ON THE BOOTSTRAP MULTIVARIATE EXPONENTIALLY WEIGHTED MOVING AVERAGE (BMEWMA) IN SETTING CONTROL LIMITS AND P-VALUES FOR INTERPRETING OUT OF CONTROL SIGNALS Page No: 4233-4243

By: Ikpotokin O and Ishiekwene C C

Keywords: Quality characteristics, BMEWMA control chart, out-of-control signals, control limits, p-values

Abstract: The effects of control limits that are too narrow increase the rate of false alarms, while those that are too wide may not be able to identify special causes of variability in any given process. It is of this view that control chart methodology that can detect small to moderate shifts in the mean vector should be developed so that the probability of detecting or not detecting false alarm rate should be minimized. The bootstrap multivariate exponentially weighted moving average is proposed in setting control limits, while p-value method was introduced to identify out of control signals.



[View Complete Article]