Computational modeling, machine learning, and statistical data analysis are increasingly utilized to mitigate chemistry, manufacturing, and control failures related to particle properties in solid dosage form manufacture. Advances in particle characterization techniques and computational approaches provide unprecedented opportunities to explore relationships between particle morphology and drug product manufacturability. Achieving this, however, has numerous challenges such as producing and appropriately curating robust particle size and shape data. Addressing these challenges requires a harmonized strategy from material sampling practices, characterization technique selection, and data curation to provide data sets which are informative on material properties. Herein, common sources of error in particle characterization and data compression are reviewed, and a proposal for providing robust particle morphology (size and shape) data to support modeling efforts, approaches for data curation, and the outlook for modeling particle properties are discussed.
Keywords: characterization; computational modeling; particle properties.
© 2022. The Author(s), under exclusive licence to American Association of Pharmaceutical Scientists.