On the Performance Assessment of Mean Methods in Estimating Process Capability for Skewed Distributions

This study compares the performances of Gini Mean, Clementsand Box-Cox transformation methods for estimating processcapability Indices when the distribution of the process data is(skewed) non-normal. The use of process performance index(PPI) is implored for process capability analysis (PCA) usingWeibull distribution. Data was simulated using R software witha decision interval (target point) of 1.0 and 1.5. Performanceassessment was carried out using Boxplots, descriptive statisticsand the root mean square deviation. It is observed that Ginimean dierence based process capability indices performs best inestimating the process capability indices closest to a set target forvarying distribution parameters at dierent sample sizes, followedby Clements and lastly, the Box-Cox transformation method.