Advancing Automation in Medical Device Manufacturing

Automation drives efficient medical device manufacturing operations by increasing precision, speeding up assembly, and reducing costs for complex assembly tasks. Manufacturing equipment can perform micro-assembly tasks with exact tolerance levels. Automated lines can run continuously to meet high market demands. Automation technologies enable high-precision assembly of miniaturized components and medical devices, reducing the margin of error and improving overall product quality.

However, even with these proven advantages, fully automated production remains the exception, rather than the rule. For example, many disposable medical devices still require manual operations because of their size, flexibility, and product variation. Therefore, medical device manufacturers (MDMs) and their contract manufacturers (CMs) focus on “automating those operations that have the greatest impact on quality, repeatability, and throughput, while maintaining operator involvement where engineering judgment adds value,” said Miguel Ballesteros, senior technical operations manager for Ormond, Fla.-based Command Medical Products, a contract manufacturing organization specializing in single-use fluid management devices for the infusion, general surgical, and lab/diagnostic markets.

This is often accomplished through the integration of manufacturing processes, rather than adding individual robots. Assembly equipment is now expected to perform the operation, verify the result, and electronically record the process data. “Machine vision, automated functional testing, and electronic traceability are becoming standard features instead of optional improvements,” said Ballesteros. “This allows quality to be verified throughout the manufacturing process, rather than only during final inspection.”

Another important trend in automation/assembly is the increased emphasis on supply chain risk mitigation. MDMs are increasingly looking to dual-source critical raw materials and components to improve supply chain resilience. Although multiple approved sources reduce supply risk, these materials still need to be evaluated and validated for the manufacturing process. Components can meet the same specification but behave differently during automated assembly because of differences in dimensions, material properties, or processing characteristics.

“As a result,” said Ballesteros, “automation systems are being designed with greater flexibility, sensing, and process monitoring so they can accommodate normal supplier-to-supplier variations within a validated process window. This creates a much stronger connection between supply chain strategy, process validation, and automation design.”

Because of these ongoing challenges, MDMs are now taking a tougher look at the return on investment (ROI) on automation. With labor costs being somewhat competitive with complex automation implementation in finished device assembly, “automation in material prep, inspection, packaging, and labeling is becoming more attractive by the day,” noted David Pascutti, vice president of business development for Youngsville, N.C.-based Robling Medical, a contract development and manufacturing organization (CDMO) specializing in cleanroom assembly and packaging of Class II and Class III sterile disposables, implants, and electromechanical assemblies. “These systems can be utilized across multiple product lines, making the ROI more attractive for manufacturers working on many different devices.”

While traditional ROI models focus heavily on direct labor savings, “manufacturers are increasingly considering capacity, quality, supply chain resilience, and the improved economics of capital investment resulting from recent U.S. tax changes,” added Ballesteros. “For example, one of the most significant changes is the restoration of 100% bonus depreciation for qualifying equipment acquired and placed in service after January 19, 2025. Before the change, the deduction was scheduled to decline to 20% in 2026 and then expire. Section 179 limits have also increased, allowing more qualifying capital investment to be deducted immediately. This does not make automation less expensive or replace the need for a sound business case, but it can improve near-term cash flow and after-tax payback.”

Current Trends

More MDMs are realizing that automation not only offers better consistency and efficiency than human operators at specific tasks, “it also helps reduce contamination and can provide traceable inspection data at higher speeds,” said Al Neumann, automated manufacturing systems manager for SMC, a Somerset, Wis.-based medical device contract manufacturer with worldwide locations, serving global customers.

MDMs want be the first to market with new products, but are also finding that, because of ongoing labor shortages and skills gaps, hiring CMs is often a more cost-effective way to manufacture products, especially for larger organizations. “The trend for CMs is a value-add—for example, a company provides a raw-material part, such as a molded part, and also becomes the partner that does the full assembly,” said Dr. Brian Romano, director of technology development for Bristol, Conn.-based Arthur G. Russell Company, a provider of custom automated assembly and packaging equipment. “While this may shift production costs to CMs, it allows medical device companies to leverage their assets to adapt to a changing marketplace and evolving technology.”

With the ongoing integration of technology into medical devices, MDMs are finding their products must be fast to market, support the product’s lifecycle, and be ready to adapt as the technology evolves. “Over the last decade, medical devices have moved from a 20- or 30-year product lifecycle to one now approaching five years,” said Dr. Romano. “This means that amortization of capital investment is now spread over a much shorter time period, with the acknowledgment that the automation will need to be retooled or replaced. This changes the landscape of the type, design, and lifespan of machinery.”

Not only do MDMs want to shorten their time to market, but they must do so in a way that reduces risk. CMs that can provide proof-of-concept across manual, semi-automated, and full-scale automation are in high demand as manufacturing partners. Given the shorter product lifecycle and product variability, modularity and scalability are better choices than fixed-function, high-volume machines. However, there are still markets and products where high-volume machinery is the most effective choice.

Product trends include a shift toward increasingly complex catheter technologies, including steerable catheters, smart catheters with integrated electronic sensors, and advanced devices for electrically based ablation procedures. “These innovations are driving demand for greater manufacturing precision, tighter process control, and more sophisticated production capabilities,” said Dave McMorrow, technical director for MMT Automation, the automation and integration arm of Medical Manufacturing Technologies (MMT), a provider of end-to-end automation and integration solutions to the medical device manufacturing industry. 

Other hot markets are microfluidics and miniaturized devices. Microfluidics involves precisely controlling very small amounts of liquid within tiny channels and structures. It is increasingly used in applications such as diagnostic testing, drug delivery, infusion devices, and wearable or implantable sensors, where precise control of fluids can improve performance, speed, and automation. “Although we have been working with these technologies for several years, we are seeing customers explore new applications and approaches, including microfluidics for in vitro diagnostic reagent handling,” said Karrie Ann Khattab, manufacturing engineering manager for Hudson, Wis.-based Phillips Medisize, a CDMO serving pharmaceutical, medtech, and in vitro diagnostic companies.

From product launches to M&A and manufacturing trends, the MPO Source delivers the medtech news you need twice a week. Subscribe here!

What MDMs Want

Cost reduction is the top concern for MDMs. They expect their CMs to have the knowledge and creative expertise to reduce costs, maintain quality, and maximize speed to market.

“While MDMs champion themselves on acquiring/developing the technology and building a market for it, they look to their CMs to provide the solutions that drive out cost, without sacrificing quality,” said Pascutti.

MDMs continue to push for greater consistency, higher throughput, and complete traceability. They also expect CMs to participate earlier in product development and recommend manufacturing methods, automation strategies, and validation approaches. 

Other top items on the MDM wish list are accuracy, reconfigurability, data logging, and traceability.

“Flexibility is also a priority,” said Ballesteros. “Customers want manufacturing systems that can support multiple products and expand as demand increases, without requiring an entirely new production line.”

The rise of smart and electrically enabled devices is driving demand for manufacturing equipment that can reliably process miniature electronic components and flexible conductors without compromising precision or throughput. “Integrating electrical testing and instrumentation directly into the manufacturing process is becoming increasingly important to ensure device performance, quality, and traceability,” said Morrow. 

Key to cost reduction are production processes that are designed with efficiency in mind—such as integrating process steps, combined with repeatably controlled process execution. “Ultimately, MDMs want processes that are designed with future increased automation and scalability in mind,” added Michael Wall, technical director for Somex Automation, an MMT company in Cork, Ireland, that provides custom automation for the medical and pharma industries.

Advanced Assembly and Automation

The shift toward AI-driven platforms versus traditional pre-coded machines is gaining momentum. As machines learn and adapt to product lines, they become less dependent on their out-of-the-box programming. This has especially been evident in systems using vision controls for movement, handling, or even inspections. For example, advances in robotics, machine vision, and precision feeding systems enable the automated assembly of increasingly miniature and complex catheter components. At the same time, enhanced control systems support closed-loop process control with integrated vision inspection and in-line testing, helping manufacturers improve repeatability and production efficiency. 

“AI is increasingly used for design and development of automation, including early-stage visualization and evaluation of solutions, control, and inspection systems, which reduces risk and lead times,” said Wall.

Ballesteros believes the greatest advancements so far have come from improvements in machine vision, collaborative robotics, servo motion control, and production software. AI is beginning to improve automated inspection by identifying defects that are difficult to detect using conventional vision systems. 

“These Internet of Things technologies have great potential for improving process monitoring and defect detection, as well as making automation more flexible, easier to validate, and less dependent on custom programming,” said Ballesteros.

As medical devices get smaller, components can become more difficult to handle, position, and assemble accurately. Smaller parts may require tighter tolerances and greater precision, creating challenges for both people assembling the devices and automated equipment that must reliably handle and position tiny components.

At the same time, “miniaturization is driving manufacturers to leverage newer assembly technologies and techniques, including different techniques for welding and adhesive dispensing,” said Jordan Bartz, manufacturing development manager for Phillips Medisize. “These technologies help us build smaller, more sophisticated devices while maintaining the quality and reliability required for medical applications.”

Innovative Design and Build

Successful equipment development typically requires collaboration with experienced vendors that have deep expertise in both the product and manufacturing processes. Combined with disciplined risk management and a robust quality management system, this approach helps deliver innovative, reliable equipment on time. “Conducting proof-of-concept testing prior to automation scale-up is a critical step in validating process robustness, reducing technical risk, and accelerating commercialization,” said McMorrow.

“Along the same lines, early engagement at R&D and process development stages provides equipment-based core manufacturing processes that are stable and can be scaled readily to ramp up output without delay, such as re-validation or product design changes,” said Wall.

Modularity is incredibly important for manufacturers working on multiple product lines because it allows for a single piece of equipment to potentially support many lines by changing out of specific modules. “We prefer modular manufacturing cells that can begin as semi-automated processes and expand by adding automated feeding, inspection, or testing as production volumes increase,” said Ballesteros. “This approach reduces initial investment, simplifies validation, and provides greater flexibility as products evolve.”

Since many of SMC’s projects involve products in early stages of development, its work cells are designed with modularity in mind. “Rather than building strictly to initial product requirements, we consider components and testing techniques that will carry into the next assembly phase,” said Neumann. “Lessons learned early in mechanical design and programming lead to more robust and economical cells down the line. For example, substituting linear motors for pneumatic actuators in a Phase-1 build provides tighter control and real-time position feedback—data that pneumatic actuators typically cannot supply—and lets programmers write code that carries forward into subsequent phases.”

3D printing is still an essential technology for creating early prototypes of components that are tested to see how well they function, before investing in expensive production tooling. Although 3D printing is typically not used for high-volume production, there is growing interest in using it for low-volume production and early commercial builds, particularly when speed and flexibility are higher priority than large-volume production.

“We are seeing the most value from 3D printing during the early stages of product and manufacturing development,” said Khattab. “For example, we can 3D-print fixtures and use them to quickly test and refine how a device will be assembled. Because these tools can be produced and modified quickly, manufacturers can try different approaches, identify what works and what doesn’t, and improve the process before moving on to larger-scale production.”

AI Is a Must

AI and related technologies are entering the medical device industry at all levels. For example, large language models can help with business and administrative tasks. Machine learning is often used for anomaly detection and predictive maintenance. Agentic AI can also assist in writing code. “Each of these applications has the potential to save time and money, especially in an era of workforce shortages and skills gaps,” said Dr. Romano.

AI’s primary impact on automation is machine learning versus static outputs from initial programming. AI also improves process monitoring, predictive maintenance, and machine vision. Rather than replacing operators and techs, AI will provide earlier identification of process drift and recommend corrective actions before defects occur.

“AI is used frequently to review code,” said Newmann. “When 5,000 or more lines of code are downloaded for a modular workcell, there will be inefficiencies or mistakes. AI has uncovered dead code, hard-to-find edge cases, inefficient programming, logic errors, and even makes style-change suggestions. Although AI code currently still requires rework, there is a definite advantage to using it. The better we prompt, the better the results.”

AI is also driving the rapid development of simulations/digital twins to improve automation and assembly. For example, simulation is becoming an effective tool for equipment development. It allows engineers to evaluate production flow, cycle time, equipment utilization, and facility layout before capital equipment is purchased. “Digital twins represent the next step by combining simulation with real production data,” said Ballesteros.

Simulations have significant potential for improving process optimization, predictive maintenance, and capacity planning. For example, stochastic discrete event simulation/modeling can prove out processes, queues, and bottlenecks and leverage statistical information from collected historical data. This can be done offline and show multiple years’ worth of production in minutes on a computer. Physics-based models leverage the physical design of production equipment to provide a simulation environment. There are methods to link the operating code, either actual programmable logic controller code or pseudo code, to provide the functional logic. This approach creates a digital twin that truly mirrors the equipment’s design and operation and is driven by historical data. 

“Changes can be demonstrated in this offline environment to provide input and guidance to hypothesized changes without ever rewiring or cutting metal,” said Dr. Romano. “However, in both the discrete-event and physics models, the power comes when live Internet of Things data from plant floor equipment is routed to the models in the simulation environment, resulting in a true digital twin. Changes made to the model can now be tested with live data when it is brought back online. This methodology also provides insights into production issues and indicates where maintenance may be required before it is needed on the floor.”

However, noted Dr. Romano, even though machines on the floor today can be new and state-of-the-art, they are often legacy equipment. “Just today, we received a request to possibly update a 22-year-old machine to connect it to a data acquisition system, bringing it into the system with the intent to leverage AI and other data mining techniques to provide insights into the production data,” he said. “Another issue related to this is that the machine data is not always consistently structured or clean; for instance, one machine may use imperial units while others use metric units. The quantity or frequency of the data may be different. This may require some extra effort to impute and align the data so it can be mined.”

Considering the operational complexity of IoT applications, MDMs often prefer to work with experienced CMs to develop robust, scalable manufacturing solutions for both established and next-gen and established technologies. Engaging early in the product development process—before design freeze—allows manufacturability considerations to be incorporated into the device design, resulting in more robust, higher-yield manufacturing processes. “Early collaboration also accelerates process and equipment development, helping MDMs reduce development timelines, mitigate risk, and achieve a faster time to market in an increasingly competitive industry,” said McMorrow.

Moving Forward

Future manufacturing systems will be increasingly connected and adaptive. Advances in machine vision, force monitoring, servo controls, and process analytics will allow equipment not only to perform an assembly operation, but also to recognize variation in the components being processed and verify that the assembly remains within a validated process window.

“This is especially important when the same raw material or component comes from multiple approved suppliers,” said Ballesteros. “Two components can both meet the drawing specification but behave differently during automated assembly because of differences in dimensions, stiffness, friction, or other material characteristics. Newer automation methods can measure these differences during the process, monitor trends by supplier or lot, and identify process drift before it results in nonconforming product.”

AI seems to evolve daily in the breadth of its applications.

“As manufacturers, we continue to learn how AI can help deliver a quality product sooner, with fewer complications,” said Neumann. “Where we once wondered what a new technology would look like in a couple of years, we now wonder what artificial intelligence will look like in a couple of weeks.”

“Technology is advancing rapidly, but successful implementation still depends on developing stable manufacturing processes before automation is introduced,” added Ballesteros.

A limiting factor is the ongoing workforce shortage and skills gap driven by COVID (retirement of experienced workers) and a lack of new workers in the field who can develop and support automation and assembly systems. Because medical device and diagnostic manufacturing companies are finding it difficult to maintain adequate engineering and technical staffing, “they are asking the OEM of the assembly equipment to play a greater role in the ongoing maintenance of machinery and the training of personnel,” said Dr. Romano. “This makes continuity on the production floor more difficult for manufacturers.”

“It is an exciting time to be involved in the automation field,” said Neumann. “Advances such as AI-assisted code review, closed-loop motion control with linear motors, and increasingly modular workcell design are reshaping how quickly and economically we can move from early-stage builds to full production.”

“For me,” said Ballesteros, “the most exciting advancement is the integration of manufacturing equipment, inspection systems, and production data into a single controlled process. Instead of detecting problems after production, manufacturers can identify process variation as it occurs and respond before nonconforming product is produced. This shift toward real-time process control will have a greater impact on quality and manufacturing efficiency than any individual automation technology.”

Mark Crawford is a full-time freelance business and marketing/communications writer based in Corrales, N.M. His clients range from startups to global manufacturing leaders. He has written for MPO and ODT magazines for more than 15 years and is the author of five books.