Conveners
Session: Quality by Design I
- Sonja Kuhnt (Dortmund University of Applied Sciences and Arts)
Session: Quality by Design II
- Jacqueline Asscher (Kinneret College, Technion)
Session: Hybrid Modelling I
- Pierantonio Facco (University of Padova)
Session: Hybrid Modelling II
- Dongda Zhang (University of Manchester)
Session: xAI
- Sonja Kuhnt (Dortmund University of Applied Sciences and Arts)
Session: Active Learning
- Alberto J. Ferrer-Riquelme (Universidad Politecnica de Valencia)
Session: PAT & Chemometrics
- Raffaele Vitale (Université de Lille)
Session: Data Science
- Alberto J. Ferrer-Riquelme (Universidad Politecnica de Valencia)
Design of experiments for process scale-up can be described as a double-edged sword for the pharmaceutical industry: intensification of experiments expands the knowledge of the process (uncertainty reduction) but increases resource expenditure. On the other hand, moving forward without enough process understanding is the first stone in a path of deviations, lack of quality, and even safety...
Capability indexes can be used to estimate how likely a given supplier of raw materials is to meet customer's requirements for these raw materials. It is therefore usually used by a customer operating a process as a criterion for selecting raw material suppliers. However, both univariate and multivariate capability indexes provided so far in the literature assume that the specifications are...
The current state-of-the-art in vaccine and pharmaceutical R&D is based on the “Quality-by-Design” paradigm, emphasizing risk-based and data-driven decisions. A key aspect is the classification of process parameters into critical and non-critical based on a series of Designs of Experiments (DoE). This process aids in understanding the relationship between Critical Process Parameters (CPPs) and...
The Quality by Design (QbD) approach has been widely adopted in the development of both novel and generic pharmaceutical formulations1. Extending these principles to the analytical domain, Analytical Quality by Design (AQbD) has emerged as a structured framework for optimizing analytical methodologies2. The aim of the present work was to outline a comprehensive framework for development of a...
At the present moment, vaccine manufacturing processes are mostly done in a fed-batch culture (i.e., nutrients are fed daily to the bioreactor). Feeding occurs at determined intervals and the volumes fed do not take into account possible changes in cell concentration and nutrients in real time. This can lead to variability in cell culture batches and scalability may also be an issue, as it is...
Process analytic technologies (PAT) are routinely used to rapidly assess quality properties in many industrial sectors. The performance of PAT-based models is, however, highly related to their ability to pre-process the spectra and select key wavebands. Amongst the modeling methodologies for PAT, partial least squares (PLS) (Wold, Sjöström and Eriksson, 2001) and interval partial least squares...
In the pharmaceutical industry, drug solubility is a critical quality attribute. For example, drug solubility in organic solvents mixtures is usually screened in drug development to select the best solvent system for crystallization in such a way as to design the manufacturing process. Solubility is also important in the final product because it has a direct impact on the way the drug is...
Keywords: Bioreactor modeling, Feed forward neural network, Hybrid semiparametric model, Physics-informed neural network, Fed-batch reactor
Abstract
Bioreactors are fundamental to bioprocess technology, yet the complexity of bioreactor systems continues to challenge effective digitalization and optimization. The intricate, dynamic nature of cell...
Keywords: Biopharma 4.0, Deep learning, Physics Informed Neural Networks, Bioreactors, Digital Twin
Abstract
Hybrid modeling combining First-Principles with Machine Learning (ML) is becoming a pivotal methodology for Industry 4.0 enactment. The combination of ML with prior knowledge generally improves the model predictive power and transparency while reducing the amount of data for process...
Keywords: Digital Twin, Hybrid Modeling, Machine Learning, Supercritical Carbon Dioxide Extraction, Process Optimization
Abstract
Supercritical carbon dioxide (ScCO2) extraction is a separation process that presents several advantages over traditional extraction methods of nonpolar solutes, eliminating the need for harmful organic solvents and costly post-processing steps required to...
End-to-End (E2E) models and Digital Twins in the pharmaceutical industry enhance efficiency, improve decision-making, allow for real-time monitoring, optimization, predictive analytics, and ultimately strengthen quality control and reduce costs. A key component of E2E models is the use of Monte Carlo simulations to capture uncertainties and variability within complex processes.
Typically,...
Despite advancements in Systems Biology, developing purely ODE-based mechanistic models remains challenging due to incomplete knowledge of parameters or computational inefficiencies. In such cases, hybrid and data-driven approaches provide viable alternatives. To facilitate seamless simulation and analysis alongside classical ODE-based models, it is advantageous to encode data-driven models in...
Introduction: The blood-brain barrier (BBB) severely restricts the passage of drugs into the brain, posing a significant challenge in treating central nervous system disorders such as glioblastoma (GBM). Therefore, there is an urgent need for advanced in vitro models that accurately characterize both BBB permeability and GBM behavior [1]. The evolution from two-dimensional (2D) to...
Inception for Petroleum Analysis (IPA) [1] is a deep convolutional network inspired from state-of-the-art computer vision architectures. IPA showed improved performance, compared to PLS, without depending on complex pre-processing operations thanks to its several computational blocks. The network begins with three stacked convolutions, followed by a multi-branch module consisting of four...
In artificial intelligence (AI), the complexity of many models and processes often surpasses human interpretability, making it challenging to understand why a specific prediction is made. This lack of transparency is particularly problematic in critical fields like healthcare, where trust in a model's predictions is paramount. As a result, the explainability of machine learning (ML) and other...
Convolutional Neural Networks (CNNs) have been increasingly used to build NIR based chemometric models with applications ranging from chemical sample analysis to food quality control. In the latter, NIR spectroscopy combined with CNNs enable rapid, non-destructive SOTA predictions of important quality parameters such as dry matter content in fruit [1].
The lack of a standard CNN architecture...
In process robustness studies, experimenters are interested in comparing the responses at different locations within the normal operating ranges of the process parameters to the response at the target operating condition. Small differences in the responses imply that the manufacturing process is not affected by the expected fluctuations in the process parameters, indicating its robustness. In...
Design of experiments is one of the main Quality by Design (QbD) tools within the process industry.
However, "classic" DoE is increasingly challenged by modern techniques such as Bayesian Optimization and Active Learning.
These innovative methods are promoted as faster and more intuitive, offering greater flexibility in experimentation.
In this talk, I will provide a direct comparison of...
Bayesian Optimization (BO) has been recently shown as an efficient method for data-driven optimization of expensive and unknown functions. BO relies on a probabilistic surrogate model, commonly a Gaussian Process (GP), and an auxiliary acquisition function that balances exploration and exploitation for a goal-oriented experimental design, with the aim of finding the global optimum under a...
Carefully designing experiments is crucial for gaining a deeper understanding of process behaviour. Design of Experiments (DOE) is a well-established active learning methodology with an extensive track record of solid contributions to research and industry in various areas, including screening, modelling, optimisation, specification matching, and robust design. Based on a reduced set of...
The development of pharmaceutical tablet manufacturing processes typically involves
time-consuming and resource-intensive development campaigns supported by equally
demanding laboratory analysis. These campaigns are designed to construct the safe
operating space to deliver the desired product quality. GSK is driving the digitalization
of tablet manufacturing via the use of digital twins...
In industrial practice, the development of pharmaceutical tablet manufacturing
processes typically involves time- and resource-intensive development campaigns
supported by equally resource demanding laboratory analysis. These campaigns are
designed to construct the safe operating space to deliver the desired product quality
for our patients. GSK is driving the digitalization of tablet...
Real-time monitoring of chemical processes is key for optimizing yield, preventing out of specification product, and improving overall process efficiency. Process analytical technology (PAT) tools, such as near-infrared and infrared (IR) spectroscopies, provide a real-time window into chemical processes, enabling non-destructive monitoring of analyte concentrations, and reducing dependance on...
Monitoring formulation quality during Continuous Direct Compression (CDC) and therefore remaining within product specifications is complex and cannot easily be inferred from process measurements. Process Analytical Technology (PAT) sensors allow in-line process monitoring and control of Critical Quality Attributes (CQAs), reducing the time and effort required for both sampling and off-line...
Online analysis has been widely developed to monitor the chemistry or the physics on batch or continuous processes. One of the major issues concerns the sampling part to integrate the analytical solution into the process. Optical spectroscopy is one of the most used technologies as it can be implemented directly inline and does not necessary required a sampling loop to adapt the process to the...
In the Chemical Manufacturing Industry, a diverse array of sensor technologies and data collection methods provide valuable insights into monitoring physical and chemical phenomena, equipment status, process conditions, raw material attributes, product quality, emissions, and logistics. Despite the extensive use of sensors, critical process information such as leaks, corrosion, and insulation...
Background
Monoclonal antibodies (mAbs) are highly specific proteins used in personalized therapeutics, with applications ranging from cancer treatment to autoimmune disease management. In this study, we focus on the production of 86 monoclonal antibody (mAb) molecules, each potentially having unique production characteristics.
Due to the confidential nature of proprietary data in the...