The models are configured using dropdown options with an easy-to-use graphical user interface. The model libraries have been implemented in close collaboration with industry and academia based on the best science available in the literature. Experimental data is imported and used to calibrate parameters and validate model predictions. Sophisticated approaches can then be used to optimise the process directly and automatically explore the process design space, replacing physical experimentation with in-silico simulation in R&D to perform tasks faster, cheaper and with greater understanding.
Digital twins for efficient bioprocess development
Siemens has an established library of integrated mechanistic and hybrid mathematical models of bioprocesses. Customers select, configure and calibrate their model to fit their process and objective from the many options available which cover all the main unit operations.
Leverage the power of science-based digital twins
See how science-based digital twins support decision-making across the biopharmaceutical lifecycle.
The models can be deployed as web applications for easily accessible user-friendly simulation tools that can be made available across the organisation without the need for end users to install any software or take any training. The same underlying model can be deployed online with bi-directional connection with the physical process to monitor, predict and optimise the process in real-time for increased productivity, yield and robustness.
End-to-end bioprocess digital twins
See how end-to-end bioprocess digital twins can help evaluate control strategies and derisk technology transfer.
Bioprocess digital twins begin by configuring, calibrating and validating models per unit operation. These models are used for in silico studies, offering faster, cheaper, and less risky alternatives to purely experimental approaches. While most focus on single unit operations, gPROMS enables the development and deployment of integrated end-to-end bioprocess models. These models can perform tasks not possible with multivariate experiments, facilitating efficient scale-up of your process and testing of control strategies before investing in physical equipment, ensuring right-first-time success.