Formic hosted a live Q&A event on February 25th, 2025, to answer your questions about end-of-line automation. Our automation experts got dozens of interesting questions from attendees curious about automation ROI, system reconfiguration as needs change, and training a team to operate a new automation system. If you missed the live event, we’re spotlighting questions on automation’s flexibility and the future of robotics here.
You can listen to the full recording of the Q&A here. Plus, stay tuned for the final segment of our “Ask the Expert” series.
What’s a cobot, and what’s a cobot’s typical ROI?
Danijel Lolic (DL), Formic VP of Product & Solutions: The cobot is a collaborative robot intended to operate collaboratively with people working around it. Usually, this is a joint area where force sensing and area scanner safety devices are used to be aware of people working in that area or entering the area and to adjust its speed and behavior appropriately. It’s a really great technology and has a lot of good use cases, but is only applicable for certain applications.
In terms of ROI, I think what people are looking for here is how to know that they will break even or make money with this automation investment. It boils down to whether you’re looking to buy a component that is a robot, which has certain costs associated with it, and then associated costs to program, service, and be responsive for it on a day-to-day basis. If you want to use a collaborative robot to palletize boxes, you’ll need a system. You will need someone who has a tool on it, programming that you may or may not be responsible for, a platform, a base, a guide, and some conveyor. So, it boils down to what the application is and what the system is. I would say you can get a collaborative robot to stack boxes on a pallet fairly economically.
However, if you’re going to use that same collaborative robot to case pack something, you’re going to have a much tougher time doing that within the same cost confines. That ROI depends on what it is you’re solving for at an application level and how much of the ownership you’re going to be responsible for on a day-to-day in terms of servicing and programming.
I don’t have an easy answer to this question, other than that you should consider more than just the initial investment and look at what else automation typically requires. You also have to consider how much you expect your business to evolve over a certain period. If you’re buying an automation system and it will stay static and operate the same way for the next five years, then the ROI calculation is very clean.
Generally, though, if you know you’re going to buy and know it will have to evolve with the business every six months, every year, every 18 months, the ROI calculation becomes a lot more challenging because you have to continually put costs in that investment and the outcome will vary.
There isn’t a simple answer to the ROI of a cobot because it’s application-specific. But it’s very important to consider the application scope and what your business looks like, especially how your needs will evolve versus how consistent they are.
If my production line speed increases over time, can the automation system be upgraded or reconfigured?
Nick DeLong (ND), Formic Technical Operations Manager: I’ll shamelessly plug Robots-as-a-Service (RaaS) here because it’s a differentiator. This question is something you should consider when purchasing an automation system in the first place.
If you know that your business ramp scale is going to be very drastic, then you have the option to purchase things differently — whether it’s using CapEx or the RaaS offering. With RaaS, we have the flex option, where if you see your ramp curve doing what you expect it to do, then you have the option to upgrade the equipment wholly to a different, larger, and more capable piece of equipment. If you’re committing to a very large system purchase, then you’re inherently limited in flexibility.
DL: I agree with that. I bought a lot of equipment that I’ve been burned by — not necessarily because I did something wrong (but sometimes I did). Whether I spent $3,500 or $3.5 million on a project, when that project ran out of its useful life, either because my business changed or it no longer satisfied my needs, I was beyond its capacities and it became a noose around my neck. I always felt the weight of the opportunity cost of that $3,500, $35,000, or $3.5 million that I could be spending on enhancing my business. So, I think Nick is right that it depends on your sourcing and purchasing decisions.
You don’t know what two years, three years, or five years from now looks like. This is part of your due diligence as a buyer, as a customer, and as a user that you should be asking, “Hey, this thing that you’re proposing, how much capacity for growth do I have?”.
How quickly can my team be trained to operate an automation system like a cobot?
ND: It depends on how intuitive the cobot system is. Teaching points with a cobot is relatively easy and is easily trained to your operators. However, if the user interface on the system doesn’t act intuitively for recovery from any forces, sense collisions, or any condition that might come up during the operation, that limits your capabilities.
Some cobot systems can build pallet patterns right on the screen. So, you can use the user interface and select the layers, select the box size, and select what your pallet pattern wants to look like. But some cobot systems don’t have that.
It’s important to understand what the user interface looks like and the full depth of it, because that will help you understand how quickly and efficiently the team can operate it.
What functions for predictive and preventive maintenance do you expect in the future from robot makers?
ND: As the robot market evolves, everybody will start to implement tools related to the Internet of Things, sensors, cloud-based AI, and predictive and preventative maintenance. A lot of what robot manufacturers are starting to implement is intuitive. The robot actually monitors itself, or it will monitor its axis for wear.
For example, if it sees consistent resistance across a time period that is outside the normal range, it will throw up an alarm and say, “I’m predicting I’m having a failure on this axis.” It can tell the same results based on friction assessments and heat transfer. A lot of these advancements are for the robots themselves.
Additionally, you can also augment systems by watching performance. If you’re watching the cycle time of the system over a certain period and you start to see degradation in the system’s performance, then you can inherently predict something is going wrong. It is an indicator that allows you to dive deeper into the system and understand those parameters to manage a failure before it happens.
DL: For me, some of it boils down to everything Nick said at an individual component in an assembly level, but some of the system-level monitoring comes back to “What do you know about the application?”.
At Formic, for example, we do 24/7 monitoring and condition monitoring into the system. We know what the targets and expected outcomes are of that palletizing or case packing system, so we can track trends and performance and say, “This thing isn’t living up to its trend, not just on the condition or oil or overheating on an individual component level, but on the trend of your performance or production outcome.”
If you want to predict that a robot arm needs servicing based on predictive preventative technology, you would need to apply specific applications to a system or a production line to ensure everything is working in harmony.
Listen to the full recording of the Q&A here.