Pages

February 14, 2011

The current and future economics of 3D printing and factory production - companies like Zara that leverage new capabilities for new business models

There has been an active debate on this site about 3D printing/Additive manufacturing and if there is a significant shift that will cause localization of production. Localization of production meaning something like people making a lot of things at home.

The Economist magazine has more facts and analysis.

Engineers and designers have been using 3D printers for more than a decade, but mostly to make prototypes quickly and cheaply before they embark on the expensive business of tooling up a factory to produce the real thing. As 3D printers have become more capable and able to work with a broader range of materials, including production-grade plastics and metals, the machines are increasingly being used to make final products too. More than 20% of the output of 3D printers is now final products rather than prototypes, according to Terry Wohlers, who runs a research firm specialising in the field. He predicts that this will rise to 50% by 2020.

So the shift that is happening is still a slow and modest shift.

Slightly cheaper components is not enough reason to radically alter a larger manufacturing process but could mean that 3D Printers or additive manufacturing could be included for some steps of a factory process. A larger driving factor would be when additive manufacturing can make things that have superior performance and critical features versus traditional manufacturing.

Being able to make new products using additive manufacturing that could not be built before will be where there will be clear growth.



Additive manufacturing making lighter parts which is valuable for airplanes

Using 3D printers as production tools has become known in industry as “additive” manufacturing (as opposed to the old, “subtractive” business of cutting, drilling and bashing metal). The additive process requires less raw material and, because software drives 3D printers, each item can be made differently without costly retooling. The printers can also produce ready-made objects that require less assembly and things that traditional methods would struggle with—such as the glove pictured above, made by Within Technologies, a London company. It can be printed in nylon, stainless steel or titanium.

Aircraft-makers have already replaced a lot of the metal in the structure of planes with lightweight carbon-fibre composites. But even a small airliner still contains several tonnes of costly aerospace-grade titanium. These parts have usually been machined from solid billets, which can result in 90% of the material being cut away. This swarf is no longer of any use for making aircraft.

To make the same part with additive manufacturing, EADS starts with a titanium powder. The firm’s 3D printers spread a layer about 20-30 microns (0.02-0.03mm) thick onto a tray where it is fused by lasers or an electron beam. Any surplus powder can be reused. Some objects may need a little machining to finish, but they still require only 10% of the raw material that would otherwise be needed. Moreover, the process uses less energy than a conventional factory. It is sometimes faster, too.

There are other important benefits. Most metal and plastic parts are designed to be manufactured, which means they can be clunky and contain material surplus to the part’s function but necessary for making it. This is not true of 3D printing. “You only put material where you need to have material,” says Andy Hawkins, lead engineer on the EADS project. The parts his team is making are more svelte, even elegant. This is because without manufacturing constraints they can be better optimised for their purpose. Compared with a machined part, the printed one is some 60% lighter but still as sturdy.

Lightness is critical in making aircraft. A reduction of 1kg in the weight of an airliner will save around $3,000-worth of fuel a year and by the same token cut carbon-dioxide emissions. Additive manufacturing could thus help build greener aircraft—especially if all the 1,000 or so titanium parts in an airliner can be printed. Although the size of printable parts is limited for now by the size of 3D printers, the EADS group believes that bigger systems are possible, including one that could fit on the 35-metre-long gantry used to build composite airliner wings. This would allow titanium components to be printed directly onto the structure of the wing.

MIT Schmidt - it should be possible for a robot builder to specify what a servo needs to do, rather than how it needs to be made, and send that information to a 3D printer, and for the machine’s software to know how to produce it at a low cost.

Neri Oxman, an architect and designer who heads a research group examining new ways to make things at MIT’s Media Lab. She is building a printer to explore how new designs could be produced. Dr Oxman believes the design and construction of objects could be transformed using principles inspired by nature, resulting in shapes that are impossible to build without additive manufacturing.

Some 3D systems allow the properties and internal structure of the material being printed to be varied. This year, for instance, Within Technologies expects to begin offering titanium medical implants with features that resemble bone. The company’s femur implant is dense where stiffness and strength is required, but it also has strong lattice structures which would encourage the growth of bone onto the implant. Such implants are more likely to stay put than conventional ones.

Shapeways, a New York-based firm spun out of Philips, a Dutch electronics company, last year, offers personalised 3D production, or “mass customisation”, as Peter Weijmarshausen, its chief executive, describes it. Shapeways prints more than 10,000 unique products every month from materials that range from stainless steel to glass, plastics and sandstone.

Loughborough University has invented a high-speed sintering system. It uses inkjet print-heads to deposit infra-red-absorbing ink on layers of polymer powder which are fused into solid shapes with infra-red heating. Among other projects, the group is examining the potential for making plastic buckles for Burton Snowboards, a leading American producer of winter-sports equipment. Such items are typically produced by plastic injection-moulding. Dr Hopkinson says his process can make them for ten pence (16 cents) each, which is highly competitive with injection-moulding.

Loughborough’s process is already competitive with injection-moulding at production runs of around 1,000 items. With further development he expects that within five years it would be competitive in runs of tens if not hundreds of thousands. Once 3D printing machines are able to crank out products in such numbers, then more manufacturers will look to adopt the technology.

Optomec makes similar claims of being competitive for making units up a few thousand One of the main drivers for cost reduction using the Aerosol
Jet process is the elimination of physical tooling. The Aerosol Jet process can reduce the overall number of processing steps, which in turn can help to reduce both capital and operating costs. Since the system can process a wide range of materials and substrates, greater utilization of the capital equipment can be obtained.


Optomec's Aerosel jet is not inkjet


Optomec also has Laser Engineered Net Shaping (LENS) LENS uses a high-power laser (500W to 4kW) to fuse powdered metals into fully dense 3-dimensional structures. The LENS system can process a wide variety of metals including titanium, nickel-base superalloys, stainless steels and tool steels - all of which are commercially available in the required powder form.


Successful application of additive manufacturing could improve time to market and time to volume

It is an industrial norm that a delay of a few months makes a whole program unprofitable. Time-to-Market and Time-to-Volume are primary concerns of the handset industry (cellphone). Time-to-Market is critical due to launch costs and contractual obligations. Reliable Time-to-Volume is vital in that it allows OEMs to fully achieve their Time-to-Market targets. Time-to-volume is the time it takes to get a product to achieve full scale volume manufacturing.

Will Sillar of Legerwood, a British firm of consultants, expects to see the emergence of what he calls the “digital production plant”: firms will no longer need so much capital tied up in tooling costs, work-in-progress and raw materials, he says. Moreover, the time to take a digital design from concept to production will drop, he believes, by as much as 50-80%. The ability to overcome production constraints and make new things will combine with improvements to the technology and greater mechanisation to make 3D printing more mainstream.

Printing in 3D is not the preserve of the West: Chinese companies are adopting the technology too. Yet you might infer that some manufacturing will return to the West from cheap centres of production in China and elsewhere. This possibility was on the agenda of a conference organised by DHL last year. The threat to the logistics firm’s business is clear: why would a company airfreight an urgently needed spare part from abroad when it could print one where it is required?

As has been noted in the prior discussion, if you are shipping product by the container load then additive manufacturing will not make a difference on shipping costs or time to market for a slow boat product. However, lower volume parts that are being sent via overnight air delivery would be a place to look.

Zara fashions is a Harvard Business case study about a company whose core advantage is being able to rapidly change fashions.

The firm tripled in size between 1996 and 2000, then skyrocketed from $2.43 billion in 2001 to $13.6 billion in 2007. By August 2008, sales edged ahead of Gap, making Inditex the world’s largest fashion retailer.

Zara’ duds look like high fashion, but are comparably inexpensive. A Goldman analyst has described the chain as “Armani at moderate prices”, while another industry observer suggests fashions are more “Banana Republic”, prices are more “Old Navy”. Offering clothing lines for women, men, and children, legions of fans eagerly await “Z-day”, each Zara location’s twice weekly inventory delivery that brings in the latest designs.

Having the wrong items in its stores hobbled Gap for nearly a decade, but how do you make sure stores carry the kinds of things customers want to buy? Try asking them. Zara’s store managers lead the intelligence gathering effort that ultimately determines what ends up on each store’s racks. Armed with handheld personal digital assistants (PDAs) to gather customer input, staff regularly chat up customers to gain feedback on what they’d like to see more of. A Zara manager might casually ask: What if this skirt were in a longer length? Would you like it in a different color? What if this v-neck blouse were available in a round-neck?

PDAs are also linked to the store’s point-of-sale (POS) system, showing how garments rank by sales. In less than an hour, managers can send updates that combine the hard data captured at the cash register combined with insights on what customers would like to see.

Rather than create trends by pushing new lines via catwalk fashion shows, Zara prefers to follow with designs where there’s evidence of customer demand.

Data on what sells and what customers want to see goes directly to “The Cube” in La Coruña, where teams of some 300 designers crank out an astonishing 30,000 items a year versus 2,000-4,000 items offered up at big chains like H&M (the world’s third largest fashion retailer) and Gap

The average time for a Zara concept to go from idea to appearance in store is 15 days vs. rivals who receive new styles once or twice a season. Smaller tweaks arrive even faster. If enough customers come in and ask for, say a round neck instead of a “v” neck, a new version can be in stores with in just 10 days. To put that in perspective, Zara is twelve times faster than Gap, despite offering roughly ten times more unique products! Contrast this with H&M, where it takes three to five months to go from creation to delivery – and they're considered one of the best. Other retailers need an average of six months to design a new collection and then another three months to manufacture it. VF Corp (Lee, Wrangler) can take 9 months just to design a pair of jeans, while J. Jill needs a year to go from concept to shelves. At Zara, most of the products you see in stores didn't exist three weeks earlier, not even as sketches

Some industries will not need the advantages of rapid product changes like fashion, but many consumer products like smartphones and some categories of consumer electronics could benefit.

So more valuable and new kinds of parts for airplanes and satellites and lighter cars like the Edison 2 and impact where time to market and rapidly shifting styles matter. The home markets look to be tiny since the overall hobby and DIY markets seem more limited by the knowledge and time for the consumer. The home markets could have some long tail niches.

If you liked this article, please give it a quick review on ycombinator or StumbleUpon. Thanks
blog comments powered by Disqus