Neural networks for advanced process control

The fabrication of integrated circuits involves many non-linear processes, each with several inputs and outputs. These complexities suggest that benefits could be derived from the development and implementation of advanced process control tools and strategies. Empirical process models are one of these tools. In this research, sequential neural network models were developed to characterize critical steps in a fabrication process. The data used were collected from an industrial process, a distinguishing feature of this research. Typically the data used to train neural network models comes from designed (factorial) experiments [1]. Here, the data came from experiments related to the processes under investigation, but not specifically designed to generate data for modeling. The models performed well demonstrating the flexibility of the sequential neural network modeling process. Additionally, the models were used in a sensitivity analysis to study the output response to the various inputs. Future work will include using the models as part of a model-based supervisory control system.

[1]  A. Krogh.  What are artificial neural networks? , 2008, Nature Biotechnology.

[2]  Costas J. Spanos,et al.  Fundamentals of Semiconductor Manufacturing and Process Control: May/Fundamentals of Semiconductor Manufacturing and Process Control , 2006 .

[3]  M. Stone.  Cross‐Validatory Choice and Assessment of Statistical Predictions , 1976 .

[4]  Gary S. May,et al.  Modeling the properties of PECVD silicon dioxide films using optimized back-propagation neural networks , 1994 .

[5]  J V Tu,et al.  Advantages and disadvantages of using artificial neural networks versus logistic regression for predicting medical outcomes. , 1996, Journal of clinical epidemiology.

[6]  Kurt Hornik,et al.  Multilayer feedforward networks are universal approximators , 1989, Neural Networks.

[7]  Costas J. Spanos,et al.  Fundamentals of Semiconductor Manufacturing and Process Control , 2006 .

[8]  D. M. Titterington,et al.  Neural Networks: A Review from a Statistical Perspective , 1994 .

[9]  Gary S. May,et al.  Recipe synthesis for PECVD SiO/sub 2/ films using neural networks and genetic algorithms , 1996, 1996 Proceedings 46th Electronic Components and Technology Conference.

[10]  Guoqiang Peter Zhang,et al.  Neural networks for classification: a survey , 2000, IEEE Trans. Syst. Man Cybern. Part C.

[11]  Gary S. May,et al.  Advantages of plasma etch modeling using neural networks over statistical techniques , 1993 .

[12]  Elie Bienenstock,et al.  Neural Networks and the Bias/Variance Dilemma , 1992, Neural Computation.

[13]  G. S. May,et al.  Intelligent control of via formation process in MCM-L/D substrates using neural networks , 1999, Proceedings International Symposium on Advanced Packaging Materials. Processes, Properties and Interfaces (IEEE Cat. No.99TH8405).

[14]  G.S. May,et al.  Manufacturing ICs the neural way , 1994, IEEE Spectrum.

[15]  B. Irie,et al.  Capabilities of three-layered perceptrons , 1988, IEEE 1988 International Conference on Neural Networks.

[16]  R. Lippmann,et al.  An introduction to computing with neural nets , 1987, IEEE ASSP Magazine.

[17]  A. Deif.  Sensitivity analysis in linear systems , 1986 .

[18]  E. Bartlett,et al.  Determining functional relationships from trained neural networks , 1995 .

[19]  G. S. May,et al.  Sequential modeling of via formation in photosensitive dielectric materials for MCM-D applications , 1999 .

[20]  Russell Reed,et al.  Pruning algorithms-a survey , 1993, IEEE Trans. Neural Networks.