Fast initial response features for EWMA control charts

The exponentially weighted moving average (EWMA) control chart became very popular during the last decade. It is characterized by simple handling and good performance. It turns out, however, that the most popular EWMA scheme with fixed-width control limits—the asymptotic control limits are taken and do not change over time—detects early changes rather slowly. For the competing CUSUM chart the so-called fast initial response (head-start) feature is developed which permits rapid response to an initial out-of-control situation. Meanwhile, in some papers similar modifications for EWMA schemes are described. We compare these approaches by using precise computation techniques, which are based on numeric quadrature rules and allow higher accuracy than earlier studies. Moreover, previous comparisons are restricted to the evaluation of the detection speed by comparing the average run lengths (ARLs), that is, the parameter of interest is constant during the whole monitoring period. Here, we consider more possible change point locations, which gives the EWMA control chart user a better insight into the scheme performance for early changes. *** DIRECT SUPPORT *** A00T7005 00004

[1]  P. Robinson,et al.  Average Run Lengths of Geometric Moving Average Charts by Numerical Methods , 1978 .

[2]  S. Crowder.  A simple method for studying run-length distribution of exponentially weighted moving average charts , 1987 .

[3]  Emmanuel Yashchin,et al.  Weighted Cumulative Sum Technique , 1989 .

[4]  Marianne Frisén,et al.  Evaluations of likelihood ratio methods for surveillance. Differences and robustness , 1999 .

[5]  S. W. Roberts,et al.  Control Chart Tests Based on Geometric Moving Averages , 2000, Technometrics.

[6]  Douglas C. Montgomery,et al.  A FAST INITIAL RESPONSE SCHEME FOR THE EXPONENTIALLY WEIGHTED MOVING AVERAGE CONTROL CHART , 1996 .

[7]  James M. Lucas,et al.  Exponentially weighted moving average control schemes: Properties and enhancements , 1990 .

[8]  William H. Woodall,et al.  The Distribution of the Run Length of One-Sided CUSUM Procedures for Continuous Random Variables , 1983 .

[9]  Roger M. Sauter,et al.  Introduction to Statistical Quality Control (2nd ed.) , 1992 .

[10]  James M. Lucas,et al.  Fast Initial Response for CUSUM Quality-Control Schemes: Give Your CUSUM A Head Start , 2000, Technometrics.

[11]  J. Stuart Hunter,et al.  The exponentially weighted moving average , 1986 .

[12]  John R. English,et al.  Modeling and analysis of EWMA control schemes with variance-adjusted control limits , 1995 .

[13]  K. Waldmann,et al.  Bounds for the Distribution of the Run Length of Geometric Moving Average Charts , 1986 .

[14]  S. Steiner.  EWMA Control Charts with Time-Varying Control Limits and Fast Initial Response , 1999 .

[15]  Sven Knoth,et al.  EWMA Schemes with Nonhomogeneous Transition Kernels , 2003 .

[16]  S. W. Roberts.  Control chart tests based on geometric moving averages , 2000 .

[17]  William H. Woodall,et al.  Alarm rates for quality control charts , 1995 .

[18]  S. Crowder.  Average Run Lengths of Exponentially Weighted Moving Average Control Charts , 1987 .

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