As markets emerge and mature, patterns develop. The latest pattern is automated teamwork. Not bringing automation to the teamwork of people, but allowing the different technologies to contribute to teamwork in getting something done.
It did not dawn on me as a concept until I was writing up the story on the AMADA 6020 ATCe press brake. (Spoiler Alert) The automatic tool changer (ATC) puts the tooling in the right stations. The HMI tablet automatically moves to the right spot (so does the foot pedal), and shows the operator a video view of the bend area shown from above it. An augmented reality outline of the part is superimposed on the image so the operator knows how to position it. As the bending starts, a sensor that checks the bend angle is deployed. It gives bend angle information both to the operator and to the system; it knows just how much springback will move the bend back to the perfect angle and all of this is shown on the HMI tablet.
If you look back at that paragraph, not only is it a lot of working parts, it’s a lot of working concepts. I am pretty sure that in designing a system like this, the engineering team had to be very specific about lines of demarcation. These lines then affect the interfaces between these systems, all the way down to the physical layer of the OSI stack. Should it be an Ethernet port? A single line for serial information? A multi-output device with something like software “ports?” As architects (some, anyway) say, “form follows function.”
This strategy does in fact follow human teamwork to a certain degree. We humans have lines of demarcation, laterally as well as vertically, and those lines must be maintained for the whole to work as a single body or organism.
The lines are moving quickly now, however. Autonomous mobile robots (AMRs) and cobots are two very recent additions to the fabricating landscape, and are making these lines more virtual than real. If you think of combining an AMR and a cobot, you can have a cobot finish work in a loading operation in cutting, only to move to unloading in the bending department. In the future, perhaps an attachment change followed by a 10-minute initiation, and that cobot becomes a welder. It’s hard to say right now what capabilities are coming.
If we want to trace the roots of these concepts, we look past Deming and the production gurus and instead look a little further back to quantum physicist Richard Feynman. His tasks in the Manhattan Project included calculating absolutely hairy equations that were complex in their inputs. For example, one item that was needed for some of the equations was the density of reactive material that could maintain a chain reaction. That little number would help the team with items like how to keep the material stable until use, and how much mass was critical mass, all sorts of things important to such a task.
Feynman was quite a genius and he saw the slogging progress of these huge equations. One day he thought he could speed things up by separating the problem into blocks of more easily-solved equations, and that was a tremendous boon to the Manhattan Project team in meeting their deadlines.
Essentially, this same concept is happening now, but in automation rather than strictly in mathematics. The Automate Show next week should provide me with ample examples of this divide-and-conquer concept, which I will share as soon as I return.