[chbot] Chchrobotics Digest, Vol 157, Issue 11

[Bevin Brett]

Correction - and I still don't quarantee anything about this

If the failure rate is 3/100 = p,   then the standard deviation would be

sqrt(npq) = sqrt(n f/n (n-f)/n) = sqrt(n​*f*(n-f))/n ~= n sqrt(f) / n   for small f = sqrt(f), in your case sqrt(3) ~= 1.4

The question about "how many" was a military secret in one of the WW.  The answer is YOU DONT fix the sample size in advance.  Instead you sample one at a time and watch the numbers, and stop sampling after the fails gets low enough or high enough, thereby saving unnecessary sampling

I don't know the details, but I suspect you are asking is for  f + sqrt(f)*2 to become acceptably low, and stop if f - 2 sqrt(f) is too high

https://www.weibull.com/pubs/2015_RAMS_right_sample_size.pdf seems to be in the right ballpark

/Bevin
2015 Reliability and Maintainability Symposium<https://www.weibull.com/pubs/2015_RAMS_right_sample_size.pdf>
Athanasios Gerokostopoulos is a Reliability Engineer at ReliaSoft Corporation. He is involved in the development of ReliaSoft’s software products and the delivery of training seminars and consulting projects in the field of Reliability and Quality
www.weibull.com

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.ourshack.com/pipermail/chchrobotics/attachments/20201018/0d9809ae/attachment.html>