Data-Knowledge-Driven Modeling and Operational Adjustment for the Pharmaceutical Tablet Manufacturing Process via Wet...

Area of Science:

• Pharmaceutical Manufacturing
• Quality Control
• Process Engineering

Background:

• Pharmaceutical Quality Control (PQC) faces challenges in Pharma 4.0 due to variable material attributes and complex process couplings in Pharmaceutical Tablet Manufacturing Processes (PTMP).
• Timely adjustment of operational variables is crucial for maintaining quality in PTMP.

Purpose of the Study:

• To propose a novel data-knowledge-driven modeling and operational adjustment framework for PTMP.
• To address uncertainties and variable couplings in pharmaceutical manufacturing for improved PQC.

Main Methods:

• Integration of Bayesian Network (BN) and Case-Based Reasoning (CBR) for a data-knowledge-driven approach.
• Development of a distributed BN model for each PTMP subunit, integrated into a global BN model.
• Operational adjustment achieved through global BN reasoning using expected Critical Quality Attributes (CQAs) as evidence.

Main Results:

• Demonstrated feasibility and effectiveness of the proposed framework in a sprayed fluidized-bed granulation-based PTMP.
• Showcased improvement in terminal CQAs of the manufactured tablets.
• Comparative analysis highlighted the efficacy and merits of the method over existing approaches.

Conclusions:

• The integrated BN and CBR framework provides a robust solution for PQC in complex pharmaceutical manufacturing.
• The method effectively enhances terminal CQAs by enabling timely operational adjustments.
• This approach offers significant advantages for Pharma 4.0 initiatives in pharmaceutical production.