In manufacturing high-technology products using automated manufacturing equ
ipment it is critical to obtain process-operating conditions that simultane
ously optimize several output variables of interest. Often these output var
iables are non-traditional responses such as standard deviations, and there
may even be discrete (count or binary) responses. Statistically designed e
xperiments are very useful in optimizing these processes. We illustrate how
factorial experiments, statistical models for the responses of interest, a
nd simple optimization techniques can be successfully applied to a bonded l
eads process. (C) 2000 Published by Elsevier Science Ltd.