Self-learning fuzzy logic system for in situ, in-process diagnostics of mass flow controller (MFC)

An improvement in the yield of better quality wafers requires an accurate control of various process variables. The control should include timely diagnosis and appropriate in-situ, in-process adjustments for drifts in these variables. One such scheme, a self-learning fuzzy logic system, is developed in this study for correcting drifts in the calibration of mass flow controllers (MFC's) that control the flow of gases into a process chamber. It consists of two components; a diagnostic system and a self-learning system. The diagnostic system uses fuzzy logic to diagnose the problem and initiate suitable remedial action, The self-learning system automatically builds the knowledge base used for diagnosis. The knowledge base is initialized using clustering principle and is tuned for better performance using a set of heuristic rules. The system is capable of learning the behavior of different types of makes and models of MFC's under various flow rates. It has been tested on two different types of MFC's under different flow rates and encouraging results have been obtained. >

[1]  C. D. Jenkins,et al.  Process optimization tweaking tool (POTT) and its application in controlling oxidation thickness , 1989 .

[2]  Alice M. Agogino,et al.  Topological framework for representing and solving probabilistic inference problems in expert systems , 1988, IEEE Trans. Syst. Man Cybern..

[3]  R. Guerrieri,et al.  Equipment Models for Process Optimization and Control Using Smart Response Surfaces , 1989 .

[4]  L. Zadeh.  The role of fuzzy logic in the management of uncertainty in expert systems , 1983 .

[5]  B. Flinchbaugh,et al.  Monitoring and diagnosis of plasma etch processes , 1988 .

[6]  R. S. Gyurcsik,et al.  A model for rapid thermal processing: achieving uniformity through lamp control , 1991 .

[7]  Max Henrion,et al.  An Experimental Comparison of Knowledge Engineering for Expert Systems and for Decision Analysis , 1987, AAAI.

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

[9]  Andrzej J. Strojwas,et al.  Statistical control of VLSI fabrication processes. A framework , 1988 .

[10]  Costas J. Spanos,et al.  Automated malfunction diagnosis of semiconductor fabrication equipment: a plasma etch application , 1993 .

[11]  Costas J. Spanos,et al.  Real time statistical process control for plasma etching , 1991, 1991 Proceedings IEEE/SEMI International Semiconductor Manufacturing Science Symposium.

[12]  T. Budge,et al.  PARSEC-process analysis with recipe support for etcher control , 1990 .

[13]  Scott M. Olmsted.  On representing and solving decision problems , 1983 .

[14]  Richard O. Duda,et al.  Pattern classification and scene analysis , 1974, A Wiley-Interscience publication.

[15]  Eric Horvitz,et al.  On the Expressiveness of Rule-based Systems for Reasoning with Uncertainty , 1987, AAAI.

[16]  K. Ramamurthi,et al.  Intelligent Real-Time Predictive Diagnostics for Cutting Tools and Supervisory Control of Machining Operations , 1993 .

[17]  C. Spanos,et al.  Statistical equipment modeling for VLSI manufacturing: an application for LPCVD , 1990 .

[18]  Ross D. Shachter.  Evaluating Influence Diagrams , 1986, Oper. Res..

[19]  Edward H. Shortliffe,et al.  A model of inexact reasoning in medicine , 1990 .

[20]  Glenn Shafer,et al.  A Mathematical Theory of Evidence , 2020, A Mathematical Theory of Evidence.

[21]  Eric Horvitz,et al.  A Framework for Comparing Alternative Formalisms for Plausible Reasoning , 1986, AAAI.

[22]  Ruey-Shan Guo,et al.  Process control system for VLSI fabrication , 1991 .

[23]  Costas J. Spanos,et al.  Continuous equipment diagnosis using evidence integration: an LPCVD application , 1991 .