Which type of control system is best for personalised medicine?

(See Beckhoff at MachineBuilding.Live, 14 October 2026, on stand 214)

At present, many labs remain highly manual, even where individual instruments or discrete processes have been automated. Samples may still be moved between instruments by technicians, with data captured at different points along the way. As treatments and formulations become more individualised, the challenge is to automate the complete workflow while recording what happens at each stage.

This is not simply an issue for the future. One useful indicator of the growth of this trend comes from the UK's advanced therapy sector. Although advanced therapies represent just one part of the personalised medicine landscape, they show why flexible, traceable manufacturing is becoming more important. The Cell and Gene Therapy Catapult recorded 193 ongoing advanced therapy clinical trials in the UK in 2025, including cell and gene therapies that can place pressure on workflow automation and manufacturing flexibility.

Regulation is beginning to reflect this shift. The UK's point-of-care and modular manufacture framework recognises that some medicines may need to be produced on demand or closer to the patient. For machine builders and end users, the challenge is to create systems that can adapt as the science changes.

From fixed sequences to variable workflows

Traditional pharmaceutical production is often built around a known formulation and repeatable process. Once the production parameters are established, the automation system can be designed around consistency and repeatability. That model remains essential for many medicines, but it does not reflect the full complexity of personalised medicine or automated drug discovery.

In these environments, the route through the lab may vary according to the sample, the patient or the test being carried out. As more of this work is automated, movement, measurement and data capture need to be controlled by software rather than managed through manual intervention.

Personalised medicine complicates this because not every sample will follow the same path. One sample may need to visit one sequence of instruments, while another may require a different route entirely. If that sequence is hard-coded into a fixed automation system, every change risks becoming a mechanical, electrical or software engineering project.

Flexibility still needs proof

Personalised medicine may involve high-value materials, short shelf-life products or samples linked to a specific patient pathway. In that environment, traceability is not an administrative function added at the end, but something that must be built into the process from the start.

For more individualised processes, validation increasingly depends on monitoring and recording what happens to each sample or product as it moves through the system. This can include scanning a sample ID, logging the instrument used and associating inspection results with the correct product or patient pathway.

For end users, the practical risk is that individual tasks are automated, but the wider process remains fragmented. A liquid handler, robot or analytical instrument may perform its own function well, yet become difficult to integrate into a broader automated workflow if each device sits in its own technical island. The challenge is not simply to create more automated labs. It is to create connected labs, where the workflow, the physical process and the data record remain aligned.

The integration problem

Lab automation is inherently multi-vendor. No single automation supplier will manufacture every device used in a modern life sciences facility. Machine builders therefore need control systems that can communicate with third-party equipment, connect to higher-level software and pass data into databases or cloud systems without creating unnecessary complexity.

This is where PC-based control offers a useful approach. An industrial PC can bring machine control, motion, visualisation, measurement and software connectivity into a common environment. High-speed industrial networking allows data to move quickly from sensors and I/O devices, while PLC, motion control and additional software functions can run on the same architecture.

The point is not that conventional PLCs cannot control laboratory equipment. They can, and they remain widely used across industry. The issue is how much additional architecture is required when the same system also needs to handle data, integrate external software and adapt to changing workflows.

With PC-based control, there is less separation between the machine and the software environment around it. This makes it easier to connect the automation layer to data systems, machine learning tools or higher-level workflow software.

Transport is another part of the same discussion. In a fixed line, samples move through a predetermined sequence. In a more flexible laboratory, the transport system may need to support non-linear workflows. Linear transport systems with independently controlled movers, such as XTS, and planar motor systems with contactless movers, such as XPlanar, can support movement where products or carriers do not need to follow a conventional conveyor route.

Designing for change

For machine builders, the message is to design around adaptability from the outset. A machine may meet today's process requirement, but its long-term value will depend on how easily it can accept a new instrument, update a workflow or connect to a different data platform.

For end users, the priority is to avoid building automated islands. The laboratory of the future will not simply be a collection of faster machines. It will be a connected environment in which instruments, software and transport systems work together to support changing scientific and manufacturing requirements.

Personalised medicine makes that shift unavoidable. The priority is not simply to automate more of the lab, it is to build automation around change. The winning systems will be those that can adapt workflows, connect instruments and preserve traceability without requiring a complete redesign each time the science moves on.