Abagy Robotics Systems Logo

Member Since 2019


Abagy developed robotic welding programming software and offers robotic systems for all manufacturers, even for custom projects and one-of-a-kind parts. Forget about the welder shortage!

Content Filed Under:



Max Zverkov, ABAGY: “To solve the shortage of welders, robots must be adaptive”

POSTED 10/04/2022

The US is projected to face a shortage of 400,000 welders by 2024. A possible way out is robotization. However, it’s not that simple. Max Zverkov, CEO at ABAGY, which developed software for autonomous robotic welding, explains why.

The American Welding Society (AWS) data on the welder shortage is distressing. How often do you hear from manufacturers about this problem? 

When we talk to companies, we make various arguments in defense of robots. For example, we calculate the return on investment (ROI) and the efficiency of robotic welding per meter compared to manual welding, but the strongest argument is the ability to solve the problem of qualified welders shortage.

We can recall one more fact from the AWS data. Right now, 80% of welders are over 35 years old, and young people do not want to take on this dangerous and hard work. This means that the situation will only get worse. 

In this case, why are industrial companies in no hurry to robotize production?

An indicator of the robotization level in a country is the number of robots per 10,000 workers. According to the International Federation of Robotics (IFR), South Korea is leading here, with a rate of 932 robots per 10,000 workers. In the US, the number is 4 times lower – 255 robots per 10,000 workers. This means that there are still very few robots in factories around the world. 

The reason is many companies are sure that robotization is not suitable for them. For a long time, it was believed that only mass production, where thousands of identical products are produced, can be robotized. Therefore, the first industry where robots were actively used was automotive. 

High-mix non-serial productions considered that robots could not be effective for them. The problem was in the robot programming. Imagine that you had a crane beam or trailer frame, and each time it was pre-assembled by people it was not perfect. The small deviations each part had made it look like a unique product to a robot. In the traditional approach, you needed 2-3 days to program each part, while welding took only several hours. Of course, they didn't want to use robots.

But today, our company has solved the problem of programming. So, even if you produce one-of-a-kind products, robots can be cost-effective for you. 

Now the problem is how to change the persistent opinion about the impossibility of robotizing the non-serial market. And we try to explain this to as many companies as possible.

How do new technologies for robotic welding work?

ABAGY automatically generates trajectories for robots. All you need to do is load a 3D model of the product, select the welds, and set the necessary welding parameters. Thus, programming is not required.

Then, using machine vision, the system scans the parts and adjusts the robot's trajectories depending on the actual position and deviations of the product. Thus, with us, robots become flexible and adaptive.

Our software is compatible with all major robot manufacturers, such as Fanuc, Kuka, Kawasaki, Yaskawa, and Panasonic. It also works great with various welding equipment, including Lincoln, Miller, Fronius, Esab, etc.

You said that the teach pendant programming developed by the robot manufacturers did not allow the use of robots in non-serial production. But there is also offline programming. How are you different from these?

Programming with a teach pendant takes place at the production site. Meaning, at this moment the robotic cell cannot work. Offline programming solved the problem of robot downtime but did not solve the problem of programming. Offline systems moved the complex process of programming into a virtual environment. And it still takes a lot of time and money. After creating the program, the programmer must also check it on a real robotic cell. In our case, as I noted earlier, there is no need to program at all. The system transfers the data on the fly: robots "see" and complete the task.

Can you give examples of non-serial industries that have started using robots?

I can refer to your research. The Association for Advancing Automation (A3) published data that in 2021 non-automotive orders represented 58% of the North American total robot units sold, in particular, Metals: up 91% over 2020. 

We are discussing robotization with such industries as Shipbuilding, Bridge structures fabrication, Heavy Construction Equipment, Trailers, etc. 

You also have the option of retrofitting existing robots, how does that work?

I can give an example. ABAGY has retrofitted a robotic cell for Schenck Process which produces equipment, for example, rotors. The robotic cell was operating at the Sabetha Kansas production facility. ABAGY installed machine vision and software. As a result, the setup time has been reduced dramatically – only 10-15 minutes for a new product – and the robot can be used for many more products. In other words, you can increase the utilization of your robotic cells.

We see a great response from the market. We think that the adaptability and flexibility of robots will allow us to multiply their number in production, and thereby solve the problem of the welder shortage that started this conversation.