A Data Mining Algorithm for Monitoring PCB Assembly Quality

A pattern clustering algorithm is proposed in this paper as a statistical quality control technique for diagnosing the solder paste variability when a huge number of binary inspection outputs are involved. To accommodate this goal, a latent variable model is first introduced and incorporated into classical logistic regression model so that the interdependencies between measured physical characteristics and their relationship to the final solder defects can be explained. This probabilistic model also allows a maximum-likelihood principal component analysis (MLPCA) method to recognize the dimension of systematic causes contributing to solder paste variability. The correlated measurement variables are then projected onto the reduced latent space, followed by an appropriate clustering approach over the inspected solder pastes for variation interpretation and quality diagnosing. An application to a real stencil printing process demonstrates that this method facilitates in identifying the root causes of solder paste defects and thereby improving PCB assembly yield.

[1]  Jr. R.R. Lathrop.  Solder paste print qualification using laser triangulation , 1997 .

[2]  A. Goldstein,et al.  Stencil Printing Process Modeling and Control Using Statistical Neural Networks , 2008, IEEE Transactions on Electronics Packaging Manufacturing.

[3]  Krishnaswami Srihari,et al.  A realtime process control system for solder paste stencil printing , 1997, Twenty First IEEE/CPMT International Electronics Manufacturing Technology Symposium Proceedings 1997 IEMT Symposium.

[4]  Daniel W. Apley,et al.  A Factor-Analysis Method for Diagnosing Variability in Mulitvariate Manufacturing Processes , 2001, Technometrics.

[5]  Gregg R. Yost.  Acquiring knowledge in Soar , 1993, IEEE Expert.

[6]  D. Ruppert.  The Elements of Statistical Learning: Data Mining, Inference, and Prediction , 2004 .

[7]  Magnus Egerstedt,et al.  Process control in a high-noise environment using a limited number of measurements , 2003, Proceedings of the 2003 American Control Conference, 2003..

[8]  Andrew J. Stoddart,et al.  Using PCA to model shape for process control , 1997, Proceedings. International Conference on Recent Advances in 3-D Digital Imaging and Modeling (Cat. No.97TB100134).

[9]  J. Booth,et al.  Maximizing generalized linear mixed model likelihoods with an automated Monte Carlo EM algorithm , 1999 .

[10]  C. Neubauer,et al.  Intelligent X-ray inspection for quality control of solder joints , 1997 .

[11]  C. McCulloch.  Maximum Likelihood Algorithms for Generalized Linear Mixed Models , 1997 .

[12]  C. Weinberg,et al.  Modeling conception as an aggregated Bernoulli outcome with latent variables via the EM algorithm. , 1996, Biometrics.

[13]  Robert Tibshirani,et al.  The Elements of Statistical Learning: Data Mining, Inference, and Prediction, 2nd Edition , 2001, Springer Series in Statistics.

[14]  K. Srihari,et al.  Process development for ball grid array assembly using a design of experiments approach , 1999 .

[15]  Christopher M. Bishop,et al.  Mixtures of Probabilistic Principal Component Analyzers , 1999, Neural Computation.

[16]  J. Nelder,et al.  Generalized Linear Models for Quality-Improvement Experiments , 1997 .

[17]  R. Storer,et al.  Critical variables of solder paste stencil printing for micro-BGA and fine-pitch QFP , 2004, IEEE Transactions on Electronics Packaging Manufacturing.

[18]  Scott MacKinnon,et al.  Statistical methods for visual defect metrology , 1998 .

[19]  Michael I. Jordan,et al.  A Variational Approach to Bayesian Logistic Regression Models and their Extensions , 1997, AISTATS.

[20]  Stan Lipovetsky,et al.  Latent Variable Models and Factor Analysis , 2001, Technometrics.

[21]  Andrzej J. Strojwas,et al.  Monitoring multistage integrated circuit fabrication processes , 1996 .

[22]  Sheng Liu,et al.  A novel approach for flip chip solder joint quality inspection: laser ultrasound and interferometric system , 2001 .