Basic quantitative analysis help

[jeremy freeman]

I'm attempting to do a quantitative analysis of a set of data that, from my reading, is 'type I censored' or right-censored/suspended. I'm having trouble applying the equations and explanations from the books to the data via a spreadsheet(excel has the weibull function), i.e. 2 failures out of 24 tested. The test data appears to only have been partially accelerated with respect to time only. I haven't had any formal training or classes on reliability engineering.

[tarik]

I checked the Excel Weibull function; it seems that it can only return a PDF or CDF value for a certain time value given parameters value (which you don't have yet). I think you are still at the stage of estimating the parameters based on a set of data. For more info about parameter estimation in general, check the following link: http://www.weibull.com/LifeDataWeb/parameter_estimation.htm.

For info about the parameter estimation for the Weibull distribution, you can check the following link:
http://www.weibull.com/LifeDataWeb/estimation_of_the_weibull_parameter.htm

Once the parameters are determined, other calculations (various metrics, plots…) could be obtained.

[Dr Dave Olwell]

Jeremy,

With all respect, you probably shouldn't be doing that type of analysis in Excel, especially if you are not sure what you are doing. You'd probably benefit from taking a class on life data analysis, as well. Particularly if your analysis has a business impact, you want to know that what you are doing is correct because there are so many easy ways to go astray.

If it is a one time analysis, find a knowledgeable colleague or consultant to help you. If you are going to do it repeatedly, take a class so you have confidence that you are proceeding correctly.

Best wishes
Dr. Dave