Effect of autocorrelation estimators on the performance of the X̄ control chart

ABSTRACT Control charts are powerful Statistical Process Monitoring tools to detect departures from in-control situations. However, their power detection relies on the fact that all assumptions underlying their design are met, such as independence of data and knowledge of the process model parameters. When parameters are estimated, the average and the standard deviation of the ARL (AARL and SDARL, respectively) are used as performance measures as they summarize the variation due to the Phase I estimations. Considering these performance measures, the effect of several autocorrelation estimators on the chart performance was investigated in case of stationary processes. Further, a bootstrapping technique was developed to adjust the corresponding control limits and obtain a guaranteed ARL performance. The effect on the out-of-control ARL due to this adjustment is also presented. Results show that overestimation of the autoregressive parameter leads to higher values of both in-control and out-of-control ARL's.

[1]  Philippe Castagliola,et al.  Synthetic Phase II Shewhart‐type Attributes Control Charts When Process Parameters are Estimated , 2014, Qual. Reliab. Eng. Int..

[2]  Philippe Castagliola,et al.  A median run length-based double-sampling X¯$$ \overline{X} $$ chart with estimated parameters for minimizing the average sample size , 2015 .

[3]  Amitava Mukherjee,et al.  Distribution-free exponentially weighted moving average control charts for monitoring unknown location , 2012, Comput. Stat. Data Anal..

[4]  M. H. Quenouille.  NOTES ON BIAS IN ESTIMATION , 1956 .

[5]  Giovanni Celano,et al.  THE EXACT RUN LENGTH DISTRIBUTION AND DESIGN OF THE S2 CHART WHEN THE IN-CONTROL VARIANCE IS ESTIMATED , 2009 .

[6]  Philippe Castagliola,et al.  DESIGN OF THE c AND np CHARTS WHEN THE PARAMETERS ARE ESTIMATED , 2012 .

[7]  William H. Woodall,et al.  The Difficulty in Designing Shewhart X̄ and X Control Charts with Estimated Parameters , 2015 .

[8]  Daniel W. Apley,et al.  Performance and Robustness of Control Charting Methods for Autocorrelated Data , 2008 .

[9]  William H. Woodall,et al.  Another Look at the EWMA Control Chart with Estimated Parameters , 2015 .

[10]  S. Psarakis,et al.  EFFECT OF ESTIMATION OF THE PROCESS PARAMETERS ON THE CONTROL LIMITS OF THE UNIVARIATE CONTROL CHARTS FOR PROCESS DISPERSION , 2002 .

[11]  Charles W. Champ,et al.  The Performance of Exponentially Weighted Moving Average Charts With Estimated Parameters , 2001, Technometrics.

[12]  Maurice G. Kendall,et al.  NOTE ON BIAS IN THE ESTIMATION OF AUTOCORRELATION , 1954 .

[13]  Subhabrata Chakraborti,et al.  Properties and performance of the c-chart for attributes data , 2008 .

[14]  Daniel W. Apley,et al.  Improved design of robust exponentially weighted moving average control charts for autocorrelated processes , 2011, Qual. Reliab. Eng. Int..

[15]  William H. Woodall,et al.  CUSUM charts with controlled conditional performance under estimated parameters , 2016 .

[16]  William H. Woodall,et al.  Guaranteed conditional performance of the S2 control chart with estimated parameters , 2015 .

[17]  W. John Braun.  Run length distributions for estimated attributes charts , 1999 .

[18]  Suojin Wang.  Tail probability approximations in the first-order noncircular autoregression , 1992 .

[19]  Eugenio K. Epprecht,et al.  Effect of the Amount of Phase I Data on the Phase II Performance of S2 and S Control Charts , 2015 .

[20]  Mahmoud A. Mahmoud,et al.  The Performance of the Multivariate Adaptive Exponentially Weighted Moving Average Control Chart with Estimated Parameters , 2016, Qual. Reliab. Eng. Int..

[21]  Philippe Castagliola,et al.  Economic‐Statistical Design of the Synthetic X¯ Chart with Estimated Process Parameters , 2015, Qual. Reliab. Eng. Int..

[22]  Philippe Castagliola,et al.  Some Recent Developments on the Effects of Parameter Estimation on Control Charts , 2014, Qual. Reliab. Eng. Int..

[23]  William H. Woodall,et al.  The Conditional In-Control Performance of Self-Starting Control Charts , 2015 .

[24]  Amitava Mukherjee,et al.  Design and implementation of CUSUM exceedance control charts for unknown location , 2014 .

[25]  Mukhtar M. Ali.  DISTRIBUTION OF THE LEAST SQUARES ESTIMATOR IN A FIRST-ORDER AUTOREGRESSIVE MODEL , 2002 .

[26]  S. Provost,et al.  Inference About the First-Order Autoregressive Coefficient , 2005 .

[27]  Layth C. Alwan,et al.  TIME-SERIES INVESTIGATION OF SUBSAMPLE MEAN CHARTS , 1992 .

[28]  William H. Woodall,et al.  A Reevaluation of the Adaptive Exponentially Weighted Moving Average Control Chart When Parameters are Estimated , 2015, Qual. Reliab. Eng. Int..

[29]  F. H. C. Marriott,et al.  BIAS IN THE ESTIMATION OF AUTOCORRELATIONS , 1954 .

[30]  On robust estimation in the first order autoregressive processes , 2000 .

[31]  Thaung Lwin.  Parameter estimation in first-order autoregressive model for statistical process monitoring in the p , 2011 .

[32]  P. Young,et al.  Time series analysis, forecasting and control , 1972, IEEE Transactions on Automatic Control.

[33]  Subha Chakraborti.  Parameter estimation and design considerations in prospective applications of the X¯ chart , 2006 .

[34]  M. A. Graham,et al.  Phase I Statistical Process Control Charts: An Overview and Some Results , 2008 .

[35]  Philippe Castagliola,et al.  Effect of measurement error and autocorrelation on the X¯ chart , 2011 .

[36]  Axel Gandy,et al.  Guaranteed Conditional Performance of Control Charts via Bootstrap Methods , 2011, 1111.4180.

[37]  L. Shenton,et al.  Moments of a Serial Correlation Coefficient , 1965 .

[38]  Min Zhang,et al.  Exponential CUSUM Charts with Estimated Control Limits , 2014, Qual. Reliab. Eng. Int..

[39]  Stefan H Steiner,et al.  Assessing the effect of estimation error on risk-adjusted CUSUM chart performance. , 2012, International journal for quality in health care : journal of the International Society for Quality in Health Care.

[40]  Ronald J. M. M. Does,et al.  Guaranteed In-Control Performance for the Shewhart X and X Control Charts , 2017 .

[41]  Chris Chatfield,et al.  Introduction to Statistical Time Series. , 1976 .

[42]  Subha Chakraborti,et al.  Parameter Estimation and Performance of the$p$-Chart for Attributes Data , 2006, IEEE Transactions on Reliability.

[43]  Marian Z. Stoykov,et al.  Least Squares Bias in Time Series with Moderate Deviations from a Unit Root , 2018, Journal of Time Series Analysis.

[44]  Douglas M. Hawkins,et al.  The CUSUM and the EWMA Head-to-Head , 2014 .

[45]  P. Maravelakis,et al.  A CUSUM control chart for monitoring the variance when parameters are estimated , 2011 .

[46]  William H. Woodall,et al.  A head-to-head comparative study of the conditional performance of control charts based on estimated parameters , 2017 .

[47]  Wolfgang Schmid,et al.  The influence of parameter estimation on the ARL of Shewhart type charts for time series , 2000 .

[48]  Charles W. Champ,et al.  Effects of Parameter Estimation on Control Chart Properties: A Literature Review , 2006 .

[49]  Philippe Castagliola,et al.  Optimal design of the double sampling X chart with estimated parameters based on median run length , 2014, Comput. Ind. Eng..

[50]  Dan Trietsch,et al.  The Rate of False Signals in Ū Control Charts with Estimated Limits , 2007 .

[51]  Albert K. Tsui,et al.  Exact distributions, density functions and moments of the least squares estimator in a first-order autoregressive model , 1994 .