Blog | 5/2/2025

The Evolution of AI in Bioprocessing: From Data Analysis to Autonomous Workflows

By Daniela Hristova-Neeley, PhD, Nick McConnell, PhD, Chris Wolfram, Alexander George, PhD, Wesley Liao

blog-image.jpg

AI and automation are steadily reshaping the bioproduction industry, but their impact is emerging in phases. Rather than a disruptive overhaul, the adoption of these technologies is occurring incrementally - starting with supportive functions outside the core production process and advancing toward increasingly autonomous control over bioprocess operations.

Health Advances team anticipates four stages in the evolution of AI in bioprocessing:

1.   Analysis Support: Driving Insights Outside the Bioreactor - Ready for Primetime

Overview: The first and most active area of AI adoption in bioproduction lies outside the process itself. AI and automation are being used to support peripheral, yet critical, functions such as data management, documentation, and supply chain coordination. These systems reduce data silos and help unlock insights across projects by centralizing and analyzing bioprocess data.

Example Products: Multiple companies provide digital solutions with built-in AI capabilities to automate data dissemination and provide insights from internal data. Some examples include:

  • Apprentice: Supports bioprocessing operations with MES system with onboard AI for decision-making and workflow automation
  • Aizon: Enables digitization of batch records and comparative analysis across runs using AI
  • Glide: Automates inventory management tasks for bioprocessing environments with AI to reduce time needed from procurement and supply chain teams

Current Status: These tools are broadly available and tailored for bioproduction use cases. Implementation is relatively straightforward, as they operate outside the highly regulated core process steps, making them accessible to early adopters.

Next Development Focus: As usage expands, refinements based on user feedback are likely to enhance usability and value. Broader awareness across the biomanufacturing sector is expected to follow, setting the stage for more integrated AI applications.

2.   Process Optimization: Enabling Smarter, Faster Development - Emerging

  • Overview: AI is increasingly being applied to accelerate and improve process development. Using historical data and simulation capabilities, AI can run in silico experiments to suggest optimal process conditions. This reduces the number of physical experiments needed and supports faster, more cost-effective development.
  • Example Products: Some companies developing analytical support solutions provide additional solutions for process optimization. Additional companies are focusing on batch optimization first including:
    • DataHow: Offers digital twin capabilities to simulate upstream processes
    • New Wave Biotech: Provides AI-driven simulations for upstream and downstream process design
    • BioRaptor: Utilizes AI as part of data analytics platform to support bioprocess optimization and integrate with LIMS and ELNs
  • Current Status: Digital twins and similar solutions are gaining attention. Similar to analysis support, these can be utilized without the need to adopt new technology that directly impacts bioproduction processes. This helps mitigate the perceived risks for early adopters and is likely to accelerate adoption among bioproduction stakeholders.
  • Next Development Focus: Continued development is likely to focus on expanding the scope and accuracy of models. As these tools mature, they may support optimization across broader portions of the workflow and integrate with automation systems.

3.   Single-Process Control: AI Enters the Bioproduction Workflow - Emerging

  • Overview: Beyond supporting development and analysis, AI is beginning to influence live process control. In this stage, AI and automation are used to monitor and adjust a specific step in the bioproduction process, typically in real time and based on continuous data inputs.
  • Example Use Case: Raman spectroscopy: Used in-line to measure cell metabolites, triggering automated feeding adjustments during upstream bioproduction.
  • Current Status: Early implementations are emerging, particularly in upstream applications. Investments in process analytical technologies (P.A.T) and control infrastructure are laying the groundwork for wider use of AI in this area.
  • Next Development Focus: Further progress will depend on improved data quality, more extensive sensor networks, and robust models trained on large, representative datasets. Increased process integration will follow as infrastructure and confidence grow.

4.   Multi-Process Control: Toward Autonomous Bioproduction - Long Term Goal

  • Overview: The most advanced envisioned application of AI involves the coordination and control of multiple interconnected bioprocess steps. In this future state, AI systems will dynamically manage entire workflows, continuously optimizing process conditions across upstream and downstream operations.
  • Example Use Case: A fully integrated system monitors key parameters (e.g., pH, protein purity, conductivity) and autonomously adjusts filtration, chromatography, and buffer exchange conditions to maintain optimal output.
  • Current Status: Reaching this level of capability will require seamless integration across process steps, standardized data frameworks, and validated AI models. Progress is expected to build gradually on the success of earlier adoption stages.
  • Next Development Focus: Incremental advances in system interoperability, data infrastructure, and control sophistication will be needed. Over time, individual AI-enabled steps may become standardized and eventually integrated into larger autonomous workflows.

In conclusion, the use of AI and automation in bioprocessing is evolving through a series of expanding capabilities—from data analysis and development support to live process control and, ultimately, autonomous workflows. While adoption is still in its early phases for many companies, solutions in analysis support and process optimization are already in use, and infrastructure for more advanced applications is steadily being built.

These developments suggest a clear direction for the industry: a stepwise journey toward intelligent, adaptive, and increasingly autonomous bioproduction systems.

 

Daniela Hristova-Neeley, PhD is a Partner and co-leader of Health Advances' Diagnostics, Precision Medicine, & Life Science Tools & Services Practice.

Nick McConnell, PhD is an Engagement Manager within the Health Advances' Diagnostics, Precision Medicine, & Life Science Tools & Services Practice.

Chris Wolfram is an Engagement Manager within the Health Advances' Diagnostics, Precision Medicine, & Life Science Tools & Services Practice.

Alexander George, PhD is a Sector Specialist within the Health Advances' Diagnostics, Precision Medicine, & Life Science Tools & Services Practice.

Wesley Liao is an Analyst within the Health Advances' Diagnostics, Precision Medicine, & Life Science Tools & Services Practice.

Share this: