Ernst Hjalmar Waloddi Weibull (18 June 1887 – 12 October 1979) was a Swedish engineer, scientist, and mathematician. (source Wikipedia)
The ultimate answer to the topic question, is flexibility. This distribution can take almost every shape. If we are looking on two parameters Weibull distribution, it can take shapes of exponential, normal and log normal distributions. If you hesitant which PDF to choose, to fit the your data results – Weibull can be an always not the worse choice. In the next graphs you can see the most common distributions used by engineers for reliability purposes and the Weibull 2p, that can mimic those:
As it seems from the graphs, it looks very close fit to the 3 major distributions, so what make it possible to happened? The secret ingredient is in the math of how Weibull 2p PDF looks like:
b – is what called, the shape of the distribution, see next graph with different b (beta) shape values.
Want to know more? Please comment what topics you are interested, in next chapter will cover the bathtub curve, widely used in the life evaluation of any product.
Leave a Reply