A hybrid NIR-soft sensor method for real time in-process control during continuous direct compression manufacturing operations

Management and Production Engineering Review

“…Many studies are dealing with the optimization of the control time, Gharbi et al (2022) developed an optimal control policy for unreliable manufacturing systems evolving in a stochastic and dynamic environment based on a new stochastic dynamic programming model. Cogoni et al (2021) proposed a novel hybrid method for optimizing the quality control time in manufacturing systems by combining the online NIR PLS and sensors, enabling a robust prediction. Pacana et al (2021) studied the improvement of the production process and the optimization of control time through the use of checkpoints and proposed a new method of quality control check in complex industrial manufacturing systems.…”

Feasibility of an NIR spectral calibration transfer algorithm based on optimized feature variables to predict tobacco samples in different states

“…17,18 This allows for the variations in the physical properties and chemical composition of substances to be reflected in real time. 19–22…”

“…17,18 This allows for the variations in the physical properties and chemical composition of substances to be reected in real time. [19][20][21][22] Over the past few years, NIR spectroscopy has provided important guidance in the identication of geographical origin, 23 grade determination, 24 and detection of the chemical composition of tobacco. 25 A reliable multivariate calibration model is developed from the calibration set samples with known attribute measurements, followed by prediction of the unknown sample characteristics, while the accuracy of the prediction results is always strongly associated with the homogeneity of the samples.…”