Multivariate statistical control in conjunction with mid-infrared spectroscopy was implemented to monitor the quality of commercial detergents. The approach was developed by estimating the Hotelling T2 and Square Prediction Error Q statistics. A joint analysis of these two scalars has led to the introduction of a bivariate probability density function, which brings to the proposal of a novel normal operating region for the process. The sensitivity to detect abnormal processes is shown to be improved, with a correct identification of the detergent samples out of specifications.
Statistical control of commercial detergents production through fourier transform infra-red spectroscopy
TARIS, ALESSANDRA;GROSSO, MASSIMILIANO;
2014-01-01
Abstract
Multivariate statistical control in conjunction with mid-infrared spectroscopy was implemented to monitor the quality of commercial detergents. The approach was developed by estimating the Hotelling T2 and Square Prediction Error Q statistics. A joint analysis of these two scalars has led to the introduction of a bivariate probability density function, which brings to the proposal of a novel normal operating region for the process. The sensitivity to detect abnormal processes is shown to be improved, with a correct identification of the detergent samples out of specifications.
File in questo prodotto:
Non ci sono file associati a questo prodotto.
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
