Data preparing

[raed]

Hi:
If i want to calculate the Weibull parameter (alpha and Beta) i have to perform the following calculation:

i sort the components anodically according to their failure points, then i calculate 'H'

H=(i-0,3)/(n+0,4), where: i is the rank n is the number of the components.Why i do that and what is H exactly? (is it the acc. probability?)

thanks in advance

raed

[Adam]

H is called the Median Rank. This equation is actually an approximation for the binomial distribution for a 50% probability. The idea is to solve for the probability of failure of the ith failure out of n failures with a 50% confidence. In other words (in terms of the binomial distribution), we are looking for the probability of success of the ith trial out of n trials given a 50% cumulative probability.

For more information you can check the following link:

http://www.weibull.com/LifeDataWeb/probability_plotting.htm

[usara_reliable]

I agree with Adams,
H is generally called Bernard median rank used for the better accuracy. Beside this I is used for the rank adjustment in case of the Censored data. You can go different technique as reverse rank adjustment method . It is useful to give exact failure rank and weibull parameter calculation.

Please prefer Ebling BOOK.

Hope this help u.