Expected Overall Compliance 1

[HughL]

Dear All,

If I test a Sample of Material with multiple specified parameters I can calculate the expected compliance for each parameter.

e.g.
Parameter 1 is expected to comply 90% of the time
Parameter 2 is expected to comply 95% of the time
Parameter 3 is expected to comply 85% of the time

If any parameter fails the overall result for a Sample fails.

So from the above information is it possible to calculate the expected compliance of a Sample with all respect to all parameters?

TIA Hugh

 

[GregLocock]

That depends on whether there is any correlation between the parameters.

If they are truly independent then the probability of a pass is 73%

Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[HughL]

Greg,

Thanks for the response

I think I get and grasp your logic and get an answer of 72.675%

Thus;

85% - (85% of 10%) = 85% - 8.5% = 76.5%
76.5% - (76.5% of 5%) = 76.5% - 3.825% = 72.675% (=73ish%)

Am I correct in this?

The parameters are not independent (of course)!

Kind regards Hugh,

 

[GregLocock]

Well you got the right answer, so in the new maths that means you understand it.

In the old days

p(A AND B AND C)=p(A)*p(B)*p(C)

If they are not independent and you understand the physical dependency then Monte Carlo modelling is often a good way to understand what is going on.

If on the other hand you just have a statistical description of the interrelationship then draw a Venn diagram and sum appropriately.





Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[HughL]

Greg,

Thanks for your help again.

p(A)*p(B)*p(C) looks less complicated.

Seems I am going through 'old days' without actually having lived them!

Much appreciated.

regards Hugh

 

[PSE]

It just goes to show that not everything that is "new" is actually improved. While it might be fodder for a separate thread, to the old math. I can tell I am going to have fun when it comes to work with my son through the new math.

Regards,