When to automate with robots

Taking the leap to add robotics into a company’s operations may not be an easy decision – especially when there are so many options available. Add to that the unknown of calculating the return on investment (ROI), and some may even shy away, feeling company size or processes won’t benefit.

That couldn’t be further from the truth. To help understand the role robots and automation play in a manufacturing facility, Ryan Weaver, automation engineer with Axis New England – a Universal Robots distributor – helps sort out the questions and anxiety that some people have.

TMD: What manufacturing processes benefit from robots?

RW: Any repetitive manual process tends to be a candidate for automation, but the approach will vary depending on several factors, such as ROI, throughput, and process variability.

Traditionally, robots were suited to high-speed, low-variability tasks, with fixed guarding and custom integration. ROI for these work cells needs to provide quick payback because of the high-cost and low-flexibility.

The approach today is a bit different. Rather than being a component within a highly engineered automation system, Universal Robots are deployed as tools for handling the repetitive manual portion of a task. The overall cost of deployment is significantly reduced since there is no guarding necessary. In addition, the programming is simple and can be handled by the user, so we can consider using Universal Robots on tasks where previously we could never justify a traditional robot.

TMD: How does integration enable flexibility of a manufacturing line? Can it quickly switch from one product run to another, or are most integrations done for a specific process?

RW: We have seen installations go both ways; sometimes robots are used in a dedicated process, other times they will be used with a process where there is constant changeover. Universal Robots are well suited to tasks where there is changeover because of the ability to safely use robots without guarding. Since the workspace isn’t restricted by cages, it is easier to reconfigure for the next job than a traditional robot cell would be.

TMD: What manufacturing processes support the most effective robotic systems?

RW: We always look for the simplest solution to the problem first, then scale up as required. It is easier and less expensive to have some human tending of the robot system than it is to fully automate. For example, we will try to explore mechanical ways of fixturing parts presented to a robot before we jump straight into machine vision guidance. A lot of that peripheral automation adds layers of complexity that we only want to explore if necessary.

TMD: How does material handling determine the right robotics/automation approach?

RW: The end-of-arm tooling part of the system usually takes into consideration the material being handled, not so much the robot. As long as the part is within the robot’s payload, we focus on the type of gripper we need to pick up the part. We use a combination of electric, pneumatic, and vacuum devices to find the right tool for the job.

One thing to focus on is the variability between parts being handled. If the part will change frequently, it might be a good idea to explore a flexible gripper that can handle different part shapes and sizes. One solution for this type of application is the two-finger electric gripper from Robotiq.

TMD: Do customers look for robotics to automate the entire process or parts of the process, and how is this determined?

RW: Generally, with Universal Robots, our goal is to automate only part of the process. There is an exponential relationship between how much of a process is automated and overall integration cost. Machinists have skills that they bring to the table, and some involvement is usually beneficial in a process.

TMD: What are the speed and precision requirements that can be achieved when integrating automation/robots?

RW: Speed is usually determined by the required throughput, but we try to have the robots running as slowly as possible within a process to ensure safety in an unguarded application. For example, we may try to have the gripper pick up three or four parts instead of one at a time. That way we only have to run with 25% of the speed compared to picking up a single part.

If a process can run 24/7, the robot has a distinct advantage over a human operator because there are no required break times, shift changes, etc.

Universal Robots are repeatable to ±0.004", which allows for a high level of precision. We have seen significant quality gains for our customers where the process used to be manual, since it is always done the same way every time. People aren’t perfect; it’s in our nature. We get tired, we get bored or distracted. Robots just keep going, doing the same thing the same way, all day.

TMD: How does robotic integration/automation minimize downtime between production runs?

RW: Since programs can be stored right in the robot controller, then called up from the teach pendant in an instant, it is very easy to change from one product run to another.

TMD: How does the system adapt/expand if the process grows or changes in the future?

RW: One of the greatest advantages of using a collaborative robot is that you can easily reconfigure or change the way the robot is deployed. You can change the work cell around the robot, or you can move the robot to a new cell without having to move any guarding. In addition, since the robots are very easy to program and reconfigure, customers own their own programs instead of being tied to an integrator.

TMD: What special compliance considerations are required when integrating robots into medical manufacturing? (FDA, cleanroom requirements, etc.)

RW: While I’m not an expert in this area, I can say that since medical processes need to be validated, one of best approaches we’ve seen is to use the robot to run the process the same way as the human does. That way, there is no need to re-validate the process since nothing has effectively changed.

Universal Robots USA Inc.

www.universal-robots.com

IMTS 2016 Booth #N-6865

Tegra Medical

www.tegramedical.com

Axis New England

www.axisne.com

Robotiq

www.robotiq.com

About the author: Elizabeth Engler Modic is the editor TMD and can be reached at emodic@gie.net or 216.393.0264