cancel
Turn on suggestions
Auto-suggest helps you quickly narrow down your search results by suggesting possible matches as you type.
Showing results for 
Show  only  | Search instead for 
Did you mean: 
Try the Materials Informatics Toolkit, which is designed to easily handle SMILES data. This and other helpful add-ins are available in the JMP® Marketplace
Choose Language Hide Translation Bar
tahir
tahir
Level II

Reliability calculation with only failure

May 13, 2016 04:53 AM (4457 views)

Hi,


I have the following test data

Nr of hours                          test results

2400                                    not failed, test stopped

2400                                    not failed, test stopped

2400                                    not failed, test stopped

2600                                    failed

I tried to use the 'life distribution' but i dont get any result as i have one failure so i want to know how to to calculate reliability at 750 hours?


BR,


/Tahir

0 Kudos
1 REPLY 1
Mark_Bailey
Staff Mark_Bailey
Staff

Re: Reliability calculation with only failure

May 13, 2016 05:04 PM (4357 views)  |  Posted in reply to message from tahir 05-13-2016

With only one observed failure, you cannot estimate two model parameters (such as alpha and beta for Weibull). You can perform a WeiBayes analysis if you only need to estimate one of these parameters. Select Weibull as the distribution model at the top in Compare Distributions. Then click the red triangle next to the Parametric Estimate - Weibull for the model results and select Fix Parameter. A new section opens where you can enter the known value and then estimate the other.

Note that you can estimate the independent parameter but you cannot evaluate goodness of fit with only a single observation.