4 ms·

The Boeing Max 8 entered service in May 2017. Assuming a linear deployment rate, the 350 planes in service have seen an average life of 10 months. Assume 4 flig

The Boeing Max 8 entered service in May 2017. Assuming a linear deployment rate, the 350 planes in service have seen an average life of 10 months. Assume 4 flights/day, that's 420,000 flights so far. 2 have gone down. A best estimate of the likelihood that a plane goes down (MLE), p = X/n = 2/420,000 = 1/210,000 ~ Binomial(n=420,000, p=P(crash)). According to the Economist [1] the likelihood your plane goes down generally is 1/5,000,000. So based on the fact that the plane crashes had similar characteristics, the Boeing Max 8 is 25X more dangerous than a regular plane. 25X is the difference between surviving a commute on a bicycle vs a car [2]. [1 https://www.economist.com/gulliver/2015/01/29/a-crash-course-in-probability https://www.economist.com/gulliver/2015/01/29/a-crash-course...]. [2 https://www.riskcomm.com/visualaids/riskscale/datasources.php https://www.riskcomm.com/visualaids/riskscale/datasources.ph...]. EDIT: The Economist source that estimates a plane's p(crash) is questionable, for a passenger plane. If anyone wants to dig into this further, I found this source too: http://www.baaa-acro.com/crash-archives http://www.baaa-acro.com/crash-archives

anoncoward111 8y ago
This is excellent data and calculation work. I really appreciate it! I have a suspicion that if you were to remove all instances of terrorism and look at the crash rate of Boeing vs Tupolev, almost all Boeing planes would be way better except for the MAX 8. That is probably a meaningful piece of data too.
Scoundreller 8y ago
If I have an upcoming flight on a MAX 8, can I sufficiently compensate for any increased risk by taking the train to the airport instead of a 25km Uber drive?
- riazrizvi 8y ago
  Given the uncertainty about p(crash) that I mention, then to satisfy confidence limits, you should wear a styrofoam helmet for the full duration of your trip. And post a photo.
dis-sys 8y ago
> 25X is the difference between surviving a commute on a bicycle vs a car Meaningless and misleading comparison at best. According to numbers released by Boeing [1] itself, the original 737 designed back in 1967 had a hull loss of 1.75 per million flights, the 737 NG designed in the late 1990s to early 2000s had a hull loss rate of 0.27 per million flights. So Boeing 737 had a 7X less likelihood to crash as the results of 30 years of improvements. 25X difference is going to send the highly unsafe 737 MAX design back to the WWII level. Now think again whether WWII era aircrafts with similar crash likelihood should be allowed to carry passengers in huge volume in 2019. You number proves one thing and one thing only - FAA has the legal and moral obligations to ground all those highly dangerous 737 MAX immediately. [1] http://www.boeing.com/resources/boeingdotcom/company/about_bca/pdf/statsum.pdf http://www.boeing.com/resources/boeingdotcom/company/about_b...

anoncoward111 8y ago

This is excellent data and calculation work. I really appreciate it! I have a suspicion that if you were to remove all instances of terrorism and look at the crash rate of Boeing vs Tupolev, almost all Boeing planes would be way better except for the MAX 8. That is probably a meaningful piece of data too.

Scoundreller 8y ago

If I have an upcoming flight on a MAX 8, can I sufficiently compensate for any increased risk by taking the train to the airport instead of a 25km Uber drive?

riazrizvi 8y ago
Given the uncertainty about p(crash) that I mention, then to satisfy confidence limits, you should wear a styrofoam helmet for the full duration of your trip. And post a photo.

dis-sys 8y ago

> 25X is the difference between surviving a commute on a bicycle vs a car Meaningless and misleading comparison at best. According to numbers released by Boeing [1] itself, the original 737 designed back in 1967 had a hull loss of 1.75 per million flights, the 737 NG designed in the late 1990s to early 2000s had a hull loss rate of 0.27 per million flights. So Boeing 737 had a 7X less likelihood to crash as the results of 30 years of improvements. 25X difference is going to send the highly unsafe 737 MAX design back to the WWII level. Now think again whether WWII era aircrafts with similar crash likelihood should be allowed to carry passengers in huge volume in 2019. You number proves one thing and one thing only - FAA has the legal and moral obligations to ground all those highly dangerous 737 MAX immediately. [1] http://www.boeing.com/resources/boeingdotcom/company/about_bca/pdf/statsum.pdf http://www.boeing.com/resources/boeingdotcom/company/about_b...