Guaranteed conditional ARL performance in the presence of autocorrelation

Abstract The Average Run Length (ARL) performance of a control chart conditional on parameter estimates from a Phase I study may deviate significantly from its designed performance. To circumvent this problem, the guaranteed conditional performance measure is proposed in the literature. Hence, one does not intend to reach a specified in-control ARL exactly, but to reach an ARL value greater than or equal to a specified in-control ARL with a certain probability. In the literature, the control chart design for a guaranteed conditional performance has only been discussed for the case of independent and identically distributed data. Therefore, a novel guaranteed conditional performance approach for an autocorrelated data generating process is proposed, assuming that the process parameters are unknown. The approach is exemplified by considering an AutoRegressive dependence structure of order 1, i.e., AR(1), and by designing individuals control chart. Appropriate bootstrap schemes are provided for implementation. These bootstrap schemes extend easily from order one to general order  p for an AR( p ) process. Through simulations, performance analyses of the proposed approach and a study about the robustness with respect to distributional assumptions are provided.

[1]  Raffaella Giacomini,et al.  A WARP-SPEED METHOD FOR CONDUCTING MONTE CARLO EXPERIMENTS INVOLVING BOOTSTRAP ESTIMATORS , 2013, Econometric Theory.

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

[3]  Christian H. Weiß,et al.  Evaluation of Phase I analysis scenarios on Phase II performance of control charts for autocorrelated observations , 2016 .

[4]  Efstathios Paparoditis,et al.  Bootstrap methods for dependent data: A review , 2011 .

[5]  S. Lahiri,et al.  1 - Bootstrap Methods for Time Series , 2012 .

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

[7]  Ronald J. M. M. Does,et al.  On guaranteed in-control performance for the Shewhart X and control charts , 2018 .

[8]  S. Lahiri.  Resampling Methods for Dependent Data , 2003 .

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

[10]  Douglas C. Montgomery,et al.  Introduction to Statistical Quality Control , 1986 .

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

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

[13]  Wilbert C.M. Kallenberg,et al.  Are estimated control charts in control? , 2001 .

[14]  Wolfgang Schmid,et al.  On the run length of a Shewhart chart for correlated data , 1995 .

[15]  G. Masarotto,et al.  Bootstrap-based design of residual control charts , 2009 .

[16]  Helmut Ltkepohl,et al.  New Introduction to Multiple Time Series Analysis , 2007 .

[17]  G. Geoffrey Vining,et al.  A tutorial on an iterative approach for generating Shewhart control limits , 2016 .

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

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

[20]  Erniel B. Barrios,et al.  Control Chart for Monitoring Autocorrelated Process with Multiple Exogenous Inputs , 2016, Commun. Stat. Simul. Comput..

[21]  P. Bühlmann.  Sieve bootstrap for time series , 1997 .

[22]  D. Freedman.  On Bootstrapping Two-Stage Least-Squares Estimates in Stationary Linear Models , 1984 .

[23]  Sven Knoth,et al.  Control Charts for Time Series: A Review , 2004 .

[24]  Layth C. Alwan.  Effects of autocorrelation on control chart performance , 1992 .

[25]  Stefan H. Steiner,et al.  An Overview of Phase I Analysis for Process Improvement and Monitoring , 2014 .

[26]  Jens-Peter Kreiss,et al.  Baxter's inequality and sieve bootstrap for random fields , 2017 .

[27]  Daniel W. Apley,et al.  Design of Exponentially Weighted Moving Average Control Charts for Autocorrelated Processes With Model Uncertainty , 2003, Technometrics.

[28]  Runze Li,et al.  NEW EFFICIENT ESTIMATION AND VARIABLE SELECTION METHODS FOR SEMIPARAMETRIC VARYING-COEFFICIENT PARTIALLY LINEAR MODELS. , 2011, Annals of statistics.

[29]  D. Politis,et al.  Valid Resampling of Higher-Order Statistics Using the Linear Process Bootstrap and Autoregressive Sieve Bootstrap , 2013 .

[30]  Giovanna Capizzi,et al.  Efficient control chart calibration by simulated stochastic approximation , 2016 .

[31]  Stelios Psarakis,et al.  SPC Procedures for Monitoring Autocorrelated Processes , 2007 .

[32]  Murat Caner Testik,et al.  Model Inadequacy and Residuals Control Charts for Autocorrelated Processes , 2005 .

[33]  Russell A. Boyles,et al.  Phase I Analysis for Autocorrelated Processes , 2000 .

[34]  Erniel B. Barrios,et al.  An AR‐Sieve Bootstrap Control Chart for Autocorrelated Process Data , 2012, Qual. Reliab. Eng. Int..

[35]  Giovanna Capizzi,et al.  Practical Design of Generalized Likelihood Ratio Control Charts for Autocorrelated Data , 2008, Technometrics.