Rolling your own modeling algorithm – Weibull analysis
Weibull analysis is a well-known technique for understanding the reliability of physical assets over time and is not directly supported in Modeler. The analysis is based on understanding the failure distribution of physical assets such as bearing, switches, electrical components, pipes (think corrosion), and so on. The only inputs to the model are the times to failure. A Weibull failure distribution is fit to the empirical failure distribution. In the two-parameter Weibull model, there are Alpha and Beta parameters. The parameters give insights into the failures:
- Beta < 1 indicates infant mortality
- Beta = 1 indicates random failures
- Beta > 1 indicates wear-out
Alpha is the number of cycles where ...
Get IBM SPSS Modeler Cookbook now with the O’Reilly learning platform.
O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.
