Distribution for Warranty Prediction

The parameter estimates for the distributions are as follows:

Scale: Exponential distribution
Shape and scale: Weibull distribution
Location and scale: Smallest extreme value, normal, lognormal, logistic, and loglogistic distributions
Threshold: 3-parameter Weibull, 2-parameter exponential, 3-parameter lognormal, 3-parameter loglogistic distributions

The parameter estimates define the best-fitting parameter estimates for the distribution that you select. All other parametric distribution analysis graphs and statistics are based on the distribution. Therefore, to obtain accurate estimates, the distribution that you select for the analysis must adequately fit the data.

Example output

Distribution Parameters

Distribution	Shape	Scale
Weibull	1.26494	398.062

Interpretation

For the compressor data, the engineer selected a Weibull distribution to model the data. The shape and scale parameters that define the best fitting Weibull distribution using the maximum likelihood method are shape = 1.26494 and scale = 398.062. Minitab uses this distribution to predict future failures.