Abstract
Robust designs with dynamic characteristics, also called signal-response systems, are used in practice to determine control settings which make a system's performance robust over a wide range of input conditions. In this paper, we review existing methods for their analysis and propose the use of a joint generalized linear model. The joint model approach is based on standard regression modelling techniques so that model assumptions can be investigated using residual analysis. It has the advantage that it uses all of the data rather than relying on summary statistics, and the approach is applicable to a wide range of response types.
Additional information
Notes on contributors
M. L. Lesperance
Dr. Lesperance is Associate Professor in the Department of Mathematics and Statistics. Her email address is [email protected].
Sung-Min Park
Mr. Park is a Senior Manager at Samsung.