Abstract:
|
Two dimensional (2-D) data is used for quality characterization in certain manufacturing processes. In this research, we use a nanomanufacturing process as an example. A carbon nanotube (CNT) film is a spatially distributed thin film. The anisotropy of a CNT film, which is a spatially distributed quality index, is difficult to measure in practice due to metrology and cost constraints. It is learned that the anisotropy is highly correlated with another quality variable, the height of the CNT array, which can be measured easily in more cost-effective way. Therefore, in this research, we propose a spatial model for predicting the anisotropy using the height. The model takes the spatially distributed 2-D height as an input and provides a predicted anisotropy distribution in a 2D space. A prediction model is also developed if the anisotropy measures are partially available. The performance of the proposed model is verified by both a simulation study and real data samples.
|
ASA Meetings Department
732 North Washington Street, Alexandria, VA 22314
(703) 684-1221 • meetings@amstat.org
Copyright © American Statistical Association.