Modeling multivariate profiles using Gaussian process-controlled B-splines

Abstract Due to the increasing presence of profile data in manufacturing, profile monitoring has become one of the most popular research directions in statistical process control. The core of profile monitoring is how to model the profile data. Most of the current methods deal with univariate profile modeling where only within-profile correlation is considered. In this article, a linear mixed-effect model framework is adopted for dealing with multivariate profiles, having both within- and between-profile correlations. For better flexibility yet reduced computational cost, we propose to construct the random component of the linear mixed effects model using B-splines, whose control points are governed by a multivariate Gaussian process. Extensive simulations have been conducted to compare the model with classic models. In the case study, the proposed model is applied to the transmittance profiles from the low-emittance glasses.

[1]  C. F. Sirmans,et al.  Nonstationary multivariate process modeling through spatially varying coregionalization , 2004 .

[2]  Yongxiang Li,et al.  Pairwise Meta-Modeling of Multivariate Output Computer Models Using Nonseparable Covariance Function , 2016, Technometrics.

[3]  Shiyu Zhou,et al.  A Simple Approach to Emulation for Computer Models With Qualitative and Quantitative Factors , 2011, Technometrics.

[4]  Ran Jin,et al.  Wafer quality monitoring using spatial Dirichlet process based mixed-effect profile modeling scheme , 2018, Journal of Manufacturing Systems.

[5]  D. Schoenfeld,et al.  Using a multi-level B-spline model to analyze and compare patient glucose profiles based on continuous monitoring data. , 2011, Diabetes technology & therapeutics.

[6]  Douglas W. Nychka,et al.  Nonstationary modeling for multivariate spatial processes , 2012, J. Multivar. Anal..

[7]  Amirhossein Amiri,et al.  A case study on monitoring polynomial profiles in the automotive industry , 2009, Qual. Reliab. Eng. Int..

[8]  Chong Gu,et al.  Optimal smoothing in nonparametric mixed-effect models , 2005 .

[9]  Li Zeng,et al.  Pairwise Estimation of Multivariate Gaussian Process Models With Replicated Observations: Application to Multivariate Profile Monitoring , 2018, Technometrics.

[10]  Peihua Qiu,et al.  Nonparametric Profile Monitoring by Mixed Effects Modeling , 2010, Technometrics.

[11]  Alan E. Gelfand,et al.  Multivariate Spatial Modeling for Geostatistical Data Using Convolved Covariance Functions , 2007 .

[12]  Rassoul Noorossana,et al.  Statistical Analysis of Profile Monitoring , 2011 .

[13]  Willis A. Jensen,et al.  Monitoring Correlation Within Linear Profiles Using Mixed Models , 2008 .

[14]  Nan Chen,et al.  Bayesian hierarchical modeling for monitoring optical profiles in low-E glass manufacturing processes , 2015 .

[15]  Neil D. Lawrence,et al.  Sparse Convolved Gaussian Processes for Multi-output Regression , 2008, NIPS.

[16]  B. Silverman,et al.  Some Aspects of the Spline Smoothing Approach to Non‐Parametric Regression Curve Fitting , 1985 .

[17]  Ying-Chao Hung,et al.  A framework for nonparametric profile monitoring , 2013, Comput. Ind. Eng..

[18]  Josef Hoschek,et al.  Handbook of Computer Aided Geometric Design , 2002 .

[19]  Abdel-Salam G. Abdel-Salam,et al.  A wavelet approach for profile monitoring of Poisson distribution with application , 2019, Commun. Stat. Simul. Comput..

[20]  D. Higdon.  Space and Space-Time Modeling using Process Convolutions , 2002 .

[21]  Debashis Paul,et al.  A GENERALIZED CONVOLUTION MODEL FOR MULTIVARIATE NONSTATIONARY SPATIAL PROCESSES , 2010 .

[22]  Gerald Farin,et al.  A History of Curves and Surfaces in CAGD , 2002, Handbook of Computer Aided Geometric Design.

[23]  Rui Meng,et al.  Nonstationary multivariate Gaussian processes for electronic health records , 2019, J. Biomed. Informatics.

[24]  N. Cressie,et al.  Fixed rank kriging for very large spatial data sets , 2008 .

[25]  Hsin-Cheng Huang,et al.  Non-stationary Multivariate Spatial Covariance Estimation via Low-Rank Regularization , 2014 .

[26]  T. Gneiting,et al.  Matérn Cross-Covariance Functions for Multivariate Random Fields , 2010 .

[27]  William H. Woodall,et al.  A review and analysis of cause-selecting control charts , 1993 .

[28]  Chung-I Li,et al.  Linear profiles monitoring in the presence of nonnormal random errors , 2019, Qual. Reliab. Eng. Int..

[29]  William H. Woodall,et al.  Current research on profile monitoring , 2007 .

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

[31]  B. Ripley,et al.  Semiparametric Regression: Preface , 2003 .

[32]  Rassoul Noorossana,et al.  Statistical Analysis of Profile Monitoring: Noorossana/Profile Monitoring , 2011 .

[33]  J. Friedman.  Multivariate adaptive regression splines , 1990 .

[34]  Rui Li,et al.  Multivariate Sparse Coding of Nonstationary Covariances with Gaussian Processes , 2019, NeurIPS.

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

[36]  J. Friedman,et al.  FLEXIBLE PARSIMONIOUS SMOOTHING AND ADDITIVE MODELING , 1989 .

[37]  Jeh-Nan Pan,et al.  Detecting the process changes for multivariate nonlinear profile data , 2019, Qual. Reliab. Eng. Int..

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

[39]  Kaibo Wang,et al.  Monitoring wafers’ geometric quality using an additive Gaussian process model , 2016 .

[40]  Rui Li,et al.  Multivariate Sparse Coding of Nonstationary Covariances with Gaussian Processes , 2019 .

[41]  Jeremy E. Oakley,et al.  Multivariate Gaussian Process Emulators With Nonseparable Covariance Structures , 2013, Technometrics.

[42]  Nan Chen,et al.  Modeling tunnel profile using Gaussian process , 2015, 2015 IEEE International Conference on Industrial Engineering and Engineering Management (IEEM).

[43]  Shing I. Chang,et al.  Statistical process control for monitoring non-linear profiles using wavelet filtering and B-Spline approximation , 2010 .