A procedure to check the validity of near-infrared calibration models over time is proposed. Validation samples are analyzed at regular time intervals by both near-infrared and reference methods in order to check the validity of the near-infrared calibration model. When an invalid situation is detected, a stable polystyrene standard is measured, to determine whether this invalid situation is due to fluctuations of the instrumental response of the near-infrared spectrometer or not. A first method based on simulating simple instrumental differences enables one to correct those simple differences with only the polystyrene standard. It is shown that such an approach can correct most of the simple instrumental differences without the use of standardization procedures. However, when the invalid situation is due to more complex instrumental differences, standardization must be applied. This procedure is applied to simulated and real near-infrared data sets.
You do not have subscription access to this journal. Cited by links are available to subscribers only. You may subscribe either as an OSA member, or as an authorized user of your institution.
Contact your librarian or system administrator
Login to access OSA Member Subscription