eWEEK content and product recommendations are editorially independent. We may make money when you click on links to our partners. Learn More.
STUTTGART, Germany—I was transfixed as the huge yellow robot swung the front portion of a car’s chassis around as if it weighed nothing. Then, the robot stopped and with the future car held steady, two big electrodes came together and sparks flew, and as the smoke cleared, the robot moved the part a few inches, and did it again. I stood next to the Mercedes Benz representative and muttered, “Amazing.” We moved on and watched another robot making spot welds, at one point stopping to sharpen its electrodes, test them and then go back to work.
Bjorn, the Mercedes representative, explained while I followed him to another station in the vast factory that Mercedes likes to call the Welding Shop that each part of what will eventually become a complete Mercedes Benz automobile is tracked, and that every reading for every weld is stored in a private cloud for further analysis. He said that this means that everything that’s measureable—the voltage and amperage used for each weld, the pressure of the electrodes on the steel of the chassis, how recently the robot had cleaned each electrode and sharpened it—was stored for every part of every car.
I was on a public factory tour of the Mercedes Benz factory in Sindelfingen, Germany, along with my friend and colleague Alan Zeichick. This suburb of Stuttgart is one of the largest automobile factories in the world. Despite the factory’s huge size, everything is interconnected, which is the case with most modern production facilities, but there’s more to the interconnection than meets the eye.
Sure, the cars grow as they move from place to place, and separate production lines take care of subassemblies as you’d expect, but beneath all of the magic of seeing a modern vehicle being put together, there is a rhythm that’s conducted by cloud-linked robots and other production machinery.
By using the cloud and some dedicated software that was developed by the big German auto makers, including BMW and Volkswagen as well as Mercedes, the entire process is orchestrated so that parts are produced, assembled and then brought together in a single, unified flow. Each part that will become the car is tracked to the original order, and as each of those assemblies is tracked, the cloud-based production system tracks every part of each one.
Bjorn demonstrated how this might work by picking an incorrect part from a bin and bringing it to the assembly of a dashboard for an S-Class Mercedes. The computer-monitored assembly process tracked each of the parts, but when he tried to install it into the wrong dashboard, an alert went off. Then when Bjorn went back to the parts bins, a light on the bin showed him where to put the part in his hands, and then another light showed him where the correct part was located.
When the correct part was placed into the dashboard, the production system noted that and sent a record into the cloud. As the production workers assembled that dashboard, each of those parts was tracked, and then, using smart screwdrivers, the movements of each worker were tracked, and the placement of each screw was measured and recorded, right down to the torque reading on each screw.
Mercedes Turns to the Industrial Strength Cloud for Quality Assurance
I asked Bjorn about the amount of data that Mercedes was storing on each car. Apparently, this is proprietary information because he wouldn’t say, but he did say that every piece of information is stored in the Mercedes private cloud for 20 years. And this is for every vehicle that the company makes, from each factory and production facility around the world. This clearly is a very large big data project.
The company uses this data to determine the performance of each part and subassembly of every car over time. This means that if a trend develops, Mercedes can use big data analysis to figure out exactly what the cause of failure might be and then make sure that it doesn’t happen again.
It might be that a specific subassembly in a specific part begins to show early failure several months after full-scale production of vehicles starts. The company is able to tell exactly when and where the parts within the assembly were made, where they were assembled into a larger part of the car, when that took place and what else might have contributed to producing that part.
With that level of detail, the company can take corrective action, and at the same time issue a service bulletin to dealers and service centers explaining the issue and what action to take.
Meanwhile, Mercedes could track a point of failure to a specific welding machine that might be a little out of calibration or to a production team that might need additional training. Because every detail is stored in the cloud, and is there to be analyzed, it means that the company can be proactive about ensuring quality.
In effect, Daimler Benz, which makes Mercedes vehicles, is experiencing much of the promise of big data, and it’s showing what thoughtful use of that data can accomplish.
But I won’t pretend that my motivation for applying for a spot on the tour was strictly an examination of big data. I’m enough of a car guy to get a thrill from seeing a camouflaged Mercedes prototype on the test track, or getting to see the new S Class convertible before anyone else I know. Sadly, Mercedes wasn’t giving out free samples, but it is good to know that the company is using big data to make the cars as good as they could be.