Abstract
This study develops a statistical framework for characterizing an improved Buckypaper manufacturing process. A screening procedure based on design of experiment techniques and the Lenth method was used to identify significant process parameters. These parameters significantly affect the tensile strength of the Buckypaper produced by the process. The procedure only samples a limited number of experimental runs. A linear regression model and a nonlinear model based on the Kriging method were constructed to characterize the relationship between the parameters and tensile strength. Tradeoff curves were used to determine the best parameter settings when simultaneously considering the mean and variance of the tensile strength. Validation tests show that both models are better predictors than previous models and the nonlinear model outperforms the linear model. The statistical framework provides a systematic and effective method of characterizing the manufacture of nanotube products. © 2014 Korean Society for Precision Engineering and Springer-Verlag Berlin Heidelberg.