Novel Hybrid Model to Predict Ribbon Solid Fractions in the Roller Compaction Process.

Area of Science:

• Pharmaceutical Engineering
• Process Chemistry
• Materials Science

Background:

• Roller compaction is a key unit operation in pharmaceutical manufacturing for producing granules.
• Scale-up of roller compaction often requires extensive experimental data, which is time-consuming and costly.
• Predictive modeling can optimize the roller compaction process and reduce experimental burden.

Purpose of the Study:

• To develop and validate a calibrated model for scaling up roller compaction processes.
• To utilize minimal experimental data for accurate prediction of ribbon solid fraction.
• To enable the application of digital tools for efficient drug product process development.

Main Methods:

• Laboratory-scale roller compaction simulation to evaluate blend compression properties.
• Development of a calibrated model combining the Reynolds roller compaction model and ribbon solid fraction measurements.
• Validation of the model against experimental data from different scales and equipment.
• Extension of the model into a hybrid approach incorporating API and formulation properties for partially characterized blends.

Main Results:

• The calibrated model accurately predicts ribbon solid fraction across different scales and drug loadings.
• The hybrid model shows reasonable agreement with experimental measurements for partially characterized blends.
• Digital Design of Experiments (DOEs) were successfully implemented using the hybrid model to define operating ranges.
• The developed method significantly reduces the experimental data required for roller compaction scale-up.

Conclusions:

• A validated, calibrated model enables accurate prediction of roller compaction behavior with minimal data.
• The hybrid model offers a powerful tool for scale-up and process optimization of pharmaceutical products.
• Digitalization of roller compaction scale-up facilitates efficient development of fast-moving drug products.