Manufacturing technology
Regulatory requirements are among the strongest drivers of this shift, and the GAO findings provide a sharper edge: with nearly 4,000 devices recalled in five years and Class I recalls climbing sharply, manufacturers must be able to demonstrate that a device has been produced according to approved specifications – records that paper and disconnected applications make harder to guarantee. “The goal is not only to digitise an existing paper form,” Clerc says. “It is to create reliable data lineage and make traceability part of normal process execution.”
Requirements such as unique device identification reinforce the need to connect device genealogy with quality and supply chain records, so manufacturers know not just that an issue occurred, but which materials, lots, suppliers and markets could be affected. “Regulatory requirements are not simply constraints on digitalisation,” he says. “When addressed through a well-designed architecture, they can help establish the data discipline needed to create wider operational value.”
AI needs the thread 4,000
US Government Accountability Office
46
Growing interest in AI further raises the stakes for data quality. For Zoelfl, the logic is simple: “AI needs data.” But having data isn’t enough – it needs to be trusted, and its context understood. A digital thread supplies that context, giving AI a stronger foundation for quality investigations, non-conformance management and change-impact analysis. “The value of AI does not come from replacing existing systems or adding a chatbot to them,” Zoelfl notes. “It comes from giving AI access to information that is structured, connected and meaningful.” The next stage is likely to be agentic AI – systems that pursue tasks across multiple steps, such as investigating non-conformance or assessing an engineering change. But medical device manufacturing cannot simply optimise for speed, since decisions can affect patient safety, and Zoelfl expects adoption to be cautious, with human oversight remaining important. “The more autonomy manufacturers give AI, the greater the importance of controlling the information environment in which it operates.” Clerc flags assumptions that may unnecessarily slow manufacturers down. The first is that digital transformation requires wholesale replacement of existing equipment, when connecting selected assets and modernising in phases can achieve the same outcome.
The US FDA oversaw the recall of thousands of medical devices between 2020 and 2024.
The second is that connectivity itself creates value – “Collecting more data does not automatically improve a process,” he says – unstructured machine data can generate more complexity without contextualisation and governance.
That point is, in effect, what the Deloitte survey data describes from the other direction: medtech leaders identified leadership alignment around the
value of digital innovation as the single biggest gap holding their organisations back, cited by 65% of respondents, ahead of funding (61%) and strategy (58%). The barrier is rarely the technology itself. A third misconception treats the digital thread as primarily about the manufacturing floor or an IT initiative, when its greatest value comes from connecting functions that have traditionally worked separately. And a final misconception assumes an organisation needs a fully defined end state before it can begin. Clerc’s advice: pick a high-value use case, demonstrate an outcome, then scale. “Connected manufacturing is a maturity journey,” he says, not a one-time installation. For many complex devices – imaging systems, analysers – the relationship with a manufacturer can last a decade or more, generating valuable field-service data about real-world performance. Yet the connection between the pre-market and post-market life cycle is often incomplete. Zoelfl sees real room for improvement here: information generated by operators after shipment doesn’t always flow back to the teams responsible for the product before shipment, even as the industry has made progress in connected service itself. A complete digital thread could connect these two worlds – a practice already proven in parts of the automotive and aerospace industries.
A journey, not a destination Between them, Zoelfl and Clerc describe two halves of the same challenge: connecting the information systems that govern a product’s design and quality story and connecting the physical and operational systems that produce it. Neither requires transforming everything at once. A quality problem, a legacy equipment integration or a supply chain risk can each provide a starting point – what matters is that individual improvements build towards a shared information architecture rather than another isolated digital island.
That is the essential distinction between digitisation and digital transformation. Digitisation makes an existing process electronic, but a digital thread makes the relationships between processes visible, contextualised and usable. This distinction will matter increasingly as manufacturers deploy AI and, eventually, autonomous agents.
Legacy systems will remain, suppliers will keep operating different platforms, and regulatory requirements will keep demanding control and evidence – and, if the GAO’s recent findings are any guide, that pressure is unlikely to ease. The digital thread isn’t just a technology upgrade; it’s a fundamental mindset shift for manufacturers. It moves organisations from isolated records to connected knowledge, from documenting quality to building it in, and from explaining past failures to anticipating future outcomes. The ultimate prize is the ability to understand the complete story of a product, and to use that understanding to make better decisions. ●
www.medicaldevice-developments.com
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50 |
Page 51 |
Page 52 |
Page 53 |
Page 54 |
Page 55 |
Page 56 |
Page 57 |
Page 58 |
Page 59 |
Page 60 |
Page 61 |
Page 62 |
Page 63 |
Page 64 |
Page 65 |
Page 66 |
Page 67 |
Page 68 |
Page 69 |
Page 70 |
Page 71 |
Page 72 |
Page 73 |
Page 74 |
Page 75 |
Page 76 |
Page 77 |
Page 78 |
Page 79 |
Page 80 |
Page 81 |
Page 82 |
Page 83 |
Page 84 |
Page 85 |
Page 86 |
Page 87 |
Page 88 |
Page 89 |
Page 90 |
Page 91 |
Page 92 |
Page 93 |
Page 94 |
Page 95 |
Page 96 |
Page 97 |
Page 98 |
Page 99 |
Page 100 |
Page 101 |
Page 102
