It's the age of the robot. The speed, precision, and power of industrial robotics continues to grow, and manufacturers who don't take advantage of today's robotics are missing out on massive potential for efficiency, profits and growth.

Whether you're a production manager looking to maximize throughput or an engineer working on an improvement project, you're probably noticing tasks and processes on your factory floor that seem like good candidates for automation.

The question is, where do you start? What robots should you be looking at? How can you get them up and running? and most importantly, how can you get your company to sign off on a major new investment?

To answer these questions and more, I called up experts from some of the world’s leading robotics companies. With their help, I've put together a guide which will set you on the path toward your company's first robotic cell.

I also tapped these experts for their best advice for making a winning presentation to get your robotic automation project approved by the people who sign the checks. It’s more than ROI. Hit the button at the top or bottom of this article to get our kit for a successful pitch, including:

expert advice
plenty of useful links to online tools and services
and a starter PowerPoint presentation template

But first, read the 3 sections below: How to Tell if You Need an Industrial Robot, How to Choose the Right Robot, and How to Prepare for Your Industrial Robot Implementation.

Featuring expert advice from:

		Dean Elkins Segment Leader, Material Handling Yaskawa Motoman
		Dan Hasley Director of Sales Kawasaki Robotics
		Dick Motley Director, Authorized System Integrator Network FANUC America Corporation
		Brian Dillman Area Sales Manager – East Universal Robots

If you’re reading this, you probably already know that the potential of today’s industrial robots to supercharge the productivity of your assets is undeniable. But where exactly should you start?

How to Tell if You Need an Industrial Robot

We’ve all heard the four D’s of robotization (Dull, Dirty, Dangerous, and Dextrous)—attributes of tasks which make them ideal for robotic automation. However, I think there’s more to it than that.

Robots Can Help Solve Labor Shortages

Dean Elkins, Segment Leader of Material Handling at Yaskawa Motoman, pointed to labor issues as a common catalyst for new robot purchases: “Perhaps you've gotten a rather large volume contract, that has lots of consistency in the process. For example, lots of material handling, or lots of arc welding in a batch, and you're compressed from the labor standpoint. You may not have adequate labor to run the job, or your physical location of your factory might be such that your work force is not all that available.”

Dan Hasley, Director of Sales at Kawasaki Robotics, agreed. “If we're talking to a metal fabricator full of welders, they're probably experiencing a shortage of labor because skilled welders are getting fewer and further between. So, they're having trouble keeping up and growing their production because they can't find more people, they can't add capacity. The way out of that conundrum for them is to automate. They have to automate just to grow their business.”

Even if your production doesn’t require skilled workers, staffing can be difficult. For example, if you have a highly variable level of work from week to week, you may not be able to find workers who can manage that much flexibility in their hours. For this problem, the solution could be a staff of workers to cover the typical base level of consistent work, with robots to help handle high-volume busy times.

Measuring Asset Performance and Finding Room for Improvement

Every manufacturer has one thing in common: they all have assets on the factory floor—whether you’re making aluminum chips or potato chips. The question is, are you making the most of your production assets?

This is basic stuff for manufacturing engineers with experience with kaizen or continuous improvement projects. How much productive work is your asset chewing through per hour, per shift, per quarter? This is a KPI for the potential improvement a robot could bring to your process. Here’s an example given by Dick Motley, director of the Authorized System Integrator (ASI) network at Fanuc America. The example is from the baking industry:

“There was a system I'm familiar with that involved taking dough out of the proofing racks, ready to be baked. There was a person transferring trays of dough from rolling racks to a certain type of industrial oven,” Motley described. “You would slide these pans into the available slots in the conveyor that then rose up into the oven, and then went around the top and came back down. By the time the bread made that trip through the oven, it was baked.”

The owner of the bakery, Motley explained, noticed that the loading and unloading process was ergonomically challenging for human workers, involving reaching up high and bending low to reach trays of dough. To address this, the owner automated the oven loading and unloading processes with robots.

When they did, the bakery found an unforeseen benefit to the automation, which had to do with the efficiency of the asset in normal operation:

“What they found out,” said Motley, “was because of the bad ergonomics, the operators had only been filling a small percentage of the available slots in the oven to bake the bread. So, most of the time, that oven was just empty hot air. When they started loading it robotically and getting every slot filled with dough that was ready to be baked, the capacity of that oven went up 80%.”

The bakery also saw further knock-on benefits to the boosted capacity:

“From an energy usage perspective, if the oven's on, you're burning the gas whether you've got bread in there or not. The energy per loaf of bread went down by 50%. The bakery got way more out of the asset and on a per-piece basis, the operating cost for the energy to bake the bread went way down. And you can realize those types of benefits in any industry,” explained Motley.

The takeaway: when you look at arc-on time in welding, spindle utilisation in machining, or even seeing boxes hitting a bottleneck at the end of a packaging line, you’re seeing the opportunity for a robot to dramatically increase the utilization of those assets, kicking your production capabilities into high gear.

Which Process Should Be Automated?

When I asked the experts which tasks make the best early targets for automation, they emphasized aiming for the low-hanging fruit: dull, dirty, dangerous, or highly precise (dextrous) jobs which are mentally easy and repetitive, but physically challenging or harmful.

Brian Dillman, Area Sales Manager with Universal Robots, one of the world’s top suppliers of collaborative robots, had this to say:

“Look around the plant for a job nobody wants to do because it's mindless or there's risk of repetitive motion injuries and it's not a difficult job to do. Simple tasks like that can be automated so that you can then move your people into a more value-added position,” Dillman said.

Motley of FANUC added another good piece of advice. “In the internet age, do you see it on YouTube?” he pointed out.

If you’re curious about a possible application, you may be able to find that someone else, be it a university lab, a robotics supplier, a system integrator or an ordinary manufacturer has already done something similar. For example, here’s a very interesting video showing a FANUC robot tending a mill and a lathe in one workcell. (courtesy of Bastian Solutions)

(courtesy of Owens Design Inc)

And here’s a video showing a computer assembly task using Denso robots:

“I just hear all the time that that's the first stop. A customer seeking whether or not they want to pursue some robotic automation now goes straight to YouTube and says, ‘Hey, what's out there? What's been done successfully, and who has done it successfully?’” said Motley. He continued, “I think an important step is going to be to get in touch with either the robot manufacturers or a system integrator. Here at FANUC, the vast majority of our sales staff are very experienced technical people whose careers started in engineering. They can walk through a plant and say, "Yup, I've seen that done. Nope don't want to go there." That's another key aspect, is to just engage the experts. Ask the experts in addition to the information you can glean on the internet.”

Specialized Process Applications: Welding and Painting

Courtesy of STC Italia

According to the Robotics Industry Association (RIA) Arc welding is one of the fastest-growing application sectors in industrial robotics, largely due to a shortage of skilled workers. The question is, is it easier for a welder to learn to program a robot, or for a robot programmer to learn how to weld? The answer is usually neither, said Hasley of Kawasaki. “A robot programmer who doesn’t know how to arc weld may fail at programming an arc welding robot. You have to have a process expert in order to set up a robot system to perform that process. Painting is exactly the same: you have a robot programmer who knows nothing about painting, or boy, that car at the end of the line is going to look like a Picasso, and not in a good way,” he pointed out. “The best approach is to put together a cohesive team including robot operators and welders, so that your staff can bring together the needed expertise for these specialised applications.”

Hasley wisely advises that management should carefully consider company culture when implementing this type of application. “So, you’ve got to have a merging of the two disciplines, and sometimes organizations are not structured to do this very efficiently. They'll have the welding department who's been welding by hand for years, and all of a sudden there's the automation department and you got a bunch of people with robots who don't know anything about welding, and there becomes a culture clash.”

Robotic Automation Pitfalls and Newbie Mistakes to Avoid

When starting a robotic automation program, many customers tend to make the same common mistakes. According to Motley of FANUC, it’s important to take your time and make sure your company isn’t biting off more than it can chew, so to speak. Is everyone on board for this project?

“I've been in the industry over thirty years, counting my time prior to FANUC,” Motley said. “If there is one consistent, overarching indicator for success, the one word I would use is ownership. No matter what the business, no matter what the industry, no matter what the application, the clearest indicator of success is does the end user really feel they own that piece of automation? That goes for all the stake holders along the way. Does the purchasing person feel like he or she got a good return? Did they get a fair price? Did the maintenance person feel like they got a piece of equipment that they can keep running? Does the operator feel comfortable interacting with it on a day to day basis? So and so forth. All the stake holders along the way have to buy in and say, ‘this was a good decision’. Building that ownership is probably the single biggest thing that I think is a predictor of success.”

Hasley agreed, advising a similarly sober approach. “I think probably the biggest pitfall is having an unrealistic automation strategy. By that I mean, going after very complex processes the first time out. For example, if you've got a highly intricate assembly process that requires a lot of human touch and feel, or very soft edges, and things that are hard to repeat, or hard to even quantify, that would be a challenging one to go after with robotics or even any automation. So, that's a big pitfall that we see customers try,” said Hasley. “I think first time robot users need to need to walk before they run and start by looking at simple processes in their manufacturing environment.”

Hasley also advised that failing to take personnel considerations into account can spell trouble for a robotics program. “Allow time for the proper training, to do proper maintenance over the life cycle of the robot, and also consider safety by either using a collaborative robot or studying the safety standards, so your team knows how to properly protect workers from robots.”

Summary: Do You Need a Robot?

Granted: so far, this article has featured robot salespeople advising you to buy robots. However, the simple arithmetic of the massive efficiency and productivity of robots is undeniable. Take a stopwatch and take a look at the utilization of the machines on your factory floor. Chances are, there’s room for improvement.

When you do decide to get a robot, below are the critical considerations for your selection.

How to Choose the Right Robot for an Application

Image courtesy of FANUC.

The journey of selecting the right industrial robot can start with a phone call to a manufacturer. As Motley pointed out above, sales and support technicians at these companies can provide the consultation expertise to find the right solution for you.

But first, here are some factors to consider.

Will You Need to Hire an Integrator?

The two videos above were made by robotic integrator companies, contractors who specialize in designing and building robotic work cells. While it’s usually possible to buy your robots direct from the manufacturer, investing in an integrator could be a wise decision. It all depends on your level of in-house robotics expertise.

“If you hire people that have a lot of experience with robotic automation, you can definitely give it a shot,” said Hasley. “We have many customers at Kawasaki that are successful at doing their own integration in-house. We sell our robots to them directly.”

However, Hasley also warned that in-house expertise comes at a cost. “You also have to understand whether you're planning on using enough robotic automation to justify having full time in-house staff to design and implement those types of systems. A system integrator is going to be able to come to you with a lot of pre-engineered solutions. Things they've already done for many customers before you showed up, and they can provide you with something that's going to cost you less even though you're outsourcing the system integration. It's going to end up costing less because you don't have to start from scratch and reinvent the wheel.”

Elkins echoed that opinion. Integrators can help prevent the sprawl of costs associated with taking on a new project.

“If you're a business owner that doesn't have a staff on hand that has the aptitude to integrate, then certainly, system integrators are worth their weight in gold,” Elkins said.

“If you have to provide a lot of tooling for a welding application, somebody has to make those tools. And if your company isn't capable of doing so, if you don't have a machine shop that gives you that capability, then you'd have to look outwardly, and an integrator would be the answer for you.”

However, if your team has the mechatronic chops to build a functioning system, then it may be time to take the reins. If the work involves a trade secret or proprietary information, this may also lead your company to self-integrate.

“If you do have the capabilities of providing those types of tools, then why not try doing it yourself? I would just comment that if you're going to do an application yourself, you want to make sure that you're integrating the technology in a robust yet safe way, and follow risk assessment standards,” said Elkins. Your supplier will have resources and expertise to help guide you in the right direction as well.

When to Use a Collaborative Robot

Image Courtesy of Universal Robots

For more information about cobots, check out A History of Collaborative Robots: From Intelligent Lift Assists to Cobots by my colleague Kagan Pittman.

To be blunt, collaborative robots allow you to avoid much of the red tape and guarding associated with industrial robots, at the cost of speed. However, their ease of installation and focus on user-friendliness may make them the right fit for your application.

I asked Brian Dillman of Universal Robots why customers might choose cobots over regular, non-collaborative robot options. “I think a big part of that has to do with how it’s a less complex system to deal with,” He said. “There’s a lot of concern about how technically astute they’ll be and how they can handle automation. Robotics does not have to be as difficult as you think. A lightweight collaborative robot that easy to program and easy to set up is less threatening to a first time user.”

Universal Robots are much lighter than other options. The UR-10, the company’s largest robot, weighs just under 64 lbs., and carries a payload of 22 lbs., or 10 kg. The robot has a reach of 51 in.

“When you can take a robot out of a box by hand, and put four bolts in it, and that's essentially all you need to do to set it up, and it has single-phase 110 power, where you set it up becomes even more flexible,” said Dillman. “A large advantage of [cobots] today is the redeployability of the arms. A lot of our users actually put the robot on a mobile pedestal. People are taking it from one operation to another which is fairly unprecedented from a robot perspective. People are now looking at these products and saying, ‘well wait a minute, where else in the plant can I use it?’ and we encourage that, because it gives that first-time user the flexibility of saying if I have no single operation that has really high utilization, why not look at two or three or more applications and move the robot where I need it?”

Kawasaki’s duAro cobot has similar versatility. The dual-arm SCARA robot comes equipped with computer vision and various sensors to enable a greater degree of application flexibility, and the entire system comes installed on a rolling cart.

Image courtesy of Kawasaki Robotics.

While cobots are more user-friendly and flexible, their safety protocols restrict speed, and depending on the task, the safety features may be irrelevant. For example, an arc-welding cobot still requires guarding due to arc flash and other dangers. While the robot itself may be safe, the peripheral equipment, especially end of arm tooling, may not be.

How Much Robot Do You Need?

IRB 8700 courtesy of ABB

Should you buy a stronger, larger robot than your task requires, to allow you the flexibility to redeploy it in future?

On the other hand, should you buy the bare minimum required to handle your task to maximise efficiency?

“You know, I think it's always a good idea to give yourself a little bit of cushion,” said Elkins. “There's things that you need to look at when implementing a robot, relative to reach and payload, that have to be considered, such as moment loads, for example. Just because you have a 12-kilogram payload robot, and you're manipulating a 9-kilogram part, doesn't mean that it's necessarily a good fit if you have to offset the part way off of the gripper itself because you might exceed some loading limit. So you really need to be mindful of what you're moving, and what the work envelope's going to look lie, and what your tooling's going to look like.”

Yaskawa Motoman and other robot manufacturers offer tools and software to help customers calculate the loads on their robot. For example, the free software KUKA.Load will calculate the loads exerted on a robot on the flange, as well as secondary and tertiary loads on the arm and at the base of the KUKA robot.

Hasley of Kawasaki offered the following advice. “We always try to preach the message of the robot as a reusable asset. Robots can be used in different tasks, so you're not locking yourself into the task in front of you today. At Kawasaki, our salespeople always have a conversation with our customers before they specify the robot size and reach and payload, so that the customer really has time to consider possible future uses of the robot, and if they feel that there's a need that might exceed the specifications for the immediate job that they're looking at then we consult with them to size up the robot in some cases,” he said.

“However, there is a trade-off, because if you oversize your robot then you sometimes sacrifice speed, or you have a larger motor package on this robot that you might need so you're paying a little bit extra for electricity to run that robot that normally would. But our customers have the opportunity to consider that before they buy a robot from Kawasaki.”

However, you may not have to make this decision. Many robots on the market come with interchangeable or extendable components, allowing you some degree of flexibility, without needing to size up your robot.

“Some Kawasaki robot models have interchangeable components. For example, we have a robot that you can change out the wrist, and all of a sudden while 90% of it is the same as before, by changing out the wrist you increase your payload by a certain factor. The robot design itself gives you some flexibility without having to replace the entire robot. For example, if you wanted to increase your payload capacity or even increase your reach you can do the same thing, if you want a longer arm we can change out that same wrist component for a longer extension and allow the customer to reach larger work envelope. Instead of just resizing the robot, just buy a robot that has some flexibility, so it can be changed out in the future if you need different size.”

The Up-Front Cost: Should you Consider Refurbished/Used Robots?

Used robots listed on Ebay.

In general, many customers tend to overestimate the cost of industrial robots. Because it’s often only possible to see the sticker price through a requested quote, it’s not widely known what the cost of robots is. Dean Elkins agrees. “That has certainly been the paradigm in recent history. I think that's changing. I think that people are getting far more educated into the realities of the robot market. I think that economy of scale, and manufacturing technologies, and just user technologies themselves are driving the cost of integration and ownership down. And people are starting to understand that, by reading and attending trade shows and looking at industry information,” he explained.

Because of this perception, it may be tempting to look into used or second-hand robots. However, while most manufacturers offer certified refurbishment services, the experts warn that this may only be a good idea in certain situations, and it’s often better, and cheaper in the long run, to purchase a new robot.

“there’s a time and a place for that. The main thing that customers need to consider is that the cost of the robot itself is actually a small portion of the total cost of automation. You’re only saving a percentage of a small percentage of the cost of the implementation. Even though you might think you've got a great deal on a used robot, honestly, you're going to end up with less life expectancy and more importantly, you're going to be behind the state-of-the-art when it comes to connectivity and compatibility with other peripherals,” said Hasley.

Dick Motley of FANUC offered a simple example. “The cadence of innovation is so fast that you have to decide if you want to take advantage of the latest technology. I'll give you an example from my personal experience. I don't need a [Smartphone] 10, but I have a [Smartphone] 7. I carefully chose the attributes of a phone that mattered to me most. Someday, I probably will have the version 10, once the astronomical price comes down a little bit, and the technology is accessible at a more reasonable price point to me personally. It's similar in robotics.”

However, Motley also advised that the ‘bang for your buck’ is rapidly increasing with newer robots, as performance and ease of use increase and cost decreases. If you’re making the investment, why not take advantage of everything a new machine has to offer?

Specially Designed Robots for Cable Management

Some tasks, such as arc and spot welding or working with certain end effectors such as pneumatic tools, require cabling or hoses extending to the flange of the robot. These can be unwieldy and difficult to manage.

KUKA KR 5-2 Arc HW

To solve this problem, models are available with a hollow wrist structure, allowing you to feed the cables through part of the arm and avoid a twisting, flopping mess. FANUC's M-10 series, the Motoman MA2010 are examples. KUKA uses the HW designation in their robot model numbers to indicate hollow wrist. For example, the KR 5-2 Arc HW has a 50mm (2 in.) opening in the arm to accommodate an arc welding dress package.

When planning for an arc welding application, or another application which will require cable management to the wrist of the robot, it may be a good idea to look for a hollow-wrist robot, advises Hasley of Kawasaki. This is partly because tucking the cables away within the wrist will make cable management easier, and partly because it will make simulation and programming much easier, as the flopping cables are not

simulated, and could behave unpredictably.

Preparing for Your Robot

When the forklift sets the robot down on your factory floor, will you be ready? There are several important preparation steps to consider before setup starts, including where and how your robot will fit into your floor space, how your employees will interact with it, and how it will be maintained.

Robotic Automation Communications Plan

In a section above, we mentioned the possibility of culture clashes brought about by the introduction of a robot. Industrial workers may be apprehensive of a robot coming in, due to the idea that robots will take jobs away from workers. On top of that, employees may be disgruntled that their workspace, daily routine, or even their primary functions are disturbed by the new robot.

It’s important to communicate with your employees and operators the goals and reasons for the robot program. In order to avoid culture problems, it can be helpful to develop a consistent communications plan to help the whole team, from executives to operators, in the loop, preventing workplace strife.

Take the time to set down answers for the following questions, as well as any others that arise:

Why are we installing this robot?

What will workers be doing differently going forward?

What opportunities for learning and training are available?

What are the company’s future plans for automation?

The entire management and project team should stay consistent to the message you set out, to ensure communication is clear for everyone.

Robot Operator, Safety, and Programming Training

Nearly every manufacturer offers a training program with courses for operators and programmers, with courses covering basic robot safety, calling and running programs, all the way up to programming the robots, using peripherals and sensor inputs, and setting up PLC control.

“Nothing is going to make a robot application go south faster than lack of training,” said Elkins. “And we have a training academy, actually. We call it Motoman Academy. When you successfully complete classes here [at Yaskawa Motoman], you're granted continuing education units (CEUs). That makes you better. It just does.” While robotics training is available to customers, that doesn’t mean it’s easy, like a first aid or workplace teamwork course. “We teach in a very structured way. The training is not ad hoc. It's really purposely direct training. And even for people that might be accredited some way. CEU's allow you to keep your accreditation active. It's a great way to learn. There's also web-based learning tools. You can receive system training through your selected integrator, but it's really important that you get training.”

If cobots are your answer, Universal Robots offers UR Academy, which includes nine online training modules. UR Academy is unique because it's completely free of charge. The entire course takes 87 minutes to complete, and promises to make you competent in handling basic programming tasks in that time.

However, it may not be enough to simply send your employees off on a training 'field trip'. “The other thing that the end user has to do is they must conduct a couple of candid assessments of themselves,” said Motley. “Number one, they must assess what is their tolerance for downtime. Even the best machinery occasionally has a bad day. If something with the robot gripper or some other peripheral or, (less common, but sometimes) the robot itself, if that experiences some kind of failure, do they have a contingency plan that would allow them to limp along until some sort of expert could come and repair the systems? Ideally, they need a comprehensive support strategy to ensure they have the right personnel trained appropriately, the right selection of spare parts, and a service plan with the robot supplier or system integrator for things they choose not to do themselves.”

List of Robot Manufacturer Training Programs

If you don't see your robot on this list, check with the manufacturer to find out more about training options.

Pitching Your Robotic Automation Project

Armed with the knowledge and planning to get your first automation up and running, it’s probably time to prepare your pitch to get your management team, board of directors, or business partners to approve the capital expenditure and fund your project.

Click the button below to download our Robotic Automation Project Starter Kit, packed with resources and expert advice for making your pitch a success.