(185f) Development of a DEM-Based Digital Twin of a Continuous Direct Compression Line | AIChE

(185f) Development of a DEM-Based Digital Twin of a Continuous Direct Compression Line

Authors 

Doshi, P. - Presenter, Worldwide Research and Development, Pfizer Inc.
Toson, P., RCPE
Bohling, P., RCPE Gmbh
Trogrlic, M., RCPE
Matic, M., RCPE GmbH
Blackwood, D. O., Pfizer Worldwide Research and Development
Lee, K., Pfizer Inc.
Kimber, J., Pfizer Worldwide Research and Development
Khinast, J. G., Graz University of Technology
Jajcevic, D., Research Center Pharmaceutical Engineering GmbH, Graz, Austria
Verrier, H., Worldwide Research and Development, Pfizer Inc.
A Continuous Direct Compression line (CDC) is a minimal setup for the production of solid dosage forms and consists only of three unit operations: feeding of the individual components, mixing the individual material streams, and compressing the blend into tablets. The discrete element method (DEM) is a powerful tool for in-depth investigations of the individual unit operations, however, the computational effort does not allow to model a complete CDC line at once.

In order to create digital twin of the CDC line, reduced order models are needed. A popular implementation of the reduced order models is based on residence time distributions (RTDs) coming from tracer experiments [1] or simulations [2]. The digital twin is able to predict the content uniformity after each processing step and in the final tablets with real feeder fluctuation data and under constructed disturbance scenarios.

In addition to exemplary scenarios, the digital twin is able to create so-called funnel plots. The funnel plots show the maximum duration and intensity of disturbances that still allow the production of tablets with in-spec content uniformity without any control action. The digital twin is thus a valuable tool to develop the process control strategy for the CDC line.

References

[1] Lee et al., 2021. " Continuous Mixing Technology: Characterization of a vertical mixer using residence time distribution". Int J Pharm, in press. https://doi.org/10.1016/j.xphs.2021.01.035

[2] Toson et al. "Continuous Powder Mixing Technology: Validation of the DEM Model". submitted to J Pharm Sci.