Abstract
The failure amplification method (FAMe) uses a special type of factor known as amplification factor to amplify the failure probability so as to maximize the information in the experiment. A general strategy for model building in FAMe is proposed by using generalized linear models (GLM). The best design settings for improving the process capability are determined through carefully selected GLMs and loss functions. Two real experiments for improving the quality of printed circuit boards are used to illustrate the proposed strategy.
Additional information
Notes on contributors
Shuen-Lin Jeng
Dr. Jeng is an Associate Professor in the Department of Statistics and a member of ASA. His email address is [email protected].
V. Roshan Joseph
Dr. Joseph is an Assistant Professor in the H. Milton Stewart School of Industrial and Systems Engineering and a member of ASQ. His email address is [email protected].
C. F. Jeff Wu
Dr. Wu is the Coca-Cola Chair Professor in Engineering Statistics in the H. Milton Stewart School of Industrial and Systems Engineering and a Fellow of ASQ. His email address is [email protected].