Abstract: A new technique for parameter design is proposed. It is based on the observation that a loss function can be decomposed into two components each of which is monotone. This allows a separate data modeling and analysis to be applied to the two groups of data (above or below target). Then parameter design optimization can be performed based on the two fitted models. The technique is illustrated with the design of a heat exchanger.
Key words and phrases: Nominal-the-best parameter design, orthogonal array, signal-to-noise ratio.