CSPC: A Monitoring Procedure for State Dependent Processes

Most Statistical Process Control (SPC) methods are not suitable for monitoring non-linear and state-dependent processes. This paper introduces the Context-based SPC (CSPC) methodology for state-dependent data generated by a finite-memory source. The key idea of the CSPC is to monitor the statistical attributes of a process by comparing two context trees at any monitoring period of time. The first is a reference tree that represents the 'in control' reference behaviour of the process; the second is a monitored tree, generated periodically from a sample of sequenced observations, that represents the behaviour of the process at that period. The Kullback-Leibler (KL) statistic is used to measure the relative 'distance' between these two trees, and an analytic distribution of this statistic is derived. Monitoring the KL statistic indicates whether there has been any significant change in the process that requires intervention. An example of buffer-level monitoring in a production system demonstrates the viability of the new method with respect to conventional methods.

[1]  Meir Feder,et al.  A universal finite memory source , 1995, IEEE Trans. Inf. Theory.

[2]  George C. Runger,et al.  Batch-means control charts for autocorrelated data , 1996 .

[3]  Marion R. Reynolds,et al.  EWMA CONTROL CHARTS FOR MONITORING THE MEAN OF AUTOCORRELATED PROCESSES , 1999 .

[4]  Jeffrey F. Bromaghin,et al.  Sample Size Determination for Interval Estimation of Multinomial Probabilities , 1993 .

[5]  T. Harris,et al.  Statistical process control procedures for correlated observations , 1991 .

[6]  J. A. Buzacott,et al.  Flexible manufacturing systems: a review of analytical models , 1986 .

[7]  Solomon Kullback,et al.  Information Theory and Statistics , 1960 .

[8]  Nien Fan Zhang,et al.  A statistical control chart for stationary process data , 1998 .

[9]  Douglas C. Montgomery,et al.  Some Statistical Process Control Methods for Autocorrelated Data , 1991 .

[10]  Jorma Rissanen.  Fast Universal Coding With Context Models , 1999, IEEE Trans. Inf. Theory.

[11]  H Niemann,et al.  Identification and analysis of eukaryotic promoters: recent computational approaches. , 2001, Trends in genetics : TIG.

[12]  David D. Yao,et al.  On queueing network models of flexible manufacturing systems , 1986, Queueing Syst. Theory Appl..

[13]  George C. Runger,et al.  Model-Based and Model-Free Control of Autocorrelated Processes , 1995 .

[14]  S. Kullback,et al.  Information Theory and Statistics , 1959 .

[15]  Haim Shore,et al.  General control charts for attributes , 2000 .

[16]  JORMA RISSANEN,et al.  A universal data compression system , 1983, IEEE Trans. Inf. Theory.

[17]  Nader Ebrahimi,et al.  Information theoretic framework for process control , 1998, Eur. J. Oper. Res..

[18]  Herbert Moskowitz,et al.  Run-Length Distributions of Special-Cause Control Charts for Correlated Processes , 1994 .

[19]  Jianjun Shi,et al.  The GLRT for statistical process control of autocorrelated processes , 1999 .

[20]  Gabriel R. Bitran,et al.  A review of open queueing network models of manufacturing systems , 1992, Queueing Syst. Theory Appl..

[21]  W. D. Johnson,et al.  Properties of simultaneous confidence intervals for multinomial proportions , 1997 .

[22]  Marion R. Reynolds,et al.  Control charts for monitoring processes with autocorrelated data , 1997 .

[23]  Irad Ben-Gal,,et al.  Design of Control and Monitoring Rules for State Dependent Processes , 2000 .

[24]  Stanley B. Gershwin,et al.  Manufacturing Systems Engineering , 1993 .

[25]  George C. Runger,et al.  Average run lengths for cusum control charts applied to residuals , 1995 .

[26]  Layth C. Alwan,et al.  Time-Series Modeling for Statistical Process Control , 1988 .