Stat 567:Statistical Reliability
Accelerated Life Model Check Based on Quantiles
Using JMP Statistical Software


Plots of Log quantiles at each load versus the log of the average of the quantiles should be parallel lines with slope one.

To illustrate what we mean we use the data of Watson, A. S. and Smith, R. L. (1985) on the failure stresses of single carbon fiber. (Table 4.1, Page 81 of your textbook). Notice that in this example failure stress plays the role of time.

Open the buttons 1, 10, 20, 50. For example below we have the 10 button:

From these tables we can built by inspection the following JMP table:

For each Fraction, the columns Quatile1, Quatile10, Quatile20, Quatile50 are the quantiles at the different lengths (loads). The last column, Ave. Quantile, was obtained using JMP's calculator and is the average of the four quantiles in that row.

Notice that we have selected a log scale for both X and Y.



If the assumption of an Accerated Life Model are satisfied these lines should be parallel with slope 1. What do you think?

SAS JMP files (Mac) of classroom examples and homework


Do you have something to tell me?