Robotics Industry Insights
- This industry insights is filed under:
Robot Market: Heal Thyself -- With the Help of Medical Devices
by Winn Hardin, Contributing Editor
Robotic Industries Association Posted 02/18/2003
The areas of medical device manufacturing, biomedicine and drug discovery have shown little reaction to today’s economic hardships, prompting many robot vendors and integrators to aggressively pursue manufacturing, material handling and R&D applications in these booming industries.
‘‘The medical industry hasn’t been highly automated up to this point,’‘ said Bob Rice, business development manager for RTS Wright Industries LLC (Nashville), an international automation vendor and integrator with expertise in the medical device area. ‘‘We’re replacing more manual operations with automation, and when you think of automation in the medical device area, you often think of a robot.’‘
Cleanroom marketing manager for Staubli Corp.’s robot division (Duncan, SC), Rick Palmer, agrees. ‘‘Robots in medical devices have experienced aggressive growth during the past five years,’‘ Palmer said. ‘‘Robot manufacturers are looking for places to sell, and the drug device and pharmaceutical industries haven't seen much of a recession.’‘
Medical Applications Far and Wide
Accommodating economics aside, the robot industry is benefiting financially from two other characteristics typical of biomedical applications: the breadth of the applications and nature of the products.
Robots are in nearly all aspects of medical device development and manufacture, including R&D in drug discovery; the manufacture of extremely complex and small devices such as hearing aids and orthodontic braces for teeth; portable plastic injection systems that hermetically manufacture and package artificial arteries in a cleanroom environment; as well as grinding and polishing applications for larger implants, prosthetics and replacement hips, similar in application and design to many other metal manufacturing and finishing applications.
Most of these products are subject to strict U.S. Food and Drug Administration rules regarding validation, documentation and manufacturing practices, such as 21 CFR part 11. At the same time, manufacturing operations are typically subject to Class 100 cleanroom specifications set out in U.S. federal standard 209E and ISO 14644-1. Because of these strict regulations and the time and financial costs of validating each product, they typically have longer lifetimes.
‘‘Medical components are not high turnover product lines,’‘ said Andy Carson, vice president of sales at RTS Wright. ‘‘It’s the same thing with pharmaceutical plants that are more willing to use longer payback periods to justify investment.’‘ As a result of federal requirements, plant expansions often hinge on quality- and performance-enhancing technology more than simply pushing quantity or line speeds, limiting robotic competition in this industry.
FDA requirements also have the effect of creating the need for systems that combine robots and automated vision inspection. ‘‘Vision systems are unbelievably booming in the medical device industry because verification is a key part of the process,’‘ said RTS Wrights’ Carson.
EPSON Robotics’ (Carson, CA) sales manager for the Midwest region, Jay Hallberg, recently had success combining visual inspection and robotics to sort of needles and manufacture medical shunts. Shunts are needles with a plastic cover or overmold. The application called for three needle categories, each with three lengths and all in the same bin. The vision system determined the size of a need, determined its orientation and then brought it to the injection molding machine for integration into the plastic shunt.
Precision & Speed
The precision and programmability of robotics compared with vibrating bowls and mechanical feeders also appeals to those concerned with yield, safety and litigation issues surrounding medical devices. For instance, Staubli is one among many robot companies that has been able to capitalize on extremely precise placement of parts during the manufacturing of medical devices, such as hearing aids.
‘‘The placement of components can be so precise that humans can not repeatedly do it with any sort of economic yield,’‘ he said, adding that precise placement and smooth movements can be more important than line or robot speed.
Drug discovery is another example. Automated drug testing machines or DNA decoders use robot arms to fill thousands of miniscule wells on a single plate. Each well may represent six months of work on a single drug compound. ‘‘Smooth motion is important, an agitation strong enough to spill liquid from one well to another could be disastrous because of a false test result,’‘ said Staubli’s Palmer.
Servos and PLCs are less reliable for these sorts of movements, claims RTS Wright’s Rice. ‘‘With PLCs and custom servos, you have to worry about every little motion, like what will happen if the process is interrupted or when it boots up,’‘ Rice explained. Other examples where tight motion control is required are precision dispensing systems, such as contact lenses and large pharmacies where a robot fills orders for shipping on a per patient basis. In these applications, Rice said, speed and precision are critical to economically and efficiently improving a patient’s health.
Leveraging Electronic Lessons
Another advantage programmable robots have over their PLC ancestors and human counterparts lies in the robotics industry’s efforts to make their systems more rugged. Safety concerns, and increased demands for higher duty cycles with longer mean time between failure (MTBF) have pushed cabling and control mechanisms inside the robot arm or enclosure. The result is a cleaner machine that generates fewer particles than a mechanical system that uses exposed motors, wires and lubricants.
Although robot vendors rarely have to be concerned with FDA validation needs (this typically falls on the integrator or end-user), ‘‘…we do have to deal with other specifications related to cleanrooms for applications such as drug discovery and pharmaceutical manufacturing,’‘ explained Joseph Portelli, cleanroom program manager at FANUC Robotics America (Rochester Hills, MI). ‘‘Most robots are far cleaner than people, and whether you’re talking microelectronics or the packaging of catheters, the number one contributor of contamination is people.’‘
A typical hospital operating room compares to a Class 1,000 clean room, Portelli said. By comparison, the majority of medical device manufacturing is done in cleanrooms rated class 100 to Class 10,000.
‘‘Manufacturing semiconductor wafers is much more stringent. Contaminants are measured in microns and equipment is typically Class 1 to 10,’‘ Portelli said. ‘‘In the medical area, we’re more concerned about bacteriological and viral contamination rather than simply dust. Bacteria is a colonial organism that tends to clump so the cleanroom’s filters don’t need to be as fine and Class 100 equipment is usually sufficient.’‘
One lesson here is that tremendous cost savings can be realized by not using robots that are rated cleaner than necessary. Portelli added that the ‘‘vast majority’‘ of cleanroom injection molding, a small but growing market for the robotic industry, is done for medical devices.
Based on precision, programmability, emissions and other lessons learned by robots in support of the semiconductor and industrial manufacturing industries, robots are particularly well positioned to benefit from the boom in biomedicine and healthcare. In the late 1990s, robots in biomedical applications were similar to those sold to the automotive industry some 20 years ago.
‘‘A few years ago many of the robots used in lab automation applications were very light duty,’’‘ said Staubli’s Palmer. ‘‘In future applications you're going to see the application of industrial quality robots with very high MTBF numbers. Today, duty cycles are going up rapidly, even though in many cases, the speeds are relatively minimal. When duty cycles are high, a poor up-time can rapidly take the yield out of the system, so the industry is becoming more and more interested in reliability.’‘
Originally published by RIA via www.robotics.org on 02/18/2003