TY - GEN
T1 - A semiconductor yields prediction using stepwise support vector machine
AU - An, Daewoong
AU - Ko, Hyo Heon
AU - Gulambar, Turghun
AU - Kim, Jihyun
AU - Baek, Jun Geol
AU - Kim, Sung Shick
PY - 2009
Y1 - 2009
N2 - It is crucial to prevent low yields in the semiconductor industry. Since many factors affect variation in yield and they are deeply related, preventing low yield is difficult. There have been substantial researches in the field of yield prediction. Many researchers had used statistical methods. Many studies have shown that artificial neural network (ANN) achieved better performance than traditional statistical methods. However, despite ANN's superior performance some problems such as over-fitting and poor explanatory power arise. In order to overcome these limitations, a relatively new machine learning technique, support vector machine (SVM), is introduced to classify the yield. SVM is simple enough to be analyzed mathematically, and it leads to high performances in practical applications. This study presents a new efficient classification methodology, Stepwise-SVM, for detecting high and low yields. Stepwise-SVM is a step-by-step adjustment of parameters to classify actual high and low yield lot precisely. The objective of this paper is to examine the feasibility of SVM and stepwise-SVM in the yield classification. The experimental results show that SVM and stepwise-SVM provides a promising alternative to yield classification for the field data.
AB - It is crucial to prevent low yields in the semiconductor industry. Since many factors affect variation in yield and they are deeply related, preventing low yield is difficult. There have been substantial researches in the field of yield prediction. Many researchers had used statistical methods. Many studies have shown that artificial neural network (ANN) achieved better performance than traditional statistical methods. However, despite ANN's superior performance some problems such as over-fitting and poor explanatory power arise. In order to overcome these limitations, a relatively new machine learning technique, support vector machine (SVM), is introduced to classify the yield. SVM is simple enough to be analyzed mathematically, and it leads to high performances in practical applications. This study presents a new efficient classification methodology, Stepwise-SVM, for detecting high and low yields. Stepwise-SVM is a step-by-step adjustment of parameters to classify actual high and low yield lot precisely. The objective of this paper is to examine the feasibility of SVM and stepwise-SVM in the yield classification. The experimental results show that SVM and stepwise-SVM provides a promising alternative to yield classification for the field data.
KW - Semiconductor manufacturing process
KW - Semiconductor yield classification
KW - Support vector machine
UR - http://www.scopus.com/inward/record.url?scp=77949388943&partnerID=8YFLogxK
U2 - 10.1109/ISAM.2009.5376916
DO - 10.1109/ISAM.2009.5376916
M3 - Conference contribution
AN - SCOPUS:77949388943
SN - 9781424446278
T3 - 2009 IEEE International Symposium on Assembly and Manufacturing, ISAM 2009
SP - 130
EP - 136
BT - 2009 IEEE International Symposium on Assembly and Manufacturing, ISAM 2009
T2 - 2009 IEEE International Symposium on Assembly and Manufacturing, ISAM 2009
Y2 - 17 November 2009 through 20 November 2009
ER -