Premium
Polyolefin formulation validation using infrared spectrometric classification techniques
Author(s) -
Van Every K. W.,
Elder M. J.
Publication year - 1996
Publication title -
journal of vinyl and additive technology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.295
H-Index - 35
eISSN - 1548-0585
pISSN - 1083-5601
DOI - 10.1002/vnl.10130
Subject(s) - polyolefin , calibration , linear discriminant analysis , computer science , principal component analysis , chemometrics , process engineering , discriminant , infrared spectroscopy , process (computing) , quality (philosophy) , biochemical engineering , machine learning , artificial intelligence , materials science , mathematics , engineering , chemistry , nanotechnology , statistics , organic chemistry , physics , layer (electronics) , quantum mechanics , operating system
Discriminant analysis using principal components of mid‐infrared spectral data is demonstrated as a powerful quality validation tool to augment sophisticated manufacturing processes where rigorous multicomponent analysis is not only costly but in many cases unwarranted. In order to balance competing requirements of validation, a multiple model approach is taken, which combines the best information available about the process, the chemistry and the spectroscopy of the formulation. A systematic approach to evaluate training data is discussed as a crucial step to insuring a diverse but homogenous calibration.