In previous work, model-based methods have been developed for efficien
t testing of components and instruments that allow for their full beha
vior to be predicted from a small set of test measurements, While such
methods can significantly reduce the testing cost of such units, thes
e methods are valid only if the model accurately represents the behavi
or of the units, Previous papers on this subject described many method
s for developing accurate models and using them to develop efficient t
est methods, However, they gave little consideration to the problem of
testing units which change their behavior after the model has been de
veloped, for example, as a result of changes in the manufacturing proc
ess. Such changed behavior is referred to as nonmodel behavior or nonm
odel error, When units with this new behavior are tested with these mo
re efficient methods, their predicted behavior can show significant de
viations from their true behavior, This paper describes how to analyze
the data taken at the reduced set of measurements to estimate the unc
ertainty in the model predictions, even when the device has significan
t nonmodel error, Results of simulation are used to verify the accurac
y of the estimates and to show the expected variation in the results f
or many modeling variables.