IFAC Workshop on
Advanced Process Control for Semiconductor Manufacturing






Paper Submission

Technical Program






Keynote 1:

300mm e-Manufacturing


Mr KC Ang
Senior Vice President, Fabs Operations
Chartered Semiconductor Manufacturing Limited



Worldwide semiconductor sector is growing and the productís time to market is becoming very significant. Introduction of Nanotechnology and 300mm post a very big challenge to wafer Fab in meeting the ramp to yield and ramp to volume production of the product. This presentation describes the 300mm e-Manufacturing vision and itís journey. It describes the automation elements of AMHS and its interactions with yield and manufacturing system, together with the information flows, in realizing the e-Manufacturing. There is a need in constant improvements & changes in making the interaction among the process equipment, AMHS, application software in ensuring the success of the 300mm Fab, to meet Fab short cycle time, better yield and higher productivity.




KC Ang has served as Senior Vice President, Fab Operations for Chartered Semiconductor Manufacturing Limited since September 2002 and he is responsible for manufacturing strategy and operational excellence across all of Chartered's fabs. Prior to this, Mr. Ang worked 4 years in Silterra and 9 years in Chartered in the area of manufacturing operations. Mr. Ang holds a Master of Science degree in Engineering from the University of Texas.


Keynote 2:

Run to Run Control, Sampling, and Performance Monitoring
for High Mix Fabs


Professor Thomas F. Edgar
Department of Chemical Engineering
University of Texas, Austin, TX 78712



In an ASIC fab or foundry, there are a great many different products, and the mix of products is constantly changing.  The high cost of process equipment drives manufacturers to maximize the use of their tools and minimize idle time, leaving little room for dedication of tools to specific product process streams.  Therefore, one lot of a specific product may take a very different processing path through the fab than the next lot of that same product.

Variations in product quality often are functions of the product being produced as well as the manufacturing tools being used, which is termed manufacturing context.  Different products behave differently during processing due to factors such as differences in materials used, configuration or layout of devices and interconnects, feature size, and overall chip size.  To further complicate matters, seemingly identical tools may process identical wafers differently based on such conditions as the number of lots processed since the last maintenance event, small differences in tool construction, or minor variations in ambient conditions.  The necessity of addressing these sources of product quality variation within control system design is a focus of this presentation.  New algorithms have been developed to address the issues raised with multiple process/multiple products control.  One algorithm, called Just-in-time Adaptive Disturbance Estimation (JADE), is outlined and applied to several test cases designed to isolate typical types of process disturbances observed in the semiconductor industry.  JADE utilizes EWMA feedback control and least squares estimation techniques identify contributing contexts such as process tool, product, technology, reticle, etc.  Processing decisions such as batch scheduling, tool allocation, and sampling plans can influence controller performance as well as information content.  By judicious selections of process measurements and tools, the amount of information that is shared across different batches can be maximized.  The trace of the state error covariance matrix from the Kalman filter was used as a metric for determining the apparent value of a particular data set to the run-to-run control algorithm.




Thomas F. Edgar is Professor of Chemical Engineering at the University of Texas at Austin and holds the George T. and Gladys Abell Chair in Engineering.  Dr. Edgar received his B.S. degree in chemical engineering from the University of Kansas and a Ph.D. from Princeton University.  For the past 35 years, he has concentrated his academic work in process modeling, control, and optimization, with over 200 articles and book chapters.  Edgar has co-authored leading textbooks:  Optimization of Chemical Processes (McGraw-Hill, 2001) and Process Dynamics and Control (Wiley, 2004).  He has received major awards from AIChE (Colburn, Computing in Chemical Engineering, Lewis) and ASEE (Chemical Engineering Division, Westinghouse, and Meriam-Wiley).  Recently he has carried out modeling and control research projects jointly with AMD, Motorola, Texas Instruments, Yield Dynamics, Tokyo Electron and SEMATECH, involving 15 Ph.D. students who now work in the microelectronics industry..