Digital Twin Calibration for Biological System-of-Systems: Cell Culture Manufacturing Process
0
Sign in to get full access
Overview
- This paper presents a framework for calibrating digital twins of a biological system-of-systems (Bio-SoS), specifically a cell culture manufacturing process.
- The researchers developed a mechanistic model of the Bio-SoS and used Bayesian optimization to calibrate the model parameters to experimental data.
- The goal is to create a digital twin that can accurately simulate and predict the behavior of the real-world cell culture manufacturing process.
Plain English Explanation
A digital twin is a virtual model that mirrors a real-world system, like a manufacturing process. In this case, the researchers created a digital twin for a cell culture manufacturing process, which is used to produce biological products like medicines or vaccines.
[https://aimodels.fyi/papers/arxiv/digital-twins-forecasting-decision-optimisation-machine-learning]
To build the digital twin, the researchers developed a detailed mathematical model of the cell culture process. This model describes how the different components of the system, like the cells, nutrients, and equipment, interact and change over time.
[https://aimodels.fyi/papers/arxiv/digital-twin-generators-disease-modeling]
The researchers then used a technique called Bayesian optimization to fine-tune the model parameters so that the digital twin closely matches the behavior of the real-world cell culture process. This allows the digital twin to accurately simulate and predict the performance of the actual manufacturing system.
[https://aimodels.fyi/papers/arxiv/bayesian-co-navigation-dynamic-designing-materials-digital]
By having an accurate digital twin, the researchers can use it to test different process conditions, experiment with new designs, and optimize the manufacturing process without having to physically test everything on the real system. This can save time and money in the development and production of biologics.
Technical Explanation
The researchers developed a mechanistic model of the Bio-SoS, which encompasses the various components and interactions within the cell culture manufacturing process. This includes modeling the dynamics of the cells, nutrients, byproducts, and equipment parameters over time.
[https://aimodels.fyi/papers/arxiv/towards-validation-autonomous-vehicles-across-scales-using]
To calibrate the digital twin, the researchers used Bayesian optimization, a machine learning technique that iteratively adjusts the model parameters to minimize the discrepancy between the digital twin's predictions and experimental data. This allows the digital twin to accurately capture the behavior of the real-world system.
The calibrated digital twin can then be used to test different process conditions, evaluate new designs, and optimize the manufacturing process without the need for extensive physical experimentation. This can lead to faster development and improved productivity of the cell culture system.
Critical Analysis
The researchers acknowledge that the accuracy of the digital twin relies on the quality and completeness of the underlying mechanistic model and the available experimental data used for calibration. Limitations in model fidelity or data availability could impact the digital twin's predictive capabilities.
[https://aimodels.fyi/papers/arxiv/towards-validation-autonomous-vehicles-across-scales-using]
Additionally, the Bayesian optimization approach used for calibration assumes that the model parameters follow certain statistical distributions, which may not always hold true in complex biological systems. Further research is needed to explore alternative calibration methods that can handle more diverse parameter spaces.
[https://aimodels.fyi/papers/arxiv/bayesian-co-navigation-dynamic-designing-materials-digital]
Overall, the proposed framework demonstrates a promising approach for developing accurate digital twins of biological manufacturing processes. However, continued validation and refinement of the models and calibration techniques will be crucial for broader adoption and real-world impact.
Conclusion
This paper presents a framework for calibrating digital twins of a biological system-of-systems, specifically a cell culture manufacturing process. By developing a mechanistic model of the system and using Bayesian optimization for parameter calibration, the researchers were able to create a digital twin that closely matches the behavior of the real-world process.
[https://aimodels.fyi/papers/arxiv/digital-twins-forecasting-decision-optimisation-machine-learning]
The calibrated digital twin can then be used to test process conditions, evaluate new designs, and optimize the manufacturing process, leading to faster development and improved productivity of biologics production. While the approach shows promise, further research is needed to address the limitations and fully realize the potential of digital twins in the biomanufacturing domain.
This summary was produced with help from an AI and may contain inaccuracies - check out the links to read the original source documents!
Related Papers
0
Digital Twin Calibration for Biological System-of-Systems: Cell Culture Manufacturing Process
Fuqiang Cheng, Wei Xie, Hua Zheng
Biomanufacturing innovation relies on an efficient Design of Experiments (DoEs) to optimize processes and product quality. Traditional DoE methods, ignoring the underlying bioprocessing mechanisms, often suffer from a lack of interpretability and sample efficiency. This limitation motivates us to create a new optimal learning approach for digital twin model calibration. In this study, we consider the cell culture process multi-scale mechanistic model, also known as Biological System-of-Systems (Bio-SoS). This model with a modular design, composed of sub-models, allows us to integrate data across various production processes. To calibrate the Bio-SoS digital twin, we evaluate the mean squared error of model prediction and develop a computational approach to quantify the impact of parameter estimation error of individual sub-models on the prediction accuracy of digital twin, which can guide sample-efficient and interpretable DoEs.
Read more7/1/2024
0
Sparse Attention-driven Quality Prediction for Production Process Optimization in Digital Twins
Yanlei Yin, Lihua Wang, Dinh Thai Hoang, Wenbo Wang, Dusit Niyato
In the process industry, long-term and efficient optimization of production lines requires real-time monitoring and analysis of operational states to fine-tune production line parameters. However, complexity in operational logic and intricate coupling of production process parameters make it difficult to develop an accurate mathematical model for the entire process, thus hindering the deployment of efficient optimization mechanisms. In view of these difficulties, we propose to deploy a digital twin of the production line by encoding its operational logic in a data-driven approach. By iteratively mapping the real-world data reflecting equipment operation status and product quality indicators in the digital twin, we adopt a quality prediction model for production process based on self-attention-enabled temporal convolutional neural networks. This model enables the data-driven state evolution of the digital twin. The digital twin takes a role of aggregating the information of actual operating conditions and the results of quality-sensitive analysis, which facilitates the optimization of process production with virtual-reality evolution. Leveraging the digital twin as an information-flow carrier, we extract temporal features from key process indicators and establish a production process quality prediction model based on the proposed deep neural network. Our operation experiments on a specific tobacco shredding line demonstrate that the proposed digital twin-based production process optimization method fosters seamless integration between virtual and real production lines. This integration achieves an average operating status prediction accuracy of over 98% and a product quality acceptance rate of over 96%.
Read more8/27/2024
↗️
0
New!Benchmarking Sim2Real Gap: High-fidelity Digital Twinning of Agile Manufacturing
Sunny Katyara, Suchita Sharma, Praveen Damacharla, Carlos Garcia Santiago, Lubina Dhirani, Bhawani Shankar Chowdhry
As the manufacturing industry shifts from mass production to mass customization, there is a growing emphasis on adopting agile, resilient, and human-centric methodologies in line with the directives of Industry 5.0. Central to this transformation is the deployment of digital twins, a technology that digitally replicates manufacturing assets to enable enhanced process optimization, predictive maintenance, synthetic data generation, and accelerated customization and prototyping. This chapter delves into the technologies underpinning the creation of digital twins specifically tailored to agile manufacturing scenarios within the realm of robotic automation. It explores the transfer of trained policies and process optimizations from simulated settings to real-world applications through advanced techniques such as domain randomization, domain adaptation, curriculum learning, and model-based system identification. The chapter also examines various industrial manufacturing automation scenarios, including bin-picking, part inspection, and product assembly, under Sim2Real conditions. The performance of digital twin technologies in these scenarios is evaluated using practical metrics including data latency, adaptation rate, simulation fidelity among others reported, providing a comprehensive assessment of their efficacy and potential impact on modern manufacturing processes.
Read more9/18/2024
⛏️
0
A Digital Twin Design Methodology for Control, Simulation, and Monitoring of Fluidic Circuits
Veyis Gunes
We propose a synthesis method for the design of digital twins applicable to various systems (pneumatic, hydraulic, electrical/electronic circuits). The methodology allows representing the operation of these systems through an active digital twin, thereby enabling a more suitable and easier computer-aided design, simulation, control, and monitoring. Furthermore, our methodology enables the detection of a system's actions on its own inputs (for example, in pneumatics: backflow of gases trapped in part of a fluidic system onto its own inputs). During the simulation or monitoring phase, the approach also facilitates real-time diagnosis of the controlled system. The outputs, on the controlled physical system or its digital twin, do not depend only on the current inputs but also on the history of the inputs and the history of internal states and variables. In other words, the underlying sequential logic has a memory while an only combinational logic approach does not. These capabilities can contribute to the digital transformation of the factory of the future.
Read more9/9/2024