Robotics and welding automation have been on the agenda of manufacturing companies for quite a while. Welding robots are said to remove human error and even projected to learn with high endurance, speed and precision. But is it that simple after all?

According to Progressive Markets, a global market and business consulting firm, the global industrial robotics market is expected to rise to $88,606 million in 2025, from $37,910 million in 2016. By 2020 more than 1.7 million new industrial robots will be installed in factories worldwide, estimates International Federation of Robotics. As we can see, robotics will play a central role in the business logic of future manufacturing and welding, but how should we actually evaluate their return on investment?

Usually, investment payback time is calculated simply by comparing the cost of the produced goods before and after investment. When selecting, for example, welding wire for the production, this is approach simple enough. But with automation, wider perspective is required.

As a system thinker, I like to propose collecting all the pieces on the table before jumping into conclusions. Welding is the art of processes, materials and handcraft. Together with new production equipment, also production planning and scheduling as well as welding skills must be modified to serve a selected production concept. And that production concept is going to be changed when automation is introduced to the shop floor.

There are at least three cost categories related to the production concepts, which should be discussed when investments on automation are in question:

• Direct costs related to used resources per production stage (these are the foremost on everybody’s mind when discussing welding costs)
• Costs emerging from the current way of operating, which, however, are related to the welding work, and also
• Costs related to availability of needed resources over period of investment life time, including the effect of the ever-repeating learning curve.

In short, automation affects the whole production concept, and thus also current and coming cost structures should be studied accordingly.

Shortening the learning curve in welding

Finally, the effect of the learning curve is interesting. Typically, when something is done for the very first time, it requires even double the standard estimated production time. Depending on the complexity of the novelty, we need several exercises before achieving the standard time. I’ve written more about these different cost types in my white paper Investment calculations and production automation in welding.

What could we do to shorten our learning curve and to increase our productivity faster, then?

At Kemppi, we have developed advanced, easy-to-learn and easy-to-use user interfaces, added connectivity possibilities and included a lot of supporting functions like a gas flow sensor, and improved ignition routines within our welding systems. A good example of this is our A7 MIG Welder, and Kemppi’s Wise software solutions are another. They provide means to focus the arc and ensure the penetration in very reliable ways, helping to guarantee the steady quality of the welds straight from the start. What is common to all our solutions is combining deep welding process knowledge with understanding of opportunities in welding automation from both the welder’s and the robot operator’s perspectives.