The computation of accuracy of quality parameters by means of a monte carlo simulation

Abstract

The bootstrap methodology is applied to control quality of an industrial production according to the specifications required by th costumer. A comparison between the approach based on the capability index and normality assumptions, and the bootstrap approach is given using real and simulated data. The procedure allows to estimate tail probabilities even in regions not supported by data, with an accuracy independent of the sample variance also when data are no “nearly” normal.

Keywords:

• bootstrap
• capability index
• statistical process control