Detecting outliers in complex profiles using a χ2 control chart method

The quality of products or manufacturing processes is sometimes characterized by profiles or functions. A method is proposed to identify outlier profiles among a set of complex profiles which are difficult to model with explicit functions. It treats profiles as vectors in high-dimension space and applies a χ2 control chart to identify outliers. This method is useful in Statistical Process Control (SPC) in two ways: (i) identifying outliers in SPC baseline data; and (ii) the on-line monitoring of profiles. The method does not require explicit expression of the function between the response and explanatory variables or fitting regression models. It is especially useful and sometimes the only option when profiles are very complex. Given a set of profiles (high-dimension vectors), the median of these vectors is derived. The variance among profiles is estimated by considering the pair-wise differences between profiles. A χ2 statistic is derived to compare each profile to the center vector. A simulation experiment and manufacturing data are used to illustrate applications of the method. Comparing it with the existing non-linear regression method shows that it has a better performance: it misidentifies fewer non-outlier profiles as outliers than the non-linear regression method, and misidentifies similarly small fractions of outlier profiles as non-outliers. [Supplementary materials are available for this article. Go to the publisher's online edition of IIE Transactions for the following free supplemental resource: Appendix]

[1]  Fred Spiring,et al.  Introduction to Statistical Quality Control , 2007, Technometrics.

[2]  Xue Z. Wang,et al.  Multidimensional visualisation for process historical data analysis: a comparative study with multivariate statistical process control , 2005 .

[3]  Jionghua Jin,et al.  Automatic feature extraction of waveform signals for in-process diagnostic performance improvement , 1998, SMC'98 Conference Proceedings. 1998 IEEE International Conference on Systems, Man, and Cybernetics (Cat. No.98CH36218).

[4]  Susan L. Albin,et al.  Determining the number of operational modes in baseline multivariate SPC data , 2007 .

[5]  J. O. Rawlings,et al.  Applied Regression Analysis , 1998 .

[6]  Lan Kang,et al.  On-Line Monitoring When the Process Yields a Linear Profile , 2000 .

[7]  Charles W. Champ.  Introduction to Statistical Quality Control, Fourth Edition , 2001 .

[8]  William H. Woodall,et al.  Phase I Analysis of Linear Profiles With Calibration Applications , 2004, Technometrics.

[9]  Douglas C. Montgomery,et al.  Using Control Charts to Monitor Process and Product Quality Profiles , 2004 .

[10]  Vipin Kumar,et al.  Finding Clusters of Different Sizes, Shapes, and Densities in Noisy, High Dimensional Data , 2003, SDM.

[11]  Jye-Chyi Lu,et al.  Wavelet-based SPC procedure for complicated functional data , 2006 .

[12]  Adrian E. Raftery,et al.  How Many Clusters? Which Clustering Method? Answers Via Model-Based Cluster Analysis , 1998, Comput. J..

[13]  Lloyd S. Nelson.  Column: Technical Aids: Exact Critical Values for the Analysis of Means , 1983 .

[14]  Nien Fan Zhang,et al.  Forecasting and time series analysis , 1976 .

[15]  L. S. Nelson.  Exact Critical Values for Use with the Analysis of Means , 1983 .

[16]  Hans-Peter Kriegel,et al.  LOF: identifying density-based local outliers , 2000, SIGMOD '00.

[17]  Mik Wisniewski,et al.  Applied Regression Analysis: A Research Tool , 1990 .

[18]  Emily K. Lada,et al.  A wavelet-based procedure for process fault detection , 2002 .

[19]  Ravi Kothari,et al.  On finding the number of clusters , 1999, Pattern Recognit. Lett..

[20]  William H. Woodall,et al.  Statistical monitoring of nonlinear product and process quality profiles , 2007, Qual. Reliab. Eng. Int..

[21]  F. S. Stover,et al.  Statistical quality control applied to ion chromatography calibrations , 1998 .

[22]  S. P. Wright,et al.  Comparing Curves Using Additive Models , 2002 .

[23]  Mahmoud A. Mahmoud,et al.  On the Monitoring of Linear Profiles , 2003 .