O'Reilly logo

IBM SPSS Modeler Cookbook by Scott Mutchler, Tom Khabaza, Meta S. Brown, Dean Abbott, Keith McCormick

Stay ahead with the world's most comprehensive technology and business learning platform.

With Safari, you learn the way you learn best. Get unlimited access to videos, live online training, learning paths, books, tutorials, and more.

Start Free Trial

No credit card required

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 ...

With Safari, you learn the way you learn best. Get unlimited access to videos, live online training, learning paths, books, interactive tutorials, and more.

Start Free Trial

No credit card required