculation, most likely theory that fits the evidence is that one of the pilots deliberately crashed to commit suicide
4) Malaysia 17 - SA-17
5) Emirates 521 - Crew didn't add power during a rejected landing

(For the sake of completeness here are the other three are EgyptAir 667, Singapore Nov2017, and Ethiopian Jul2020. These all occurred during ground handling.)

So considering the 5 777 hull losses which occurred in flight, we have at least one (Malaysia 17) and possibly two (if you include Malaysia 370) which is/are not in any way due to a problem with the aircraft. Of the remaining three, two were crew induced so that we really only have one (BA38) where the causal factor was a design/engineering issue.

The upshot is that over millions of departures the statistical anomalies should wash out, so that by considering the whole body of work we can get a picture of how safe or unsafe an aircraft is...and the result is that the MD11 is profoundly unsafe...which is precisely why it is no longer operated by any passenger carrier. Consider that the last MD11 was built in October 2000 making it just over 25 years old, while DAL, UAL, and AA all operate aircraft that are the same age or older (apparently DAL has a 757 which is 35 years old).

 

I agree with everything you have to say except 1 thing. The 777 is a much better aircraft. It also was a clean sheet design not based of something designed in the 1960s. I’m not trying to rationalize the safety record. I’m just trying to present some evidence on when people say they won’t get on one again. And trust me, I have that exact emotional reaction after this most recent accident. So I completely understand. I just wanted to point out some specifics.

The only thing I disagree with you on is that it was retired from pax use due to safety. It was retired due to its fuel consumption, lack of range, excessive MX costs, and suitable competitor aircraft that performed better (777 and A330). Multiple major Us and EU airlines operated the type for over 10 years in the US and 20 years in Europe without any “major” incidents. The MD11 was built 10 years too late and is a design of the past. While the 777 in some cases can be considered the best widebody aircraft ever built.

 

Agree. I've never flown the MD11/10 or the DC10, so from an outsiders perspective it looks like turning the DC10 into an MD11 really caused some problems. But I will never set foot on one again, especially after seeing these statistics.