AI-based inline prediction of tablet properties for accelerating formulation development

Oct. 7, 2025, midnight - by Nico Kühnel , Jan Streckenbach , Pierre Braun , Moritz Rosch , Friederike Gütter , David Dormagen , Tim Landgraf

Remove from favorites

Add to favorites

Development of solid oral dosage forms depends on timely knowledge of critical quality attributes (CQAs) such as tablet hardness, tensile strength, solid fraction, and tablet weight. These CQAs are traditionally obtained by destructive offline tests that consume material and slow down iteration. A machine learning model was trained to predict these properties inline from process features during compaction. It enables faster screenings of blends and process settings while reducing reliance on destructive end-testing.

Download PDF