A New Approach to Production Process Capability Assessment for Non-Normal Data

The process quality capability indicators Cp and Cpk are widely used to measure capability. Traditional metric estimation methods require data be explicit normally distributed. Often, the actual obtained from production regarding measurements of features incomplete do not have a normal distribution. This means that use traditional estimating may lead erroneous results. Moreover, in case qualitative characteristics where two-sided tolerance limit is specified, it should very difficult. problem arises when meet postulate normality distribution and/or one-sided has been defined for process. Therefore, purpose this article was present possibility using Six Sigma method relation numerical paper proposes power transformation multiple-criteria decision analysis (MCDA) asymmetry coefficient kurtosis coefficient. task minimize Jarque–Bera statistic, which we test An appropriate methodology developed presented on an empirical example. In addition, variable after transformation, determined, selected evaluation indices were calculated.