This article, writes Darragh McLoughlin, explores the development of an automated catheter production line using digital twin technology, robotics, and SCADA integration. The project demonstrates how simulation-driven design can improve precision, compliance, and scalability in medical device manufacturing.

Introduction

Catheter manufacturing is a cornerstone of modern healthcare, enabling minimally invasive procedures that improve patient outcomes. Despite its critical role, much of the production process remains manual – introducing inefficiencies, variability, and challenges in scaling. These limitations are becoming increasingly problematic as global demand rises, driven by ageing populations and the prevalence of chronic illnesses. Manual assembly, while effective for small batches, struggles to maintain consistency and throughput in high-volume environments. 

Side by side comparison for a digital twin.

Automation offers a compelling solution. By integrating robotics, machine vision, and advanced control systems, manufacturers can achieve repeatable precision, reduce defect rates, and ensure compliance with stringent regulatory frameworks such as ISO 13485 and FDA 21 CFR Part 11. This article explores a simulation-driven approach to automating catheter production, developed as part of an honours degree in automation and robotics at the Technological University of the Shannon.

The project demonstrates how digital twin technology can be leveraged to design, simulate, and validate an automated catheter production line before physical deployment, reducing risk and accelerating implementation.

RoboDK workflow layout.

Industry context

The medical device industry operates under some of the most rigorous quality and safety standards in manufacturing. Catheters, used in cardiovascular interventions, drug delivery, and diagnostic procedures, require exceptional precision and sterility. Any deviation in assembly can compromise patient safety and regulatory compliance, making process control paramount.

Historically, catheter production has relied on manual labour, which introduces variability and limits scalability. Operators must handle delicate, flexible materials, often under cleanroom conditions, which increases labour costs and fatigue-related errors. As demand grows, these limitations have prompted manufacturers to explore Industry 4.0 technologies – robotics, AI-driven vision systems, and IoT-enabled monitoring – to create smart factories capable of predictive maintenance, real-time analytics, and adaptive control. 

SolidWorks CAD model of full cutting assembly.

Companies such as Medtronic and Abbott have already embraced automation strategies, integrating collaborative robots and vision systems to improve throughput and traceability. However, widespread adoption remains slow due to high initial costs, technical complexity, and the need for extensive validation under Good Automated Manufacturing Practice (GAMP 5). This project addresses these challenges by presenting a modular, simulation-based approach that reduces commissioning time and supports compliance from the outset.

System architecture

At the heart of the proposed system is a Universal Robots UR5e collaborative robot, selected for its compact footprint, integrated force/torque sensors, and ease of programming. The UR5e performs critical tasks such as tube placement, cutting, and packaging, replacing manual handling with precise, repeatable motion. Its collaborative design allows safe operation alongside human workers, making it suitable for hybrid workflows in cleanroom environments.

Final placement of catheter into packaging box.

The robotic system is supported by a Siemens S7-1500 PLC, which acts as the central controller for process coordination. Using Siemens TIA Portal, ladder logic was developed to manage conveyor sequencing, robotic triggers, and safety interlocks. Communication between devices is handled via Profinet, ensuring deterministic data exchange and synchronisation across the system. This architecture enables real-time control and scalability, allowing additional robots or inspection stations to be integrated without major redesign.

Digital twin simulation

One of the most transformative aspects of this project was the use of digital twin technology to validate and optimise the catheter manufacturing process before physical deployment. A digital twin is a virtual representation of the production environment, enabling engineers to simulate workflows, predict performance, and identify potential issues without incurring the cost or risk of real-world trials.

Using RoboDK, the UR5e robot’s kinematics were modelled alongside CAD representations of the cutting station and conveyor system. This allowed for collision detection, reachability analysis, and cycle time estimation under realistic constraints. The simulation environment mirrored the robot’s payload limits, joint configurations, and tool centre point calibrations, ensuring that virtual paths could be translated seamlessly into executable programs. 

SCADA incorporating the PLC, HMI, UR5e, Conveyor, Cognex Camera and Power BI used in this project.

Beyond motion planning, the digital twin facilitated predictive validation of mechanical forces. MATLAB was employed to estimate torque requirements during cutting operations, confirming that the UR5e’s integrated force sensors could maintain safe handling pressures. These simulations demonstrated that the system could operate within a torque range of 4.4-6.6 Nm, even less than ±2mm positional offsets, reducing the risk of material damage and ensuring compliance with ISO 13485 and GAMP 5 standards.

The benefits of this approach extend beyond design optimisation. By validating workflows virtually, manufacturers can reduce commissioning time, minimise downtime during integration, and accelerate regulatory approval through documented simulation results. This capability is particularly valuable in medical device manufacturing, where process validation and traceability are critical for compliance.

Vision system integration

Quality assurance in catheter manufacturing is non-negotiable. Even minor defects – such as microcracks, contamination, or misaligned connectors – can compromise product integrity and patient safety. To address this, the system incorporated a Cognex In-Sight 3800 vision sensor configured for high-speed, high-resolution inspection.

The vision system employed edge detection algorithms to verify dimensional accuracy and alignment, while AI-based defect classification (Cognex SnAPP) enhanced detection reliability. Unlike traditional rule-based inspection, SnAPP uses adaptive learning to improve accuracy over time, reducing false positives and negatives. During simulation, the system achieved a defect detection accuracy exceeding 97%, with real-time pass/fail decisions communicated directly to the Siemens PLC. 

RoboDK – path trajectory and optimisation.

This integration enabled automated rejection of defective units and supported full traceability through SCADA dashboards. By eliminating manual inspection, the system not only improved consistency but also reduced labour costs and operator fatigue. Furthermore, the adaptability of the vision system allows inspection of transparent and semi-rigid materials, which are common in catheter production.

SCADA monitoring and data analytics

To replicate industrial monitoring, a SCADA-style dashboard was developed using Siemens WinCC and Power BI. WinCC handled real-time control and alarm management, while Power BI provided advanced analytics on cycle times, defect rates, and operator interactions. Simulated datasets were used to emulate live production, demonstrating how cloud-connected dashboards can support predictive maintenance and regulatory reporting.

Key performance indicators such as Overall Equipment Effectiveness (OEE), downtime trends, and defect analysis were visualised through interactive charts, enabling data-driven decision-making. This dual-layer monitoring approach enhanced operational transparency and supported compliance with ISO and FDA standards. By integrating SCADA with cloud analytics, the system lays the foundation for smart manufacturing environments where real-time insights drive continuous improvement.

Regulatory compliance

Medical device manufacturing operates under some of the most stringent regulatory frameworks in industry. For catheter production, compliance with ISO 13485, Good Manufacturing Practice (GMP), and GAMP 5 is essential. These standards mandate risk-based process validation, traceability, and documented control of every step in production.

The automated system was designed with these requirements in mind. PLC logic and vision algorithms were structured for validation, ensuring repeatable execution and accurate defect detection. SCADA dashboards supported electronic batch records and audit trails, aligning with FDA 21 CFR Part 11 requirements for electronic data integrity. Packaging automation incorporated Unique Device Identification (UDI) and label verification, meeting FDA 21 CFR Part 801 standards for product tracking.

By embedding compliance into the system architecture, the project demonstrates how automation can simplify regulatory adherence. Automated data logging, alarm handling, and audit trail generation reduce the burden of manual documentation and support faster approval during audits. This approach reflects best practices in Industry 4.0-enabled medical device manufacturing, where compliance is integrated rather than retrofitted.

Performance and validation

Simulation results confirmed the feasibility and benefits of the automated system. The average cycle time per catheter was approximately 60 seconds – three times faster than manual assembly, which typically requires 180 seconds per unit. Defect rates dropped from 8.5% to 2%, and daily output increased from 160 to 480 units in an eight-hour shift.

These improvements were attributed to synchronised robotic handling, automated inspection, and real-time control logic. MATLAB torque simulations validated the mechanical design of the cutting station, confirming safe handling forces and alignment tolerances. Vision system testing achieved a defect detection accuracy exceeding 97%, with reliable identification of cracks, misalignments, and contamination.

Although the project was developed in a simulated environment, the results mirror performance benchmarks reported by leading manufacturers. The integration of digital twin technology, AI-driven vision inspection, and SCADA monitoring demonstrates a scalable model for catheter production that can transition smoothly into physical deployment.

Environmental and economic impact

Automation in catheter manufacturing offers significant environmental and economic advantages. By reducing material waste and energy consumption, the system supports compliance with EPA regulations and Ireland’s Circular Economy Action Plan. SCADA and IoT-enabled monitoring facilitate real-time energy tracking, helping facilities meet ISO 50001 energy management standards.

From an economic perspective, the system’s modular design allows for phased deployment, making it adaptable for companies with budget constraints. The estimated cost of hardware and software components – including the UR5e robot, Siemens PLC, Cognex vision system, and simulation licences – was approximately €57,000. While this represents a substantial initial investment, long-term savings from reduced labour, improved quality, and predictive maintenance support a strong return on investment.

Automation also enhances sustainability by enabling predictive maintenance, which extends equipment lifespan and reduces unplanned downtime. These benefits align with global trends towards greener manufacturing practices and cost-efficient production strategies.

Lessons learnt

Developing this system provided valuable insights into the complexities of automating medical device manufacturing. One of the most significant lessons was the importance of cross-disciplinary expertise. Successful implementation required knowledge of robotics, PLC programming, mechanical design, vision system calibration, and regulatory compliance. Each subsystem had unique challenges – from tuning Cognex vision algorithms for transparent materials to validating torque requirements in MATLAB for delicate cutting operations.

The project also highlighted the critical role of simulation in reducing deployment risk. Digital twin technology allowed engineers to identify bottlenecks, optimise robotic paths, and validate control logic before physical integration. This approach not only saved time and resources but also ensured that compliance requirements were addressed early in the design phase.

Another key takeaway was the value of modular architecture. By designing the system as a collection of interoperable components – robotics, vision, PLC, SCADA – the solution remains adaptable for future upgrades. This flexibility is essential in an industry where product variants and regulatory standards evolve rapidly.

Outlook

As Industry 4.0 technologies continue to mature, the future of catheter manufacturing lies in fully connected smart factories. Digital twins will evolve from static simulations into dynamic models that integrate live sensor data, enabling real-time optimisation and predictive maintenance. AI-driven vision systems will become more sophisticated, capable of detecting micro-defects and adapting to new product geometries without manual reprogramming.

IoT-enabled sensors will enhance traceability and environmental monitoring, supporting compliance with sustainability regulations and energy management standards. Cloud-based SCADA platforms will provide remote access to production data, enabling global manufacturers to monitor multiple facilities from a single dashboard.

Lean and Six Sigma methodologies will remain central to continuous improvement, ensuring that automation delivers not only speed and precision but also cost efficiency and waste reduction. Ultimately, the convergence of robotics, AI, and digital simulation will redefine medical device manufacturing, making processes more resilient, scalable, and compliant.

Conclusion

This project demonstrates the feasibility and advantages of automating catheter manufacturing through a simulation-driven approach. By integrating collaborative robotics, AI-powered vision systems, and SCADA-style monitoring within a digital twin framework, the system achieves significant improvements in cycle time, defect reduction, and regulatory compliance.

Although developed in a simulated environment, the architecture mirrors industry best practices and aligns with ISO 13485, GAMP 5, and FDA 21 CFR Part 11 standards. The results – tripling output, reducing defects by more than 75%, and enabling predictive analytics – underscore the transformative potential of automation in the medical device sector.

As demand for high-precision, cost-effective healthcare solutions grows, automation will become indispensable. This project provides a blueprint for manufacturers seeking to transition from manual workflows to smart, scalable production systems that meet the challenges of tomorrow.

Additional resources

Additional resources for this project are available via a Linktree link, which includes access to the SolidWorks CAD models of the cutting assembly, interactive Power BI dashboards for SCADA-style monitoring, and a working simulation of the digital twin developed in RoboDK. https://linktr.ee/DarraghMcL?utm_source=qr_code LinkedIn (3) Darragh McLoughlin | LinkedIn

Author: Darragh McLoughlin is a graduate electrical engineer, currently working on automation and control projects in the semiconductor industry. He recently completed an honours bachelor’s degree (BEng) in Automation and Robotics, where his thesis focused on digital twin technology and smart manufacturing for the medical device sector.