On efficient median control charting

Abstract

In most manufacturing processes, we face sudden outliers, and in such situations, median control charts are more outliers-resistant than mean control charts. In ideal circumstances, a median chart presents less efficiency than a mean chart. In order to overcome the efficiency loss in uncontaminated environments by maintaining its resistance ability in contaminated environments, we have suggested an auxiliary information–based set of median type control charts using the coefficient of variation and coefficient of kurtosis of auxiliary characteristics for efficient process monitoring. The performances of these control charts are evaluated in terms of average run length (ARL) and extra quadratic loss (EQL) under the univariate, bivariate, and trivariate normally distributed process environments. The effects of contaminated environments are also examined on the ARL performance of different median-based charting structures. The illustrative examples concerning median type control structures are also provided for procedural details.

Keywords:

• auxiliary information
• average run length (ARL)
• coefficient of variation
• coefficient of kurtosis
• contamination
• extra quadratic loss (EQL)
• median control charts
• trivariate normal distribution