Biopharma Manufacturing Needs Connected Data and Workforce Readiness to Scale New Therapies

Biopharma’s ability to make advanced therapies reliably and at scale is lagging behind its scientific progress, says Nicolas Pivet, Cytiva’s vice president and general manager of manufacturing and digital solutions. RNA-based treatments and cell and gene therapies have advanced, but production systems still struggle to reproduce processes consistently as they expand.

Pivet argues that familiar obstacles—including paper batch records, disconnected laboratory information systems, fragmented automation, separate enterprise resource planning tools and isolated data—are signs of a deeper problem. Many facilities developed over decades, with digital tools added to address individual needs rather than designed around connected manufacturing operations. The result can be abundant data without enough timely insight to guide decisions.

Build visibility before adding advanced tools

For Pivet, the first priority is observability: seeing what is happening across a process well enough to detect deviations early and respond. Earlier insight can support better decisions and greater confidence in them, forming a base for more advanced manufacturing approaches.

That requires connecting information from processes, equipment, analytical systems and workflows, while making it accessible, trustworthy and meaningful in context. Once those links are in place, manufacturers have a stronger foundation for predictive analytics, automation, digital twins and artificial intelligence. Pivet cautions against beginning with AI applications before addressing visibility, connectivity, data integrity and operational context. As he puts it, AI can expose weak foundations rather than repair them.

The sequence matters because a model or automated system can only support sound decisions when the information feeding it is reliable and relevant. Better-connected data also helps teams understand how a change in one part of production may affect other steps. That makes it easier to act on emerging problems instead of simply collecting records after the fact.

Organizational trust shapes the pace of change

Digital transformation can be held back more by organizational reluctance than by technical limits. Manufacturers may see changing established practices as riskier than continuing with them. Pivet says that, in 2025, more than a third of biopharma companies were still at an early exploration stage or had not begun their digital transformation.

Trust is a major factor in whether change takes hold, and it depends on dependable performance. Pivet sees faster progress at organizations that start with real production challenges, check whether an approach delivers the intended result and expand it gradually. That path can make the benefits clearer to the people who will use the new systems and reduce the uncertainty surrounding broader change.

He expects future manufacturing equipment to be designed as part of connected, interoperable systems, with visibility and integration built in from the start. That shift would make it easier to link equipment with automated workflows, rather than repeatedly adding connections to tools that were designed to operate separately.

Prepare workers for more data-focused roles

Technology is only part of the transition. Automation and generative and agentic AI are changing tasks across production, engineering, data, quality and management teams. Pivet points to data literacy, systems thinking, familiarity with automation, the ability to interpret models and cross-functional collaboration as increasingly useful skills. As these capabilities spread, workers may spend less time gathering information and reacting to events and more time interpreting results and improving processes.

Skills gaps remain a challenge: Pivet cites the Global Biopharma Index, which reported severe shortages in digital and AI-related skills at 29% of organizations. He expects more hybrid roles within five years, potentially including digital process engineers, manufacturing data scientists, process intelligence specialists and manufacturing systems architects.

Preparing for those roles means developing skills across the organization, with functional teams learning together rather than assigning digital change to one department. Immersive training using mixed and augmented reality can let operators, technicians and engineers practice with equipment, workflows and troubleshooting scenarios in realistic digital settings. This may help workers build confidence before entering live production, where hands-on practice can be limited by production schedules, equipment access and good manufacturing requirements. Pivet suggests immersive learning could eventually play a role in biopharma training similar to flight simulators in aviation.

Transfer process understanding, not just instructions

Technology transfer can falter even when process documents are complete. Receiving teams may get specifications and procedures without the reasoning behind key parameters or an explanation of how the process responds when conditions change. As therapies grow more complex, that missing context can make it harder to reproduce a process at a new site.

Pivet says effective transfer must carry process understanding along with the written instructions. Digital twins can help teams explore assumptions virtually before trying them in physical production. Used this way, they may help clarify how a process behaves and address uncertainty before it moves into manufacturing.

Connected infrastructure underpins execution

If a manufacturer can prioritize one area, Pivet recommends digital infrastructure. Reliable information flows support technology transfer, workforce development, predictive analysis, process understanding and operational improvement. Without them, organizations may struggle to spread knowledge, repeat successful work or make sound decisions consistently.

His central point is that scientific progress alone does not guarantee practical access to new therapies. The ability to turn that progress into predictable production depends on connected information, visible processes and workers prepared to use digital tools. How well manufacturers build those foundations may shape their ability to scale therapies in the coming decade.

The original article appeared in GEN - Genetic Engineering and Biotechnology News, which is available at GEN.