Hacker News Reader

Parent Full thread

throw0101b

> Might I introduce you to the Tupolev Tu-134?

Also perhaps the 737-MAX. :)

* https://en.wikipedia.org/wiki/Maneuvering_Characteristics_Au...

View on HN

Given how much the max family and 8 are flown, I don't think this holds up.

Some searching says...

1. >1000 flights per day for the family

2. 2 US carriers accounting for more than 200 Max 8 per day

So one would have to get to 20,000 flights after two crashes to get 2 nines. We're well past that threshold

From inception to the second crash, numbers were abysmal. The 737 Max was crashing at a rate of about one per hundred thousand flights, two order of magnitude worse that the 737 NG (one fatal crash per 10 million flights).

Said differently, the Max 8 was working safely 99.999% of the time, while the 737 NG was working safely 99.99999% of the time. An order of magnitude better than 99.99%, but two orders of magnitude worse than expected...

It is certainly a lot safer now. Hopefully even better than the 737 NG.

The probability of not having a crash after 20,000 flights, with 99.99% chance per flight, assuming no serial correlation, is 13%.

Yes, there is a better, more accurate method than my napkin math, which was only to provide a baseline most could understand to see we are well beyond the two nines

Reply on news.ycombinator.com