A Distribution-freeCUSUMChart for Monitoring Variability of AutocorrelatedProcesses

A Distribution-freeCUSUMChart for Monitoring Variability of AutocorrelatedProcesses

0.00 Avg rating0 Votes
Article ID: iaor201528931
Volume: 31
Issue: 6
Start Page Number: 923
End Page Number: 934
Publication Date: Oct 2015
Journal: Quality and Reliability Engineering International
Authors:
Keywords: statistics: sampling, simulation
Abstract:

We present a distribution‐free tabular cumulative sum chart for monitoring the variability of an autocorrelated process. A quantity known as the asymptotic variance parameter is employed as a measure of the variability, and a distribution‐free tabular cumulative sum chart is applied to variance estimates calculated from batches of nonoverlapping samples. The proposed chart is applicable to a stationary process with a general marginal distribution and a general autocorrelation structure. It also determines control limits analytically without trial‐and‐error simulations. The performance of the proposed chart is tested on stationary processes with both normal and nonnormal marginals with various autocorrelation structures.

Reviews

Required fields are marked *. Your email address will not be published.