Role of neural networks and genetic algorithms in developing intelligent quality controllers for on-line parameter design

Parameter design methods, in general, do not take into account the common occurrence that some of the uncontrollable factors are observable for products and processes, during operation and production, respectively. This paper introduces a methodology that facilitates on-line parameter design for products and processes utilizing the extra information available about observable uncontrollable factors. Implementation of the proposed methodology leads to a quality controller that operates in two distinct modes: identification mode and on-line parameter design mode, identification mode involves establishing a model that relates quality response characteristics with significant controllable and uncontrollable variables. On-line parameter design mode involves optimization of the controllable variables with respect to desired levels of output quality parameters, with consideration to levels of the observable uncontrollable variables. A plasma etching semiconductor manufacturing process is used as a testbed for the proposed intelligent quality controllers. Results reveal that the proposed quality controllers can be used for on-line parameter design of manufacturing processes. Results also reveal that significant improvements in quality (measured in terms of average deviation of process outputs from target) over off-line parameter design approaches are to be expected in production processes with some level of control on uncontrollable variables. Even in the absence of any control on uncontrollable variables, the proposed controllers always perform better than traditional off-line robust parameter design techniques; however, the improvements may not be significant.

[1]  Michael I. Jordan,et al.  Advances in Neural Information Processing Systems 30 , 1995 .

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

[3]  C. Spanos,et al.  Statistical experimental design in plasma etch modeling , 1991 .

[4]  Richard P. Lippmann,et al.  An introduction to computing with neural nets , 1987 .

[5]  D. Manos,et al.  Plasma etching : an introduction , 1989 .

[6]  C. Ferrar.  Glow discharge processes; Sputtering and plasma etching , 1981, IEEE Journal of Quantum Electronics.

[7]  S. Hyakin,et al.  Neural Networks: A Comprehensive Foundation , 1994 .

[8]  Esther Levin,et al.  A statistical approach to learning and generalization in layered neural networks , 1989, Proc. IEEE.

[9]  Megan Pledger.  Observable Uncontrollable Factors in Parameter Design , 1996 .

[10]  A. Lichtenberg,et al.  Principles of Plasma Discharges and Materials Processing , 1994 .

[11]  John H. Holland,et al.  Adaptation in Natural and Artificial Systems: An Introductory Analysis with Applications to Biology, Control, and Artificial Intelligence , 1992 .

[12]  Sidney Addelman,et al.  trans-Dimethanolbis(1,1,1-trifluoro-5,5-dimethylhexane-2,4-dionato)zinc(II) , 2008, Acta crystallographica. Section E, Structure reports online.

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

[14]  D. Montgomery.  USING FRACTIONAL FACTORIAL DESIGNS FOR ROBUST PROCESS DEVELOPMENT , 1990 .

[15]  A. Linden,et al.  Inversion of multilayer nets , 1989, International 1989 Joint Conference on Neural Networks.

[16]  R. N. Kackar.  Off-Line Quality Control, Parameter Design, and the Taguchi Method , 1985 .

[17]  David E. Goldberg,et al.  Genetic Algorithms in Search Optimization and Machine Learning , 1988 .

[18]  P. E. Riley,et al.  Study of etch rate characteristics of SF/sub 6//He plasmas by response-surface methodology: effects of interelectrode spacing , 1989 .

[19]  Gary S. May,et al.  An optimal neural network process model for plasma etching , 1994 .

[20]  Genichi Taguchi.  System Of Experimental Design: Engineering Methods To Optimize Quality And Minimize Costs , 1987 .

[21]  Casimir C. Klimasauskas,et al.  Dynamic neural control for a plasma etch process , 1997, IEEE Trans. Neural Networks.

[22]  R. H. Myers,et al.  Response Surface Alternatives to the Taguchi Robust Parameter Design Approach , 1992 .

[23]  G. Geoffrey Vining,et al.  Combining Taguchi and Response Surface Philosophies: A Dual Response Approach , 1990 .

[24]  Gunar E. Liepins,et al.  Genetic algorithms: Foundations and applications , 1990 .

[25]  G. Geoffrey Vining,et al.  Taguchi's parameter design: a panel discussion , 1992 .

[26]  Jacek M. Zurada,et al.  Introduction to artificial neural systems , 1992 .

[27]  Sholom M. Weiss,et al.  Computer Systems That Learn , 1990 .

[28]  Kurt Hornik,et al.  FEED FORWARD NETWORKS ARE UNIVERSAL APPROXIMATORS , 1989 .

[29]  Edward A. Rietman,et al.  Use of neural networks in modeling semiconductor manufacturing processes: an example for plasma etch modeling , 1993 .

[30]  Alice M. Agogino,et al.  Use of influence diagrams and neural networks in modeling semiconductor manufacturing processes , 1991 .

[31]  Anne C. Shoemaker,et al.  Robust design: A cost-effective method for improving manufacturing processes , 1986, AT&T Technical Journal.

[32]  Ravindra Khattree,et al.  ROBUST PARAMETER DESIGN : A RESPONSE SURFACE APPROACH , 1996 .

[33]  David Haussler,et al.  What Size Net Gives Valid Generalization? , 1989, Neural Computation.