Michael Barbella, Managing Editor12.10.21
The timing couldn’t have been better.
Three years ago, NuTec Tooling Systems Inc. began building a syringe coating machine for a large pharmaceutical firm. The client wanted to mass produce plastic syringes with a glass-like coating to provide an alternative to the more costly glass versions typically manufactured by its competitors.
The machine NuTec constructed included four Epson Cleanroom SCARA robots (developed by Epson Robots), each of which were strategically placed at various points in the apparatus to precisely and cost-effectively automate the syringe manufacturing process.
The automated process coats syringes at a rate of 38 parts per minute, passes the parts through various inspection stations, then siliconizes the hypodermics before changing temporary caps to final caps and subjecting them to a final X-ray inspection. The robots handle the syringes both before and after their glass-like coating is applied; in the final stages, the robots apply inner and outer covers to full containers of syringes and applies labels with a laser marker.
The machine was operational in November 2020, enabling NuTec’s pharmaceutical customer to work with the government and manufacture massive quantities of syringes for use in battling COVID-19.
“Epson’s high-speed G6-Series SCARA robots with Epson RC+ software enable precision processes with exceptional repeatability assembly pick and place capabilities,” Brent Martz, NuTec Tooling Systems sales and marketing director, said in an Epson news release. “The ease of use and application versatility within the Epson RC+ development environment plus an ISO-3 rating and compliance with cleanroom standards makes them ideal for this project and the medical sector in general, where speed and precision are vital to the manufacturing process.” Bürkert
Indeed, speed and precision are vital to medtech manufacturing, particularly as the types of medical devices that can be automatically assembled are constantly expanding. MPO’s feature “Complex Coupling” details the trends and market forces impacting the medtech assembly/automation sector. Prasad Akella, founder and chairman of Drishti Technologies Inc., was among the more than half-dozen experts interviewed for the feature; his full input is provided in the following Q&A:
Michael Barbella: What current trends are shaping the medical device assembly and automation sector?
Prasad Akella: Perhaps the biggest trend is that medical devices are getting increasingly more capable, and consequently, complex. If we layer in the fact that medical device manufacturing is highly regulated, it’s natural to conclude that continuously improving the most fundamental practices—current good manufacturing practices (cGMP)—becomes even more critical to ensure compliance and the safety of these sophisticated products. For example, consider a simple device like a glucose monitor with the patch that constantly transmits sugar levels to one’s smartphone and, very likely, a cloud service. Or the latest and greatest in heart valves: These products are connected to the internet and often use advanced materials with just the right physical properties to keep us all healthy. Manufacturing them well requires deep skills.
Now, when one looks at the competitive field, medical device manufacturers are realizing they have to out-compete and out-manufacture the competition. The cGMP requirements from yesterday aren't good enough for today and certainly not for tomorrow. These manufacturers have to focus on the “current” in the cGMP and go beyond the regulatory requirements to develop even better manufacturing practices that increase efficiency, quality, safety and ultimately revenue.
Barbella: What factors are driving the need for automation in medical devices?
Akella: The need for automation in medical devices is driven by the same factors all of manufacturing is driven by—cost, volume, speed, quality and personalization. Automation is inexorable!
That said, I believe that there is a fundamental underlying assumption in your question: That the primary path to scale is automation, since we have maximized what we can do with manual processes. I flag this as being especially important since every medical device manufacturing line I have seen—from blood pressure monitors to heart valves to MRI machines—is manned by people. Because people can easily perform tasks that a piece of automation might be challenged to do, certainly at a viable price point.
So, in addition to automation, the bigger need is to drive a lean culture and impact the 70-plus percent of your lines that are manual. To provide them with the tools and processes to build these medical devices safely and at scale—while simultaneously empowering them to make better and quicker decisions. Tools and processes that augment humans and help them perform manual assembly tasks to the best of their ability. This could be augmented reality, poka yoke systems or, in Drishti’s case, data, analytics and insights from AI and computer vision systems.
Barbella: What new innovations have been developed within the medical device assembly and automation space? What specific market needs do these innovations address?
Akella: Significant energy is being focused on increasing quality, gaining visibility into operations, and supporting decision-making at all levels, and for good reason. The risk to end users is massive, and the fines and penalties that may be levied against an organization with defective products are hefty.
A new advancement in this area is computer vision and AI (“AI-on-video,” if you will), which can help detect defects as they occur, rather than wait for an end-of-line inspector to catch and flag them. The economic value stems from the fact that the longer a defective unit goes down the line, the more value is being added to a fundamentally useless unit. This new inspection methodology can, therefore, avoid significant scrap and rework, which is expensive for manufacturers. It also alleviates the heavy burden on the inspector to catch the defect, because having prevention mechanisms in place upstream increases the likelihood that problems will be caught earlier. This system also helps manufacturers identify opportunities to train the team, either in the moment with feedback (think of your spell checker running behind the scenes in your email client) or in a training station (think of your golf swing being videotaped for you to learn from after the fact).
Barbella: How is Industry 4.0 affecting medical device assembly and automation?
Akella: Industry 4.0 is having a profound effect on all manufacturing sectors, and medical device manufacturing is no exception. At Drishti, Industry 4.0 is all about providing manufacturers and their employees with more data and insights, better training for their front line teams, and, most importantly, reducing their decision making time. In the medical device world, that means four things:
Barbella: How is the design of assembly systems being impacted by the kinds of medical devices produced?
Akella: Products, including medical devices, progress alongside advances in material science, electronics, software, and manufacturing. Given our focus at Drishti, I will focus on the process advances, rather than on the actual tooling required to deliver these amazing products.
One of the most important learnings of the late 20th century was the notion of concurrency—during the entire process from design to production to customer experience. The central idea being that quickly and continuously feeding back data to upstream players can enable changes that dramatically improve manufacturability while simultaneously reducing costs of design changes and improving the aesthetics and usability of products. With the advent of computer vision technologies like Drishti, it is now possible to create and use data in ways never possible before, and at a much faster pace.
As a member of the operations team, for example, you’re working closely with the product team to develop the manufacturing process in parallel with the product development process—to build in quality and quality control measures early. To do this, you are likely using manufacturing data and learnings from a Drishti system on a similar production line elsewhere in the company.
Next, during the new product introduction (NPI) and prototyping process, you are likely helping product designers build and test early designs. While the NPI team is figuring out how to build the early prototypes on the given production line, Drishti cameras capture every single step of that product’s development on video as the NPI process progresses through the different builds. AI can measure workstation activity and cycle times and, because Drishti captures every single cycle, even in these earliest stages of a product design, Drishti can help shape the design of the product itself. How? Consider an example from a heart valve line: Drishti’s assembly time data highlights both the highly variable and elevated cycle times in the station where the line associate has to grasp the valve at a particular location to work on it. Drishti analytics identify the slower cycle time, while the companion video highlights the grasping problem. The product design team, now aware of the challenge, can go back to the blueprints and create a texture on that curvature that doesn’t compromise the product, but also enables the line associate to grip in, increasing the fluidity of the assembly process.
Barbella: What are some of the challenging aspects of medical device assembly and automation?
Akella: Medical device assembly is, in many cases, very precise. Whether it’s threading a mesh implant or packaging the Spanish language instructions in the box that’s heading to Mexico, this type of assembly requires focus and accuracy 100 percent of the time, as the impact on a patient’s life is direct and significant. That’s a hefty ask for human beings. That’s why augmentation from computers is so important; computers don’t fatigue or get distracted.
Barbella: How did COVID-19 impact medical device assembly and automation processes or technology, if at all?
Akella: The coronavirus had a profound impact on all sectors of manufacturing, and medical device was one of the hardest hit for a few reasons. First, because the industry is still highly manual, the introduction of personal protective equipment (PPE) on the line had an impact on production. Steps that were previously done bare-handed had to be done while wearing rubber gloves, and that lengthened cycle time. Second, with non-essential personnel away from the line, visibility into production was severely limited, making it hard to manage to the desired KPIs. Essential personnel were forced to quarantine or leave the workforce, making absenteeism spike. Finally, many medical devices saw a drop in demand as many procedures were put on hold, while demand for other devices like respirators, ventilators and PPE, skyrocketed.
Medical device manufacturers had an unprecedented need to pivot production without their full staff and with many variables in assembly that hadn’t previously existed. Drishti was able to support manufacturers during this time by giving non-essential personnel remote access to assembly operations, helping them rapidly train and onboard line associates, and providing the data and tools needed to meet changes in demand.
Barbella: How might medical device assembly and automation evolve over the next half-decade?
Akella: When the tasks are complex and require cognition, robots simply aren’t up to the challenge. And that’s not going to change any time soon—meaning, not in our lifetimes. So, companies will need to deliberately design for human assembly. A few examples to highlight this:
Three years ago, NuTec Tooling Systems Inc. began building a syringe coating machine for a large pharmaceutical firm. The client wanted to mass produce plastic syringes with a glass-like coating to provide an alternative to the more costly glass versions typically manufactured by its competitors.
The machine NuTec constructed included four Epson Cleanroom SCARA robots (developed by Epson Robots), each of which were strategically placed at various points in the apparatus to precisely and cost-effectively automate the syringe manufacturing process.
The automated process coats syringes at a rate of 38 parts per minute, passes the parts through various inspection stations, then siliconizes the hypodermics before changing temporary caps to final caps and subjecting them to a final X-ray inspection. The robots handle the syringes both before and after their glass-like coating is applied; in the final stages, the robots apply inner and outer covers to full containers of syringes and applies labels with a laser marker.
The machine was operational in November 2020, enabling NuTec’s pharmaceutical customer to work with the government and manufacture massive quantities of syringes for use in battling COVID-19.
“Epson’s high-speed G6-Series SCARA robots with Epson RC+ software enable precision processes with exceptional repeatability assembly pick and place capabilities,” Brent Martz, NuTec Tooling Systems sales and marketing director, said in an Epson news release. “The ease of use and application versatility within the Epson RC+ development environment plus an ISO-3 rating and compliance with cleanroom standards makes them ideal for this project and the medical sector in general, where speed and precision are vital to the manufacturing process.” Bürkert
Indeed, speed and precision are vital to medtech manufacturing, particularly as the types of medical devices that can be automatically assembled are constantly expanding. MPO’s feature “Complex Coupling” details the trends and market forces impacting the medtech assembly/automation sector. Prasad Akella, founder and chairman of Drishti Technologies Inc., was among the more than half-dozen experts interviewed for the feature; his full input is provided in the following Q&A:
Michael Barbella: What current trends are shaping the medical device assembly and automation sector?
Prasad Akella: Perhaps the biggest trend is that medical devices are getting increasingly more capable, and consequently, complex. If we layer in the fact that medical device manufacturing is highly regulated, it’s natural to conclude that continuously improving the most fundamental practices—current good manufacturing practices (cGMP)—becomes even more critical to ensure compliance and the safety of these sophisticated products. For example, consider a simple device like a glucose monitor with the patch that constantly transmits sugar levels to one’s smartphone and, very likely, a cloud service. Or the latest and greatest in heart valves: These products are connected to the internet and often use advanced materials with just the right physical properties to keep us all healthy. Manufacturing them well requires deep skills.
Now, when one looks at the competitive field, medical device manufacturers are realizing they have to out-compete and out-manufacture the competition. The cGMP requirements from yesterday aren't good enough for today and certainly not for tomorrow. These manufacturers have to focus on the “current” in the cGMP and go beyond the regulatory requirements to develop even better manufacturing practices that increase efficiency, quality, safety and ultimately revenue.
Barbella: What factors are driving the need for automation in medical devices?
Akella: The need for automation in medical devices is driven by the same factors all of manufacturing is driven by—cost, volume, speed, quality and personalization. Automation is inexorable!
That said, I believe that there is a fundamental underlying assumption in your question: That the primary path to scale is automation, since we have maximized what we can do with manual processes. I flag this as being especially important since every medical device manufacturing line I have seen—from blood pressure monitors to heart valves to MRI machines—is manned by people. Because people can easily perform tasks that a piece of automation might be challenged to do, certainly at a viable price point.
So, in addition to automation, the bigger need is to drive a lean culture and impact the 70-plus percent of your lines that are manual. To provide them with the tools and processes to build these medical devices safely and at scale—while simultaneously empowering them to make better and quicker decisions. Tools and processes that augment humans and help them perform manual assembly tasks to the best of their ability. This could be augmented reality, poka yoke systems or, in Drishti’s case, data, analytics and insights from AI and computer vision systems.
Barbella: What new innovations have been developed within the medical device assembly and automation space? What specific market needs do these innovations address?
Akella: Significant energy is being focused on increasing quality, gaining visibility into operations, and supporting decision-making at all levels, and for good reason. The risk to end users is massive, and the fines and penalties that may be levied against an organization with defective products are hefty.
A new advancement in this area is computer vision and AI (“AI-on-video,” if you will), which can help detect defects as they occur, rather than wait for an end-of-line inspector to catch and flag them. The economic value stems from the fact that the longer a defective unit goes down the line, the more value is being added to a fundamentally useless unit. This new inspection methodology can, therefore, avoid significant scrap and rework, which is expensive for manufacturers. It also alleviates the heavy burden on the inspector to catch the defect, because having prevention mechanisms in place upstream increases the likelihood that problems will be caught earlier. This system also helps manufacturers identify opportunities to train the team, either in the moment with feedback (think of your spell checker running behind the scenes in your email client) or in a training station (think of your golf swing being videotaped for you to learn from after the fact).
Barbella: How is Industry 4.0 affecting medical device assembly and automation?
Akella: Industry 4.0 is having a profound effect on all manufacturing sectors, and medical device manufacturing is no exception. At Drishti, Industry 4.0 is all about providing manufacturers and their employees with more data and insights, better training for their front line teams, and, most importantly, reducing their decision making time. In the medical device world, that means four things:
- You have more information on your manual assembly lines than you’ve ever had before.
- The information is presented to you when, where, and how you need it to take decisive action.
- Every data point available to you is also backed by video to simplify root cause analysis, training and collaborating across the company on problems now made visible.
- Your people, including your line associates, are empowered to make better decisions, faster, on a continuous basis.
Barbella: How is the design of assembly systems being impacted by the kinds of medical devices produced?
Akella: Products, including medical devices, progress alongside advances in material science, electronics, software, and manufacturing. Given our focus at Drishti, I will focus on the process advances, rather than on the actual tooling required to deliver these amazing products.
One of the most important learnings of the late 20th century was the notion of concurrency—during the entire process from design to production to customer experience. The central idea being that quickly and continuously feeding back data to upstream players can enable changes that dramatically improve manufacturability while simultaneously reducing costs of design changes and improving the aesthetics and usability of products. With the advent of computer vision technologies like Drishti, it is now possible to create and use data in ways never possible before, and at a much faster pace.
As a member of the operations team, for example, you’re working closely with the product team to develop the manufacturing process in parallel with the product development process—to build in quality and quality control measures early. To do this, you are likely using manufacturing data and learnings from a Drishti system on a similar production line elsewhere in the company.
Next, during the new product introduction (NPI) and prototyping process, you are likely helping product designers build and test early designs. While the NPI team is figuring out how to build the early prototypes on the given production line, Drishti cameras capture every single step of that product’s development on video as the NPI process progresses through the different builds. AI can measure workstation activity and cycle times and, because Drishti captures every single cycle, even in these earliest stages of a product design, Drishti can help shape the design of the product itself. How? Consider an example from a heart valve line: Drishti’s assembly time data highlights both the highly variable and elevated cycle times in the station where the line associate has to grasp the valve at a particular location to work on it. Drishti analytics identify the slower cycle time, while the companion video highlights the grasping problem. The product design team, now aware of the challenge, can go back to the blueprints and create a texture on that curvature that doesn’t compromise the product, but also enables the line associate to grip in, increasing the fluidity of the assembly process.
Barbella: What are some of the challenging aspects of medical device assembly and automation?
Akella: Medical device assembly is, in many cases, very precise. Whether it’s threading a mesh implant or packaging the Spanish language instructions in the box that’s heading to Mexico, this type of assembly requires focus and accuracy 100 percent of the time, as the impact on a patient’s life is direct and significant. That’s a hefty ask for human beings. That’s why augmentation from computers is so important; computers don’t fatigue or get distracted.
Barbella: How did COVID-19 impact medical device assembly and automation processes or technology, if at all?
Akella: The coronavirus had a profound impact on all sectors of manufacturing, and medical device was one of the hardest hit for a few reasons. First, because the industry is still highly manual, the introduction of personal protective equipment (PPE) on the line had an impact on production. Steps that were previously done bare-handed had to be done while wearing rubber gloves, and that lengthened cycle time. Second, with non-essential personnel away from the line, visibility into production was severely limited, making it hard to manage to the desired KPIs. Essential personnel were forced to quarantine or leave the workforce, making absenteeism spike. Finally, many medical devices saw a drop in demand as many procedures were put on hold, while demand for other devices like respirators, ventilators and PPE, skyrocketed.
Medical device manufacturers had an unprecedented need to pivot production without their full staff and with many variables in assembly that hadn’t previously existed. Drishti was able to support manufacturers during this time by giving non-essential personnel remote access to assembly operations, helping them rapidly train and onboard line associates, and providing the data and tools needed to meet changes in demand.
Barbella: How might medical device assembly and automation evolve over the next half-decade?
Akella: When the tasks are complex and require cognition, robots simply aren’t up to the challenge. And that’s not going to change any time soon—meaning, not in our lifetimes. So, companies will need to deliberately design for human assembly. A few examples to highlight this:
- Design for the resolution of the human eye or for the dextrous capabilities of the human hand.
- Continue investing in the training of people.
- Present information so even those of us not trained in probability and statistics can make quick and informed decisions on behalf of our employers.