(337bd) Monitoring and Control of an Integrated Crystallization-Filtration-Drying Platform for Continuous Pharmaceutical Manufacturing

My main interests lie on developing an advanced decision-making framework for chemical processes based on numerical tools. During my PhD, I am focusing on developing a pharmaceutical model library, PharmaPy, of active pharmaceutical ingredients (APIs) and applying it to digital design and control of manufacturing process.

Research Interests

This abstract highlights the research conducted during a PhD program, focused on the development of an advanced decision-making process for chemical processes using numerical tools. Recent disruptions in the international supply chain of critical medicines and APIs have emphasized the importance of continuous manufacturing (CM) as a key technology for achieving process intensification (PI) within the pharmaceutical sector. Still, there are many challenges associated with process development in CM, such as streamlining material flows across interconnected unit operations. But, this issue can be addressed through model-based simulation studies using digital twins. Digital twins provide a valuable tool for understanding process dynamics and investigating the design space (DS) of any complex system.

The overarching goal of my research is to demonstrate an advanced pharmaceutical manufacturing development framework that encompasses design, optimization, and control, by utilizing a digital twin of the physical system, thus ensuring robust process development in CM. In general, my research approach is structured around the development of a tool comprised of three key components. Firstly, it focuses on process modeling and the development of mechanistic models to describe and analyze chemical processes, particularly within pharmaceutical manufacturing for API purification. Secondly, it explores model-based optimization and control techniques applied to continuous purification process. The digital design can be investigated using the integrated flowsheet simulations using a sequential simulation-optimization approach with the aim of achieving desired critical quality attributes (CQA) for the API and effectively managing variations in set point trajectories under process uncertainties and disturbances. Lastly, the real-time optimization framework research involves the monitoring and real-time control of continuous pharmaceutical crystallization using process analytical technology (PAT), feedback control to achieve the desired crystal properties, purity, and yield within an intensified process unit.

In order to achieve the goals for my research, I have focused on the process development of continuous pharmaceutical purification steps, encompassing crystallization, filtration, washing, and drying. Thus, digital twins of the selected processes have been constructed using a set of mechanistic models for each step of the process, employing the object-oriented platform PharmaPy (Casas-Orozco et al., 2021). The platform's effectiveness has been tested through practical applications involving paracetamol, aspirin, and cannabidiol, wherein the system was optimized for maximum throughput and target residual moisture content/purity of the final crystal product. The model was able to optimize the system even in the presence of short-term disturbances in process parameters, such as filter fouling and variations in mass flow rate from the crystallizer. Additionally, a real-time monitoring framework is formulated using a moving horizon estimation framework, and a soft sensor for monitoring the final moisture content of the cake is demonstrated in aspirin production. These validation experiments not only confirm the functionality of the digital twin as a guide for mapping critical process parameters to achieve desirable critical quality attributes of the final product across multiple operational steps but also demonstrate its effectiveness as a real-time process monitoring and fault detection tool, capable of identifying critical process disturbances.

Overall, this research contributes to the advancement of advanced manufacturing methodologies in CM, facilitating improved process understanding, control, and efficiency within the pharmaceutical industry.

References

[1] Casas-Orozco, D., Laky, D., Wang, V., Abdi, M., Feng, X., Wood, E., Laird, C., Reklaitis, G. V., & Nagy, Z. K. (2021). PharmaPy: An object-oriented tool for the development of hybrid pharmaceutical flowsheets. Computers and Chemical Engineering, 153, 107408. https://doi.org/10.1016/j.compchemeng.2021.107408

[2] Hur, I., Casas-Orozco, D., Reklaitis, G. V., & Nagy, Z. K. “Intensified Continuous Purification Platform for Pharmaceutical Systems-Application to the Continuous Manufacturing of Pure Cannabidiol Crystals”, in AICHE Annual Meeting, 2022

[3] Hur, I., Wu, WL., Eren, A., & Nagy, Z. K. “Moisture content monitoring in continuous drug substance isolation manufacturing platform”, in AICHE Annual Meeting, 2021

[4] Hur, I., Casas-Orozco, & Nagy, Z. K. “Dynamic Flowsheet Simulation and Application of Soft Sensors on an Intensified and Integrated Purification Step for Pharmaceutical Upstream Manufacturing”, in AICHE Annual Meeting, 2021