(652c) The Control Strategy Evaluation Tool: A User-Friendly Toolchain Based on High Fidelity DEM Simulations

Pharmaceutical process development can require a substantial amount of materials for experimental characterization. Consequently, experimental work is usually restricted to a small design space. In order to explore a larger design space and perform more optimization studies, in silico modelling and high-fidelity digital twins offer the possibility to perform numerous experiments virtually.

The Research Center Pharmaceutical Engineering (Austria) have recently started an FDA funded research project to develop an approach to assess control strategies based on high-fidelity digital twins. This high-fidelity digital twin platform relies on simulating typical pharmaceutical materials, equipment, and processes using verified numerical methods and algorithms. Using a continuous direct compression (CDC) line as an example, a DEM model is employed to simulate powder flow and predict the residence time distribution (RTD) in a horizontal blender connected to a three-paddled feed frame. The RTD is then used to parametrize reduced order models (ROM) that are used to predict system responses at various process settings.

The ROMs are used as core components of a control strategy evaluation tool (CET). The CET offers the possibility to build a digital twin of continuous manufacturing lines and provides a systematic framework for implementing discharge strategies. It includes (1) a modular simulation model of pharmaceutical manufacturing lines, (2) a library of typical disturbance scenarios, and (3) an algorithm for implementing a control strategy in terms of material discharge. The CET can be used to evaluate different discharge locations, determine the effect of various types of disturbances, or specify discharge thresholds. The CET functionality is demonstrated using selected scenarios and compared to traditional approaches in terms of key performance indicators, such as the waste amount and final product quality.

Disclaimer: This abstract reflects the views of the authors and should not be constructed to represent FDA’s views or policies.