Building an accelerated Digital Twin to understand miscible fluid blending

About the customer

AbbVie, an American research-based biopharmaceutical company headquartered in Lake Bluff, Illinois, United States (U.S.), focuses on discovering and delivering transformational medicines and products in several key therapeutic areas, including immunology, oncology, neuroscience, eye care, virology and women’s health.

The following case study explores research conducted jointly with AbbVie, published in AAPS PharmSciTech by authors John Thomas, Kushal Sinha, Gayathri Shivkumar, Lei Cao, Marina Funck, Sherwin Shang and Nandkishor K. Nere.

Their challenge

A common challenge in biopharmaceutical manufacturing is the effective mixing of stratified miscible fluids with large disparities in density and viscosity. The differences between fluid densities and viscosities can lead to order-of-magnitude increases in blend time compared to the blending of single-fluid systems. Additionally, the mixing performance in two-fluid systems can be strongly dependent on the position of the impeller relative to the fluid interface. Due to the complexities of the fluids involved and engineering design limitations, manufacturers struggle to scale up production.

The challenge extends further when considering the limitations of three common approaches. First, literature correlations are scarce for two-fluid systems with large density and viscosity disparities. Second, numerical modeling is often limited by the complex 3D, time-evolving nature of transport physics, which restricts the applicability of Reynolds-averaged Navier-Stokes (RANS)-based and other time-averaged modeling methods. Lastly, experimental approaches are frequently constrained by limited equipment access, high material expenses and specialized labor costs.

Our solution

To overcome the limitations of traditional methods, AbbVie implemented a high-performance Digital Twin strategy using Simcenter™ M-Star™ software, which is part of the Siemens Xcelerator business platform of software, hardware and services. By using advanced lattice-Boltzmann transport algorithms with graphics processing unit (GPU)-based hardware, Abbvie developed a high-fidelity digital replica of their physical mixing tanks capable of running real-time numerical experiments.

Although the Digital Twin significantly accelerates the design process, it remains grounded in physical reality. Precise experimental measurements of fluid properties are supplied to the model, and experimental blend-time and power number data are used to validate the simulation output. This integrated approach ensures that the results are fully converged and reliable for the specific operating conditions of interest.

Results

Simcenter M-Star delivers a fundamental shift in simulation efficiency for AbbVie: complex fluid mechanics are resolved in hours, and full process hours are completed within days of total turnaround time. This high-speed Digital Twin generated transient processing insights with a level of fidelity that rivals experimental measurement—at a significantly lower cost both computationally and financially.

Overall, the Digital Twin strategy delivered value across three critical dimensions. First, it ensured output reproducibility, providing high-fidelity predictions that were functionally indistinguishable from measurements taken on physical equipment. Second, it provided resource scalability by drastically reducing the time and materials required to obtain comparable process data. Finally, Simcenter M-Star ensured process applicability, allowing AbbVie to leverage multiple in-silico experiments to generate robust process design correlations that directly informed their manufacturing strategy.