A number of conferences have been held recently dealing with the general topic of “mass customization”. The main thrust of mass customization is that because an ever-increasing portion of manufacturing is software-driven, intuitively it should be possible to easily customize virtually any product. However, in some ways mass customization is encountering the same kinds of obstacles to widespread adoption that hamstrung the infant videotext field when it arose as precursor of the Internet. Therefore, to help identify obstacles to widespread use of ‘mass customization’ it’s useful to review why videotext didn’t ‘catch fire’ while the internet did. We’ll also discover why the evolution of software-defined radios (SDR’s) may help business evolve into what I call SDMC’s – software defined manufacturing companies.

In 1985, before the Internet became widely utilized, many major banks and commercial consumer interests (credit card companies, for example) were busily promoting a technology called ‘videotex’ (videotext). Just as there are many conferences on the future of mass customization being held today, in the videotext field many conferences were held discussing various strategies which would promote use of the new technology.

Among the concepts discussed were issues such as standardization of the graphics which would be carried on what were envisioned as linked private networks. I recall having learned discussions about the virtues of something called NAPLS (North America Presentation Level Standard) graphics.

Then there was the matter of the underlying transport layer. Commonly, discussions focused on X.25 networks being linked between cities, and the major impact of Moore’s law was yet to be felt as tariffs for moving data were still relatively high.

The evolution of several new standards facilitated the internet’s eventual explosion as a major consumer communications channel. Almost overnight the small but growing early 1980’s private networks such as Prodigy and CompuServe were connected over a consistent and expandable network.

Although certainly key, it was not just the invention of hypertext markup language (HTML) that made the ‘net work. Other discrete developments contributed such as the definition of the communications layer with the OSI model, the relative ease of teaching a new computer user how to interact with the new icon-based user interfaces, and standardization of graphical content thanks to CompuServe’s work on the GIF format and the Joint Picture Expert Group which built the .jpg standard.

As I like to do, let’s zoom out to the high level vision of the internet and list the key paradigm shifters that collided to make the net ‘pop’.

• A document linking language • A smooth and somewhat consistent windowed user interface (UI)

• The OSI transport model which defined how TCP-IP would work in a layered fashion

• Consistent graphical content

• Last, but not least, a killer ap that pulled all of these individual improvements into a single software environment: The Browser.

In much the same way, we read how various conferences are being held that almost sound like repeats of the pre-internet days. If videotext were a contemporary concept, it might well be called “mass customization.”

Fortunately, the high level strategic lessons of the Internet’s “youth” can be used as a learning template to resolve the issues that prevent widespread adoption of mass customized product. Right now, there isn’t much.

The whole notion that somehow ‘mass customization’ could be a workable concept has is rooted deeply in the mass communication mindset. Certainly, as modern advertising channels developed as a one-to-many medium (e.g. from manufacturers to consumers) there was little room for feedback. A few exceptions were luxury cars and boats, where owners had some degree of interaction with designers and builders, but in the main customization was something ‘not done often’ because of cost.

We can use the automobile as a prototype to see how customization evolves over time; from virtually none to a thriving business today. In the earliest days of the automobile, Henry Ford wanted to make cars any color so long as it was black. So the first level of customization in cars was color.

Over time, additional product choices were introduced including power brakes, air conditioning, different levels of radios and now multimedia and navigation systems. So complete has the adoption of previous ‘features’ been that at least air conditioning is becoming a ‘standard’ items on most cars. Similarly, anti-lock brakes are in transition now from ‘optional’ to standard.

Customization in autos, when you think about it, has also differentiated. On the one hand there are full-up custom shops which bring to mind names like Chuck Barris and Orange County Choppers, to the factory option sellers such as Ford’s Special Vehicle Operations  (SVO) now the Ford Performance Group/SVT, and the TRD unit at Toyota (Toyota Racing Development). Then you have a brand-specialty group. Jim Dines, for example, is legendary among big Beamer drivers, and Carol Shelby has his following among Mustang drivers.

While there is a certain level of ‘generalized customization’ available from the auto supply shops like J.C Whitney, perhaps the most widely truly customized car was the Volkswagen Beetle. Tens of thousands of these we from being generic low price, high utility value cars to being transformed into truly unique sand rails, dune buggies (I’ll let you sort out the differences and there are many) to the Avenger body kits, and perhaps the most widely touted in the late 1960’s – the Bradley GT.

What’s particularly important to note in our discussion here is that through present day, the amount of ‘customization’ of most products, such as autos, has been largely changes of appearance. Metal flake paint. Special wheels, a suspension bolt-on, and perhaps a few more CD’s in the changer. The underlying car is mostly still the same: four wheels, a couple of seats, a steering wheel, and a roof over it with heating and cooling to keep up with Mother Nature.

There’s also one other important aspect: Obsolescence. Virtually every auto made today is designed with a typical maximum service life on the order of 200,000 miles. Trucks, especially the big names like Peterbuilt and Kenworth go millions of miles, so it’s not like it can’t be done. Just Detroit likes the planned obsolescent business model because it results in higher sales.

As I’ve written in other papers, the whole notion of the ‘annual model’ didn’t really exist until just going into the Depression. Alfred Sloan was its champion at General Motors in the mid 1920's.

In the analysis of modern mass production, we can generalize that business models over time tend to evolve from many-to-many models, to few-to-many, to one-to-many as specialization is improved.

As an example, in the early days of medicine, there were lots of small drug and patent medicine makers selling a broad spectrum of products (many of questionable value) to the mass market.

Over time, and driven by consumer protection legislation plus improvements in medicine, a ‘few-to-many’ modelspace appeared. This was predominately by large, well-capitalized companies, each pursuing improved medicines across a still relatively broad spectrum.

However, as individual research programs paid off, new mass markets were discovered and these, at least till generic meds came into the space, were hugely successful one-to-many models. Take the case of Pfizer’s building of the erectile dysfunction market with Viagra, as a prime example.

To a lesser extent, this has occurred in almost all market segments such that there are only a handful of world class auto makers, a small number of LCD display makers, and so forth.

Once a technology has been pushed to near its practical limits, and as segment competition limits pricing power, the only ways to increase the bottom line are through expert segmentation by feature set enhancement, or by cost-reducing manufacturing.

Thanks to the patent and intellectual property frenzy, the easiest tack for most companies has been to cost-reduce manufacturing and assembly. That’s why more and more auto subassemblies, television sets, and what have you are being assembled in Mexico, and why you may not be able to find a “Made in America” television, refrigerator, ice maker, or dress shirt.

The public policy issues are immense, but easily solved with the application of money to the rule-maker class. While it certainly makes more sense to make goods close to their eventual point of consumption, it doesn’t do anything for the cost-driven bottom line. This is why manufacturing formerly done in the U.S. is now mostly done in Asia and India: These countries have been at the forefront of workplace automation (Japan’s auto assembly lines for example, or the appliance lines in Taiwan and South Korea, of the huge “human automation” done in India and China.

How this whole system of constantly revising the means of production in order to reduce costs ends has been hotly debated in business circles.

One crowd argues that we will trip over a ‘business singularity’ at which point there will suddenly be infinite numbers of ways to make products more efficiently. While that sounds like a lot of ‘wishing and hope’ (not to mention a perversion of Moore’s Law to non-computing areas) it’s apparent to even the ‘singularity’ adherents that we need to get there pretty damn fast because Peak Oil is here and now that we’re cutting down the rain forest in the Amazon, the clock is ticking on resource depletion.

It occurs to me that there are only a few possible outcomes for our collective future:

1. Business hits some magical singularity and making money reverts to a new kind of values basis. I rate this outcome low probability, but worth hoping for.

2. Business (or the ‘powers-that-be’) manufacture a sufficiently large war/series of wars to provide enough infrastructure and people destruction that growth in output becomes possible from the infrastructure that remains. This is a ‘maybe’ but not ‘likely’ given the linguistics from www.halfpasthuman.com  for late summer. The Middle East flare-up due then seems driven by corporate profiteering/profit engineering needs masking the true agenda of resource and energy acquisition.

3. The world simply starts to run out of (pick anything you want here…) oil, wheat, grains, coffee, tires, rice, home heating oil, natural gas, fresh potable water, public support of the fiat currency debacle…well, you get the idea.

“What does all this after to do with mass customization?” you’re asking.

I’d propose that concurrent with the migration of manufacturing, over the course of the Industrial Age that an archaic version of “craftsmanship” has disappeared. True, in terms of absolute mean time-between-failures, commoditization has reduced failures at the macro level. But the art of repairing goods has been largely supplanted with the ‘black box replacement’ strategy.

In 1920 if you had a problem with the carburetor on your automobile (they weren’t common enough to be called ‘cars’ or ‘rides’ just yet) a skilled mechanic would look at the problem. Maybe it was dirt in the bowl, perhaps a fouled jet, or even something like a leaky gasket. Whatever the problem, it could often be fixed right on the spot. By the time the 1950’s got here, the car repair field has started moving toward a more ‘gross’ repair strategy. A transmission might be swapped out for a rebuild and the engine of a car might be swapped out for a ‘short block’.

By the 2000’s we had moved into an age where all kind of additional subassemblies are now being replaced at the slightest provocation. Is it worth trying to troubleshoot a single capacitor or resistor failure on your car’s cabin climate control module? Ha! Nope, not just the erring box gets replaced but often several subassemblies just to ensure that the problem wasn’t caused elsewhere.

Most mechanics don’t have the time (at typical $65-75/hour of shop time) to tear apart much of anything. They simply order in parts, replace them, double their cost, tack on their time, and pass it all off to the customer.

Along the way a very few manufacturers were able to buck the trend, but even here it was only a momentary holding measure. Land Rover 109’s and Volkswagen Beetles come to mind along with Jeeps surplused from the military and Post Office, and my all-time gotta get one someday rig, a Toyota Land Cruiser. For the most part, the auto industry slid happily into the 100,000 mile lifespan, black box it when it breaks kind of mindset.

The examples seem to frame an undercurrent that has been very dangerous to the planet:

• Consumer items early in the Industrial Age had high value and manufacturing was in the many-to-many mode. Prices were high, goods restrained; products were maintained because their replacement would be prohibitively costly.

• Somewhere around the midpoint of the Industrial Age we see that consumer goods became more widely available, prices came down, and the standard of living generally went up. However, falling prices also removed incentive to repair and maintain really cheapo goods.

• As the Industrial Age faces its largest challenges ever (with the potential failure of the fiat/pseudo-money cultural paradigm) the zenith of the Industrial Age with nearly instant obsolescence and a culture within which everything (including human relations) is disposable faces collapse just as it was turning into a single corporate state provider of all goods to all people.

The linguistic work of www.halfpasthuman.com  suggests that this won’t become widely apparent until perhaps early October of this year. However, from that point forward into 2009 a society-wide paradigm shift occurs with 2009 being the “Year of Transition.”

“Transition to what?” you may wonder. Presumably it will mark a major transition to a resource constrained lifestyle because we see leading indicators of this in the news headlines almost every day lately:

• Limitations on the amount of rice available for purchase

• Talk about the impact globally of soaring food prices.

• The ‘push-back’ on the cattle industry from an increased emphasis on the use of corn for ethanol production.

• Almost continuous increases in the price of oil, due in part (depending on who you listen to) to either Peak Oil or simply exploitation by oil interests.  Notice how many gas stations are limiting credit card purchases to $50 these days?

With the whole world facing a combination of food shortages, oil and energy issues, not to mention a quickly spreading recession that could go global as US consumers tighten the belt, there’s a natural question that arises in any discussion of “mass customization” – how does it fit?

Another/ further consideration to be weighed is the role of ‘distance to market’ and how it has evolved over the last 40-years.

Recall that globalization has been a late-stage event within the current economic macro-cycle. The Volkswagen Beetle first appeared in America about four years after the end of the last Kondratieff (Kondratiev) “Trough War”, e.g. World War II.

However, it didn’t really start picking up until the late 1950’s when Japanese companies, utilizing the imported expertise of American consultant like W. Edwards Deming, brought a new kind of highly efficient mass production that was able to penetrate the American marketplace because while America made good products, cheaper foreign-made products could be priced lower because the combination of low overseas labor and production costs plus transportation costs to land the goods in the U.S.A. was less than the production and labor costs domestically.

Not surprisingly, off-shoring production has become a runaway trend in manufacturing, although a number of authors, such as Barry Lynn in “End of the Line: The rise and fall of the Global Corporation” argue that as labor rates rise offshore, and the dollar declines as it becomes weighed down with consumer debt, the trend will reverse.

And that gets us to an important economic fork in the road when considering the concept of “mass customization”.

I’d offer that there is a curious dichotomy facing mass customization. It boils down to the notion that mass customization will not make widespread gains until one of three possible tracks becomes widely recognized:

• The High Track envisions a dramatic increase in disposable income which will facilitate increased customization because with higher disposable incomes, people will likely be able to spend money on uniquely customized goods.

• The Middle Track envisions continued deterioration of the corporate off-shore business model and a reinvention of business models based on the “micropreneur” concept which we’ll delve into in greater depth in a moment.

• The Low Track outlines the case where mass customization becomes (or shades into) custom or short-run manufacturing due to resource or transportation constraints from possible future events like an Oil Embargo or a general economic collapse.

So no matter what happens to the economy some variety of mass customization or short run manufacturing is bound to flourish.

The way traditional mass manufacturing companies are organized requires a substantial amount of capitalization in order to produce a ‘profit’. In other words, a company must have a manufacturing plant, suppliers of raw materials, workers, machinery, packaging, shipping, warehousing and distribution in order to make money.

The large multinational corporations now occupy unassailable positions because the huge capital expenditure required to facilitate a new player entering a particular vertical market becomes a formidable barrier to entry.

What’s more, market share is also a barrier to entry.

If you wanted to begin Monday morning building lawnmowers, without massive advertising expenditures and distribution deals (paid for one way or the other) you would not likely have sufficient sales to achieve something I call the “magic of utilization.”

It’s a concept borrowed from the airline industry where I first learned and lived it: If an airplane lease costs $20,000 per day and the airplane moves 200 passengers, then each passenger must pay $100 toward the airplane lease overhead in order to make ends meet. This is in addition to fuel, flight crews, advertising, administrative costs, maintenance and what have you.

But now image a situation where an airplane carries 2,000 passengers per day. In this case, each passenger’s ticket need only contribute $10 toward the lease and the balance of the ticket price can go to other operating expenses.

The same holds true for manufacturing plants. A manufacturing plant that can operate at break-even on one shift will bleed a company to death if there is only one hour a production a day. Or, as shareholders dream it, the company’s products will be so demanded by consumers that the plant will run three shifts every day of the year at peak efficiency.

In the traditional manufacturing model specialization was determined by corporate ownership.

In 2004 that changed when Jim Lewis, president of Micro Logic Corp, envisioned and launched a general manufacturing company called eMachineShop.com. The business model was built on the confluence of several new technologies and really started to move things in the direction of software defined companies:

• A widespread migration of product design into the Computer Aided Design (CAD) world and from there,

• Manufacturing of products directly via Computer Assisted Manufacturing (CAM)

• The ability to distribute free or very low cost software via the Internet, and

• The realization that because machine set-up for manufacturing could be largely reduced to a software exercise, a generalized manufacturing company could be built which could hit appropriate utilization levels to become profitable while at the same time being (if the term makes sense) “product agnostic.”

In other words, as long as eMachineShop could build utilization by building a customer base, then generalized manufacturing could be realized. • Having an extensive background in workflow automation, Lewis also built an enterprise software platform to manage as much of the customer interaction as possible in order to focus on building the generic manufacturing platform.

In addition to starting the www.emachineshop.com  platform, Lewis also started a printed circuit board house, www.pad2pad.com  which featured free PCB layout software.

Now that can have a sense of how the generalized manufacturing company can operate, the next area to discuss is…

A Micropreneur is a scaled down Entrepreneur. A Micropreneur can be one person (or just a few) leveraging all available technologies in order to build a competitive business model without the traditional high capital costs associated with new product design and launch.

Thanks to the existence of the internet and now the generalized manufacturing company for metal, plastic, and fiber products, not to mention printed circuit boards, plus the new class of 3-D printers and advancements in stereo lithography, it’s possible for a person with a good idea and some basic web skills to create an entire company.

To show you how micropreneuring works, let’s look at how a person with a good idea for a car part could become a micropreneur:

• First, the idea is designed - perhaps using a free CAD software packaged such as the www.emachineshop.com  or if it’s a printed circuit board based product, using the free layout tool from www.pad2pad.com  and the free schematic design tool from www.capilano.com  (DesignWorks Lite).

• Next, the product is produced and the finished parts are shipped to the micropreneur.

• If assembly is required, the Micropreneur can do that – or part time local labor can be employed.

• Now the product is sold. The ease of web site creation and the availability of formerly nonexistent sales channels like eBay and even Craigslist have opened up wide new vistas for creative marketers. If the product is well accepted in its target market, then it can be ‘ramped up’. The Micropreneur can either scale up production runs (usually at lower unit costs), or the now-viable concept can be sold to a company in the desired vertical market.

A key point about Micropreneuring is that it is adaptable to very small market niches. If a person is restoring a 1939 Chevrolet, for example, and needs to design a replacement for the vacuum-advanced distributor, they can easily turn into a Micropreneur by simply making more parts than they need for their own antique car restoration project and putting a few on the shelf in the garage. Over time, these parts can be sold (perhaps at some profit) to the originator of the design/build project.

The One Last Piece Like a good detective story, we’ve found most of the pieces of this particular business puzzle. There’s just one part left – and that’s a part that doesn’t exist…or at least didn’t until this week's report.

That’s a repository where Micropreneurs, Product Designers, and end users (Consumers) could all meet and exchange designs and schedule manufacturing.

The result of this week's report is a draft of something I call the Public Design Library – a draft of which is now up at www.publicdesignlibrary.org. OK, so I went a little overboard on this week's report...

.

It provides a starting point – much like planting a seed – to give mass customization and short run production projects some fertile soil from which to grow.

Here’s the invitation: If you have ever designed anything in a compatible format (.dxf – which stands for digital exchange file) the vision is that over time the Public Design Library should be able to grow just as text-oriented libraries grow.

With luck, over time, this will provide a place where if you want something like a piece of custom furniture, you can view finished pieces, download a proprietary or public plan from a design, and have it built.

Or, if you’re a Micropreneur, launch that “Next Big Thing”.

Or, if you want to build a cabin, download a set of plans.

And one important point:

If the economy does go through some massive contraction, the development of the generalized manufacturing company can provide a backbone of geographically dispersed manufacturing sites which might serve consumer interests better than the megalithic corporate giants of today.

In short: Better products, more suited to individual needs, locally produced and with the savings accruing to the end users, not a corporate bottom line somewhere offshore. That's because if the communications infrastructure of the world holds up, there's no reason why a concept like eMachineShop's couldn't be used to route orders to virtually thousands of local software defined manufacturing companies worldwide, which in turn might move the world away from the mass consumption model and toward a quality-oriented best use of resource model where quality and craftsmanship would again return to their historical prominence.

Just as the Open Source movement in software has begin to challenge the traditional software development model, so too I expect the "public design/ customer-local-dispersed manufacturing model in software defined manufacturing centers to be an important coping tool as traditional capitalism becomes ever more cannibalistic as resource restraints grow.  It's maybe one way "Together, we can make it..."