<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
</head>
<body style="overflow-wrap: break-word; -webkit-nbsp-mode: space; line-break: after-white-space;">
<div><br>
</div>
Derek,
<div>5 hours per day is way too low. Airplanes are very expensive, airlines are low profit margin businesses (which is why they are so interested in other, more highly profitable side business like credit cards), and airplanes only earn revenue when they are
in the air. The airline’s incentive is to keep their airplanes moving as much as possible. Something like a 737 or A-320 is far more likely to be spending as much as 14 hours in the air each day. It can be interesting (to an aviation geek like me, at least)
to go to something like Flight Aware (<a href="https://www.flightaware.com">https://www.flightaware.com</a>), pick a random flight, then track it backwards using the Track Inbound Plane button to see where it’s been in the past few days. Here’s one particular
United Airlines 737-900ER’s (UA hull #3475)’s activity for just today:</div>
<div><br>
</div>
<div>— Newark, NJ (depart 7:48am Eastern) to Minneapolis, MN (arrive 9:41am Central) as UA 1750</div>
<div><br>
</div>
<div>— Minneapolis, MN (depart 10:59am Central) to Denver, CO (arrive 11:55am Mountain) as UA 2802</div>
<div><br>
</div>
<div>— Denver, CO (depart 1:59pm Mountain) to Minneapolis, MN (arrive 4:59pm Central) as UA 732</div>
<div><br>
</div>
<div>— Minneapolis, MN (depart 6:05pm Central) to Denver, CO (arrive 7:20pm Mountain) as UA 2027</div>
<div><br>
</div>
<div>— Denver, CO (depart 9:00pm Mountain) to Seattle, WA (arrive 11:00pm Pacific) as UA 2851</div>
<div><br>
</div>
<div>That’s a bit over 14 hours in the air (well, it includes taxi time too but you get the idea). That very same plane is scheduled to fly out of Seattle back to Denver at 5:00am as UA 1232 to begin another day of revenue generation.</div>
<div><br>
</div>
<div>The bigger planes on long haul routes are likely to have fewer hours in the air because of time zone issues and the like. Turn-around times on international flights are usually a lot longer than 1-2 hours. I would guess 9-12 hours per day would be reasonable.</div>
<div><br>
</div>
<div>All that said, see Figure 1 on page 7 of AC 25.1309-1A, the Probability vs. Consequence Graph. Probable failure conditions are expected to require no more than Abnormal procedures to safely address. An example would be something like needing to go through
a non-normal checklist. Happens all the time, and the passengers on the plane probably don’t even know it happened. Improbable failure conditions are expected to involve no more than Emergency procedures (e.g., engine out) and possibly incur some airplane
damage (e.g., UA 328 in 2021, QF 32 in 2010). Surely the passengers would be aware of this, and it may make the local and regional news. Depending on the amount of airplane damage, it may even make the national news. But not likely the international news.
QF 32 did make international news but that’s probably due to it being the first major failure of a really big airplane. Extremely Improbable failure conditions, the ones that result in adverse effects on occupants or a catastrophic accident (e.g., SWA 1380
in 2018, Lion Air 610 in 2018, Ethiopian 302 in 2019) are the ones you see in the international news. Definitely a big deal, but—fortunately—they don’t happen all that often.</div>
<div><br>
</div>
<div>So if you double or triple your numbers below to account for 10-15 flight hours per day instead of the 5 you used, you get:</div>
<div><br>
</div>
<div>— 1 X 10^-5 equates to 2.5 to 3.75 Abnormal procedures per day</div>
<div><br>
</div>
<div>— 1 X 10^-7 equates to one Emergency procedure or Airplane damage every 30 to 45 days</div>
<div><br>
</div>
<div>— 1 X 10^-9 equates to one Catastrophic Accident every 6 to 10 years</div>
<div><br>
</div>
<div>Keeping in mind that a lot of airplane incidents and accidents are not mechanical (e.g., pilot error, meaning we need to exclude those events from this analysis), these calculated failure rates are probably not that far from what we are actually seeing
in practice.</div>
<div><br>
</div>
<div><br>
</div>
<div><br>
</div>
<div>— steve</div>
<div><br>
</div>
<div><br>
</div>
<div><br id="lineBreakAtBeginningOfMessage">
<div><br>
<div>On Aug 21, 2024, at 1:41 PM, Derek M Jones <derek@knosof.co.uk> wrote:</div>
<br class="Apple-interchange-newline">
<div>
<div>Paul, Steve,<br>
<br>
<blockquote type="cite">“Can you (or anyone on the list) help me understand how the committee arrived at 10^-5, 10^-6, 10^-7, 10^-8 as targets?”<br>
</blockquote>
...<br>
<blockquote type="cite">(1) Probable failure conditions are those having a probability greater than on the order of 1 X 10^-5.<br>
(2) Improbable failure conditions are those having a probability on the order of 1 X 10^-5 or less, but greater than on the order of 1 X 10^-9.<br>
(3) Extremely improbable failure conditions are those having a probability on the order of 1 X 10^-9 or less.<br>
</blockquote>
<br>
Say 25K commercial aircraft<br>
https://about.ch-aviation.com/blog/2022/06/30/june-2022-global-fleet-size-analysis-by-ch-aviation/<br>
<br>
Assume they each fly 5 hours per day, 365 days per year (I'm sure<br>
people on this list have much more accurate numbers)<br>
<br>
1 in 10^-5 per hour equates to 1.25 failures every day<br>
1 in 10^-6 per hour equates to 0.875 failures every week<br>
1 in 10^-7 per hour equates to 1.125 failures every 90 days<br>
1 in 10^-8 per hour equates to 0.456 failures every year<br>
1 in 10^-9 per hour equates to 0.9125 failures every 20 years<br>
<br>
-- <br>
Derek M. Jones Evidence-based software engineering<br>
blog:https://shape-of-code.com<br>
<br>
_______________________________________________<br>
The System Safety Mailing List<br>
systemsafety@TechFak.Uni-Bielefeld.DE<br>
Manage your subscription: https://lists.techfak.uni-bielefeld.de/mailman/listinfo/systemsafety</div>
</div>
</div>
<br>
</div>
</body>
</html>