(567b) Development of a DEM Based Digital Twin of a Tablet Press Feed Frame for the Evaluation of Continuous Manufacturing Control Strategies.

Pharmaceutical process development can require a substantial amount of material for experimental characterization, and experimental work is therefore typically restricted to a small design space (e.g. narrow material variation and operating range). In order for the industry to explore a larger design space and carry out more optimization studies, in silico modelling and high-fidelity digital twins offer the possibility to perform numerous virtual experiments.

The Research Center Pharmaceutical Engineering (Austria), supported by the U.S. FDA, have recently started a project to develop an approach for assessing drug manufacturing control strategies based on high-fidelity digital twins, taking a direct compression line as a use-case. In this high-fidelity digital twin platform, typical pharmaceutical materials, equipment, and processes were simulated using verified numerical methods and algorithms [1,2,3].

This work presents simulations of a tablet press feed frame using the discrete-element-method (DEM) model. Due to the large hold-up masses and mean residence times in the feed frame, it was computationally unreasonable to resolve the full Residence Time Distribution (RTD) with DEM simulations. Therefore, a DEM extrapolation technique colloquially called "Relay Race" was used to predict the full RTD. The RTD obtained in this fashion was then used in a reduced order model to predict responses to various disturbances at different feed frame throughputs and tablet press speeds. These predictions were compared to experimental data. Analysis of the DEM data provided a deeper insight into the process, revealing flow patterns that are very challenging to determine experimentally.

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

References

[1] Siegmann et al., 2020. Powder flow and mixing in different tablet press feed frames. Adv Pow Tech 31, 770–781. https://doi.org/10.1016/j.apt.2019.11.031

[2] Toson et al., 2019. Explicit Residence Time Distribution of a Generalised Cascade of Continuous Stirred Tank Reactors for a Description of Short Recirculation Time (Bypassing). Processes 7, 615. https://doi.org/10.3390/pr7090615

[3] Siegmann et al.,2021. Massively speeding up DEM simulations of continuous processes using DEM extrapolation. Pow Tech 390, 442–455. https://doi.org/10.1016/j.powtec.2021.05.067