PDA

View Full Version : variance of the probability density function with sudden death


SimSch
November 13th, 2008, 07:08 AM
Hello I have the following problem:
With Weibull++, guarantee data from a element and the sudden death method I calculate the reliability function. Then I want to get information about the variance of the probability density function. The problem is I use to get the weibull function the sudden death method. And because of this, I think is not right to calculate the variance in the data sheet because there are only the earliest failures.
Is there a possibility to get the variance of the function like I get the mean time to failure?
Thanks for your help!

David
November 13th, 2008, 10:51 AM
Hi SimSch,

Just to clarify, you would like to calculate the variance of the probability density function (pdf). Is this correct? Or are you trying to calculate the variance of the probability of failure (1 - reliability). Unfortunately, Weibull++ does not have the capability to estimate the variance of the pdf. This is something that can be done by hand though.

I hope this helps.

SimSch
November 14th, 2008, 05:10 AM
Hallo,

I want to calculate the variance of the probability density function. It is easy to calculate the variance if you have all failures. For example:
First failure 100 h
Second failure 200 h
Third failure 250 h
Fourth failure 400 h
=>variance is 12500 h^2
=>Standard deviation is 111,803399

But if you work with sudden death then you don’t have the information about all failures. For example:
1 failure at 100h
10 survived at 100h
1 failure 200 h
10 survived at 200h
It is not right to calculate the variance with the formula above, because you need the information about all failures and you have only information about the first failures of a lot.

Or do you think I am wrong? Have you a solution for my problem?

Is it possible to calculate the variance of the probability of failure with Weibull++?

Thanks!

David
November 14th, 2008, 09:52 AM
For the data set indicated (100, 200, 250, 400), assuming a Normal distribution, the std is actually equal to 125. Therefore, the variance is equal to std^2 = 15625.

It sounds like what you are looking for is the variance of the data based on the underlying distribution. You can get this using the Function Wizard in Weibull++ 7. From the Folio menu, select Insert General Spreadsheet. Now, from the Data menu select Function Wizard. From the list of available functions on the left side, select AVAR. Then click on Select... below Data Source to select the data sheet for which you want the variance to be calculated. When finished click on Insert and the variance for the data will be displayed in the spreadsheet.

I hope this helps.

SimSch
November 17th, 2008, 01:19 PM
Thanks a lot!
That is exactly what I need.