Camshaft to all: I´m okay!
future of production | At Ilsenburg , the physical world of things is fused with data networks to form a cyber-physical system.
Ilsenburg on the edge of the eastern Harz area. A large hangar in an industrial zone. Robots are moving about busily in assembly cells, indicator lamps are flashing red and green, and forklift trucks are on the go between yellow lines. In the middle of all this, a laser pointer is burning coded information onto a camshaft. This looks like an incidental process, but for the product it is a crucial step in its short life – and, for thyssenkrupp Presta in Ilsenburg, this marking process is symbolic of their vision of manufacturing for the future. Here, this future has already begun.
A little earlier, the individual parts of the camshaft – shafts and cams – were still lying in large boxes, which likewise carried a bar code. The coded numbers of the batch had been affixed by the supplier. Before the individual parts were allowed into the first cell, their batch number was stored on an RFID chip on the small mounting trolley. The steel shaft and the intake and exhaust cams then disappeared into a hermetically sealed assembly cell, where they were “married up” – as they say here – by an industrial robot. After that, the cams were positioned immovably on the shaft at the correct angle to, and distance from, each other. Then came the crack testing.
“Up to this point, all we know about the camshaft is the batch from which its individual parts originate,” says Manfred Muster, division manager for project handling. It is still a non-entity – but as long as no cracks are found, the camshaft will now receive an identity: It will get a “name” lasered onto its side. From this very moment and for the rest of its life, the camshaft will carry in a Data Matrix code (similar to a QR code) information such as serial number, drawing number, part number, and customer identification. It now has proof of identity, because nothing can happen without a name from this point on: Every camshaft has to be registered individually with the next machine on its journey. “In this way, the machinery knows whether the product has the correct status and is actually allowed to be assembled on this machinery, or whether a prior step in the process has been missed,” Muster says.
Individualized products, not mass production
In this way, the physical world of things is fused with data networks at Ilsenburg to form a single system, a “cyber-physical system” – and this is exactly how the future of industrial production is intended to look. The prerequisite is that as many elements in the production process as possible should have a name, a history, and a web interface. The goal is the “smart factory,” which is self-controlling, adaptable, and flexible. What flexibility means here, among other things, is the ability to manage small batch sizes. This means a move away from mass production and toward individualized products.
One product that is already made in a very individualized way is the automobile, and pressure to innovate is correspondingly high in that industry: Whereas only 31 percent of companies in mechanical and plant engineering use “intelligent” production facilities, the figure for automotive suppliers is as high as 80 percent, according to a survey conducted by the market research firm Pierre Audoin Consultants in 2013 on behalf of the IT service provider Freudenberg IT. “Automotive suppliers already need to be capable of guaranteeing on-demand production for their customers,” it says. This will soon also be the case for many other sectors.
The Ilsenburg plant is already providing a foretaste of this. In some places here, there are five large industrial robots in one cell working at high speed, presenting each other with parts to be cooled, heated, aligned, and pressed. Workers have one main task: to listen in to the automatic processes as they proceed, with bar code or Data Matrix code scanners, RFID readers, and tablets with data-acquisition modules. What they are monitoring here is not individual machines or processes, but the whole system – and a very modern one, too.
Around 650 workers at this plant are engaged in assembling camshafts. Unlike casting or forging, the process of assembling makes it possible to combine a variety of materials and, for example, to use high-strength materials in the cams. This process is cheaper, and it is possible to reduce the weight – but, on the other hand, this way of manufacturing is more complex. One kilometer away, at the new thyssenkrupp Valvetrain production plant, they produce integral modules with the camshafts already built into the cylinder head cover. Around 150 workers are employed here. The two Ilsenburg plants together turn out around 11.5 million camshafts per year. They are global market leaders.
At both plants, the codes on the modules are checked again and again before the next step in the production process can happen – and then the decision to stop or go is taken immediately. “There are various exclusion criteria: NA, NIO, or ‘No number’,” Muster says. NA means “Not Applicable,” and NIO stands for “Not In Order.” To demonstrate the point, he swaps over an intake and an exhaust cam. By return, the next assembly cell sends the incorrectly sorted module back: “NA!”
Each camshaft has an identity and is linked to the net through an interface. It also possesses a memory. Muster calls the database behind this the “formulation,” and because it is instantly available at all times, it is easier to modify subsequent steps in the process in order, for example, to achieve better manufacturing tolerances. Assembly lines and machines can be influenced by a new “formulation” at any time. This is flexibility as understood in Industry 4.0.