Made-to-measure – by assembly line
How does industry work in the digital age, and what does that mean for the employees? We visit the cleaning technology firm Kärcher.
“Any customer can have a car painted any colour that he wants, so long as it is black.” Henry Ford’s comment sums up the problem of industrial mass production. It makes things efficiently and quickly – but they are always the same.
However, Kärcher’s B40 scrubber-drier is different. Made in Schwabia in south-west Germany, the B40, like its big sister, the B60, is about as big as a shopping trolley, though much bulkier. With a manually-controlled electric motor, it swabs floors and immediately vacuums them dry again. Customers can choose rollers or discs for the wiping process, decide how broad or narrow these should be, what sort of steering they want and whether to have automatic interior tank cleaning: in fact, customers can choose from roughly 40,000 different combinations. In the end, practically no two machines ordered are alike. The customer can even choose the colour – though most opt for anthracite and the classic Kärcher yellow.
“In the old days we used to have a catalogue with just a very few variations,” says Carsten Schlenker, factory manager at the firm’s main location in Winnenden, near Stuttgart. “An order picker would get the necessary parts ready in a cart and then the machine was assembled from them on the production line.” By “the old days”, he means before 2015. Since then industry 4.0 has arrived in Winnenden.
Over the decades the Kärcher company premises have fused with the town: there are modern office buildings and a handsome auditorium. But some departments are based in former houses the firm acquired as it expanded.
Yet though a lot may look very traditional, production here is extremely modern. In cooperation with the Fraunhofer Institute for Industrial Engineering in Stuttgart, Kärcher has developed a completely digitalised assembly line for its mobile cleaning juggernauts. That does not mean that robots have taken over. Assembly is still done by the employees, but they are assisted by computer technology. Instead of the “picker” going around selecting the necessary components, all the parts for all the options are ready to hand at the line. Indeed, far too many parts for a human to keep track of. This is done by the emerging product itself: it knows which variant it is destined to be and what parts are required from a radio chip on the assembly trolley linked to the ordering system.
At each of more than 12 assembly points, the chip transmits a signal to the storage rack of components. The relevant compartments light up in different colours to signal what parts are to be selected and installed. In this “pick-by-light” procedure, the compartment with the component required next lights up in green. Parts to be assembled later show up in blue. Boxes containing nothing that is required immediately stay dark. If a worker delves into the wrong box, a red warning light comes on.
The component lists and packing slips of the old days are a thing of the past. A screen displays the parts to be installed at this point in the process. It can also show pictures and videos for more complicated stages (or less experienced workers). And at the end of the assembly line, a radio signal means the testing station knows precisely what the customer has ordered for this particular machine and what functions must be tested and how.
There are 10 production lines in the Winnenden factory, five of them for floor cleaners, including the first for Kärcher to be converted to modern digital technology. This line turns out some 8,000 wash-and-dry machines a year – about 60% of all floor cleaners to leave the Winnenden works. It took a year for the new line to be created alongside the working production lines. A provisional one was tested using cardboard boxes. “That allowed us to be flexible testing what works and what doesn't, and we could alter individual processes and stations quickly and simply,” says Wolfgang Thomar, the divisional manager of production engineering. “We were able to include a lot of feedback from the workers, ranging from ergonomic requirements to the removal of packaging.“
From mechanical engineer to IT expert
An individually assembled Kärcher scrubber-drier is finished in 120 minutes – about 30 minutes faster than under the old method. “By doing away with the picking process, we were able to eliminate a source of error,” says Matthias Braun, a shift supervisor. “I can also react more flexibly and complete a particularly urgent order in one or two days. Previously, everything was fixed about a week in advance.” Kärcher now promises to deliver any appliance within 14 days at the latest. “But we deal with 90% in less than a week,” says Braun. Along with changes in production, his job description is changing too: away from traditional mechanical engineer to IT expert.
Other jobs have disappeared completely, such as the piece picker who sorted the small components into the right boxes. Two of the eight employees a shift who used to work here are no longer needed.
Fortunately, business is good for Kärcher. It is a family firm and does not reveal the size of its profits, but in 2016 turnover rose by 5%, well above the average for the sector, to about €2.33bn. That is why workers who are no longer needed in one section can be retained without problem and redeployed in another, says Schlenker, the works manager. At the end of 2016 Kärcher employed 11,862 staff, about 500 more than the year before and about twice the number of 10 years ago.
Progress in industrial production is usually discussed in terms of complete automation. Tales of successes circulate, such as Changying Precision Technology, a mobile phone factory in Dongguan, China, which replaced 90% of its staff with robots and achieved a 250% production increase and an 80% reduction in errors. Kärcher is also working with a very high rate of automation – but only with its standard products. These are the “yellow group” of portable high-pressure cleaners or window vacuum cleaners that have for years accounted for about 50% of the company‘s turnover. Only with these “end-user products” are the quantities high enough for automated production. They do not have the range of variants seen in the scrubber-drier machines for business customers, the “grey group”.
The stack takes charge of reordering
“Kärcher has succeeded in combining the best of both worlds: mass production and individual customised production,” says Hermann Simon, a former economics professor who coined the term “hidden champions” (brand eins 10/2012, “Wachsen in der Nische“). Today, he is consulted by firms all over the world. “Thanks to digitalisation,” says Simon, “firms are regaining the opportunities for individual design options without having to forgo the advantages of automation.” But this kind of switch is fraught with risks and is not simple. “It makes the supply chain and stock-keeping significantly more demanding,” says Simon. “Every single component has to be permanently available, even those that are only installed in one of these 40,000 different product variants.”
Kärcher, too, is seeking to solve this problem through digitalisation and networking. All the storage stacks “know” how many of each individual component they contain. If the stock drops below a certain level, a signal is transmitted for restocking or reordering. Many parts (such as the big plastic tank-in-tank systems of the pressure cleaners) are made in the company's own factory. But screws come from another world market leader: at a digital signal from the storage stack, the firm of Würth in Künzelsau in south-central Germany supplies all the screws needed for assembly and sorts them directly at the assembly line into the relevant area of the stacks.
To some extent Kärcher has simply outsourced the tiresome and time-consuming job of picking parts to its supplier. On the other hand, this is a perfect example of how the revenue models and therefore working conditions of medium-sized firms are changing. The proportion of “added value“ that basic manufacturing accounts for at firms such as Kärcher or Würth has shrunk in recent years and will continue to do so. In contrast, there is growing importance in other areas, such as research and development, design, service, and even disposing of returned goods. That means not merely making the screws but placing them on the client’s shelf in good time and as needed; not just manufacturing the B60 scrubber-drier but also servicing it – ideally before it breaks down with a fault, but not unnecessarily early either. That requires networking and the sensors with which Kärcher equips its products. “Our focus used to be on making machines and selling them,” says Wolfgang Thomar. “Now the sale is the start of a chain of services we offer our customers.”
In the market for professional cleaning appliances Kärcher has two kinds of clients: the firms that buy and use the machines, and those where the cleaning is carried out. Kärcher offers both the opportunity to retrieve information through sensors in the cleaners. The cleaning contractors can check the state of their fleet, its use and similar parameters, while their customers can see when, where and for how long cleaning took place.
“Through the sensors in a machine we can also see when something is not working perfectly and send out a service technician,” says Thomar. “That boosts customer satisfaction with the cleaning firms because there are fewer breakdowns.” Preventive maintenance is the specialist term for this kind of intelligent maintenance, which experts such as Hermann Simon say is becoming increasingly important. The US multinational General Electric, for example, has achieved a 30% to 40% increase in the availability of its machines for business customers extracting oil or gas.
There are other examples of how digital networking can open up better service and new business areas: “We supply a major chain of discount markets with our scrubber-driers,” says Thomar. “When we see that it’s going to rain in places where this chain has markets, we know customers will tramp more dirt into the stores. That means that tomorrow our machines won’t be used just once but perhaps three or four times.” It’s also possible to check the need for additional deliveries of cleaning products in good time.
Fewer blue collars, more grey ones
In the USA there is a traditional distinction between white-collar (office) and blue-collar (overalls) jobs. Will developments like the made-to-measure assembly line in the Winnenden works lead to classical medium-sized employers such as Kärcher needing more and more white-collar instead of blue-collar workers? Plus a few robots? “We will always need workers, including in production,” says Thomar. “But there will be a shift: at present we achieve less than 10% of our turnover from service. One day it will be more than 30%.” That will impact on the labour force. “The traditional maintenance man, servicing the production machines with a spanner in his hand, will certainly be a less common sight in future here at Kärcher – in his place there will be more people with laptops.” Many office jobs will also be affected by the change. “I think the biggest increase will be in grey-collar workers – the well-qualified technicians. Both white-collar and blue-collar jobs will migrate in that direction.”
Clearly, training is crucial, not just for Kärcher but for industry as a whole. So Wolfgang Thomar is especially proud of the company’s “Lean“ Academy, an internal training department through which all staff must pass, with varying degrees of intensity and focus. The term “Lean” was coined by John Krafcik of MIT, while he was part of the International Motor Vehicle Program. During a study of Japanese companies, he observed that Toyota did “everything with half of everything”: half the people, half the space, half the inventory, half the resources, yet with very high quality. Not knowing what to call the method or approach that Toyota followed in their Operations, Krafcik used the word “Lean”.
Lean Management concepts such as kata (improvement and coaching), heijunka (production training) or poka yoke (error-proof) are a sign that it is mainly Japanese methods of business organisation and increasing production that are taught here. Managers have to experience what it is like on a mock-up assembly line. Workers have to understand what value-stream planning entails and the meaning of those performance indicators that measure what they are achieving.
In a sign of the importance the firm attaches to further training, there are similar Lean academies at the Kärcher works in Mexico, Brazil, the USA, China and Romania. “Process-related thinking will become more and more important,” says Thomar. “I myself am the best example. I'm actually an engineer, but it is my job to design processes. Because only when you have a stable process does it make any sense to digitalise it.”
Industry 4.0 stands for the fourth industrial revolution, following on from the development of mechanical production driven by water power or steam, then the division of labour at the assembly line, and the automation of production. Key concepts include networking machines and goods with the help of sensors (the internet of things) and so-called cyber-physical systems, which can make their own decisions based on sensor information.
Batch size 1 production
In mass production identical or similar orders are usually grouped into a single batch so that the production line needs the smallest number of changeovers to make the greatest number of identical items. Batch size 1 production means this is no longer necessary, as every item can be produced individually without additional outlay. This is especially relevant when a great number of configuration possibilities would make batch-optimised production very elaborate.
Kata, heijunka, poka yoke
Concepts from Lean Management. Kata describes a continuous process of improvement based on systematic learning and behavioural routines. Heijunka is a way of levelling out fluctuations in market demand and avoiding waste and bottlenecks. Poka yoke is about using technology to prevent mistakes (such as “pick by light” systems on assembly lines or the different-sized pump nozzles at filling stations, to alert customers accidentally choosing the wrong fuel).
A method for designing and testing assembly processes and workplaces, also known as “mock-up workshops” . Workshops or whole assembly lines are created using simple cardboard boxes or plastic dishes. Procedures are tested and modified on the basis of employee feedback. Changes can be made rapidly and it is easy to try out alternatives, avoiding many later adjustments to the real assembly line. Involving employees at an early stage is also seen as an advantage, as they are then more likely to accept the changes.