Achieving consistent yield and quality in modern commercial biologics manufacturing requires strong process understanding, integrated data systems, and predictive modeling. A structured AI/ML framework was applied using multi-year manufacturing data to develop hybrid and machine learning models that accurately predict key drivers of yield and product quality. The program incorporated automated data pipelines, parameter contextualization, and governance through routine review forums. Deployment resulted in higher yields, tighter quality profile, and increased overall process robustness.

Digital transformation in pharmaceutical development often fails because it prioritizes infrastructure and tools over processes, people, and culture. Novartis is taking a different approach, activating change through targeted catalysts and bottom up ownership. By combining local accountability with enterprise guardrails, the model accelerates adoption, surfaces inefficiencies earlier, and redefines data as a strategic asset—challenging traditional operating models and positioning data and digital as a source of competitive advantage.

Digital and AI transformation in bioprocessing often underperforms not due to technology, but because workforce systems are misaligned with process complexity. This talk introduces a quantitative framework for engineering workforce performance in regulated bioprocessing environments. It shows how instability, leadership dependency, and mis-sequenced retain-retrain-recruit-automate decisions create execution risk, and how organisations can intervene earlier to sustain throughput, compliance, and digital adoption.

This presentation will demonstrate how data-driven regulatory intelligence can revolutionize bioprocessing by automating compliance workflows, predicting process deviations, and accelerating scale-up decisions. Through real-world case studies and simplified AI workflows, attendees will see how raw manufacturing and quality data become actionable insights—transforming weeks-long regulatory tasks into minutes.

Digital biomanufacturing is driving the transition toward increasingly automated and autonomous process development and manufacturing. This presentation highlights recent advances in AI-assisted bioprocess modeling for CHO cell cultures, including hybrid modeling, LLM-assisted workflow automation, and specific rate estimation. Applications to process design and advanced control are presented, together with future opportunities for extending these approaches to downstream purification and end-to-end digital biomanufacturing.

Integrated continuous purification (ICP) involves highly interacting and synchronous unit operations, presenting unique challenges that require a control architecture to consistently meet product specifications and mitigate process deviations. This work presents automated control strategies for continuous viral inactivation and ultrafiltration/diafiltration within an end-to-end ICP platform. The developed model-based control approaches, maintained critical process parameters within ranges under representative perturbations of the ICP process, enhancing operational reliability across diverse process conditions

Nanobody molecules are an emerging class of biologics whose multivalent formats pose unique purification challenges due to structural flexibility and complex interactions. We developed the first high-fidelity mechanistic chromatography model for a Nanobody molecule, capturing complex elution behavior and all critical quality attributes to support late-stage process development. The validated model enables robust design space definition, scale-up across, and a predictive, digitally driven filing alternative to traditional empirical workflows.

We present an automated, data-rich platform for nanoparticle manufacturing that enables rapid, material-efficient identification of critical process parameters while ensuring reproducibility and regulatory relevance. The system integrates an impinging jet mixer with real-time, spatially resolved dynamic light scattering, coordinated through advanced control software, database management, and a user-friendly interface. Future integration of Bayesian optimization and automated Design of Experiments will further accelerate process development, demonstrated using model drug delivery systems.

The future of bioprocessing is autonomous. I will show how a CHO digital twin fuses genome-scale metabolic modeling with PAT-driven AI to forecast VCD and titer in real time. By introducing an XAI-guided, BO–enabled adaptive control framework, we move to closed-loop decision-making by updating recipes and feeding towards desired setpoints trajectories. The result is interpretable, high-performance control that enables transparent, end-to-end bioprocess optimization.

Digital twins are valuable for efficiently transferring complex processes from the laboratory to manufacturing scale and ensuring consistent results at different manufacturing sites. Here, we discuss case studies in the application of cell culture digital twins to tech transfer programs and demonstrate how modeling has allowed us to meet aggressive timelines and better serve the patients who need our products.

This presentation highlights the PatroLab digital platform, which integrates process analytical technology (PAT), digital twins, and predictive process control for advanced bioprocessing. By combing Raman spectroscopy, mechanistic models, and machine learning, PatroLab enables real-time monitoring, simulation, forecasting, and dynamic process control. Case studies will demonstrate improved process consistency, scalable process development, reduced variability, and enhanced manufacturing efficiency.

This presentation explores how digital twins, combining mechanistic process understanding, AI, and process data, enable smarter, faster bioprocess development and control. Six industrial use cases demonstrate the value of digital twins: accelerated biosimilar development using PAT and glycan modeling; reduced experimental effort in viral vector process design; media optimization through time-resolved nutrient uptake prediction; UF/DF development guided by digital membrane and recovery modeling; scale-up informed by CFD-based reactor behavior; and fully integrated digital control of continuous bioprocesses sustained for over 30 days. Together, these examples show how digital twins streamline experimentation, enhance decision-making, and de-risk scale-up—unlocking end-to-end process insight from early development to production.

CLAIRE (Cell Line AI Recognition and Evaluation) streamlines clonal CHO cell line development by combining deep-learning image analysis with automated liquid handling. It supports automated monoclonality verification, quality assessment, colony quantification, and hit-pick list generation for liquid handlers. Integrated within a user interface and paired with Lynx automation, it enables a streamlined, end-to-end workflow that reduces manual effort and compresses development timelines to 36 days.