Semiconductor Yield Modeling Using Generalized Linear …

Semiconductor Yield Modeling Using Generalized Linear Models by

Dana Cheree Krueger

A Dissertation Presented in Partial Fulfillment of the Requirements for the Degree Doctor of Philosophy

Approved March 2011 by the Graduate Supervisory Committee:

Douglas C. Montgomery, Chair John Fowler Rong Pan Michele Pfund

ARIZONA STATE UNIVERSITY May 2011

ABSTRACT Yield is a key process performance characteristic in the capital-intensive semiconductor fabrication process. In an industry where machines cost millions of dollars and cycle times are a number of months, predicting and optimizing yield are critical to process improvement, customer satisfaction, and financial success. Semiconductor yield modeling is essential to identifying processing issues, improving quality, and meeting customer demand in the industry. However, the complicated fabrication process, the massive amount of data collected, and the number of models available make yield modeling a complex and challenging task. This work presents modeling strategies to forecast yield using generalized linear models (GLMs) based on defect metrology data. The research is divided into three main parts. First, the data integration and aggregation necessary for model building are described, and GLMs are constructed for yield forecasting. This technique yields results at both the die and the wafer levels, outperforms existing models found in the literature based on prediction errors, and identifies significant factors that can drive process improvement. This method also allows the nested structure of the process to be considered in the model, improving predictive capabilities and violating fewer assumptions. To account for the random sampling typically used in fabrication, the work is extended by using generalized linear mixed models (GLMMs) and a larger dataset to show the differences between batch-specific and populationaveraged models in this application and how they compare to GLMs. These

ii

results show some additional improvements in forecasting abilities under certain conditions and show the differences between the significant effects identified in the GLM and GLMM models. The effects of link functions and sample size are also examined at the die and wafer levels.

The third part of this research describes a methodology for integrating classification and regression trees (CART) with GLMs. This technique uses the terminal nodes identified in the classification tree to add predictors to a GLM. This method enables the model to consider important interaction terms in a simpler way than with the GLM alone, and provides valuable insight into the fabrication process through the combination of the tree structure and the statistical analysis of the GLM.

iii

DEDICATION This work is dedicated to my husband, Chad, who has encouraged me, sacrificed

with me, and loved me throughout this special season of our lives together.

iv

ACKNOWLEDGMENTS

I would first like to thank Dr. Montgomery for teaching me, mentoring me, and believing in me. Your enduring patience and gentle encouragement have been invaluable to me, both in completing this work and in my own role as a teacher and scholar. I am also thankful for the helpful contributions from my committee members Dr. Pfund, Dr. Pan, and Dr. Fowler. Your questions and comments have made me a better researcher.

I am grateful for the support of my parents, who stood behind me as I took a leap of faith and pursued this degree. Without your encouragement, I would not dare to dream, and I wouldn't appreciate the value of taking the scenic route.

I have no words to express how grateful I am to my husband, Chad, who has sacrificed so much along with me to allow me to have time to work on this dissertation. Also, many thanks to Clara who has provided a strong and very special motivation for me to finish this race.

I am also thankful for the support of my friends and colleagues who have encouraged me through this extended process. Andrea, Shilpa, Busaba, Nat, Jing, Linda, Donita, Chwen, Diane, and many others have helped me persevere.

I would also like to acknowledge the organizations that supported me financially in this research. This work was sponsored in part by NSF and SRC (DMI-0432395). Also, the support I received through the Intel Foundation Ph.D. Fellowship and ASQ's Statistics Division's Ellis R. Ott Scholarship enabled me to continue and complete this research.

v

 ................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download