From: Steven Pollack (steve@familyjeweler.com)
Date: Sat Jan 05 2002 - 03:11:53 EET
Brock,
I know you are right that custom jewelry makes up a much smaller market
segment than mass production and whether this is because there are so
few skilled craftsmen to satisfy the demands of the general public or if
the lower prices of mass production are the reason more craftsmen do not
enter the market to offer custom, the size of the markets are in fact
different. The custom market may not be as small as you think though.
Chain stores and other sellers of discount jewelry account for 50% of
the market while small mom and pop stores acount for the other half.
There are about 35,000 different jewelry businesses in the U.S. so
something accounts for this high level of fragmentation and I would
submit that alot of it is the ability of small craftsmen to open their
own store and offer custom services.
But in the case of jewelry, why is custom more expensive than mass
production and does CAD/CAM enabled mass customization alter this
equation? For the following I ignore all CAD/CAM except Digital Jeweler
because this is the one instance where the cost of the CAD model
creation approaches zero at higher volumes.
Typically the cost of creating the master model, hand carving a unique
wax, is what made up the bulk of the price difference between mass
production and custom.
Digital Jeweler creates the CAD model in a matter of a minute and can
handle a high volume of model creation given ongoing development costs.
I figure that the Solidscape printer produces ring sized objects on the
order of $50 per unit based on full capacity and a one year
amortization. This is one cost that mass production does not have. If
the people from 3D Systems are correct and the Viper can produce 50 ring
sized units in 4 hours then the unit cost drops to about $12. Of course
I am still having trouble casting the resin material so it is not yet an
option but I think that there will be a breakthrough in the ability to
cast it's output.
So unit costs, at full capacity, can already drop into a range where the
differential between the mass production injection wax and the RP
created model are approaching relative parity. The next issue is the
cost of finishing. Mass production utilizes cheap labor but there is
nothing to say that the digital files produced with Digital Jeweler
cannot be forwarded to an overseas production facility where the RP
would be located. The finishing time on an RP model is longer than a
nice shiny injection wax but with labor at a dollar or two per hour that
is hardly a difference.
While we are not at a level where RP enabled mass customization can be
priced as cheap as mass production, we are getting really close.
Steven
Brock Hinzmann wrote:
> For those small numbers of people willing to pay the extra for
> customized styling, I suspect you could be right. Actually, you
> wouldn't even need to standardize the bolt locations, as long as the
> part interface can be scanned accurately into the system. The big
> companies might even welcome design experimentation and might want to
> be informed of improvements and failures (such as materials mismatches
> that result in parts falling off, crashes, and so forth) experienced
> by such lead users. I don't deny the logic and technical feasibility
> of what you say. I just think market experience indicates most people
> aren't willing to pay extra. Even in jewerly and art, where you would
> think uniqueness would be primary and where safety is not an issue (in
> most cases), the volume of custom work pales in comparison to the mass
> of manufactured stuff.
>
> Brock
>
>
> Steven Pollack wrote:
> >Brock,
> >While RP may be a ways away from making completely original designs
> for
> >standard models of cars, one thing it can do is make manufacturing
> smaller batches
> >economically feasible. Whereas auto manufacturing tool and die
> creation
> >and setup charges may only amortize at volume over 100,000 RP may
> make it feasible to
> >setup smaller production in the 1,000's of units. The CAD system in
> that
> >case, instead of offering unlimited variation in the shape of the
> hood might allow
> >for a dozen variations which is 11 more than we have to choose from
> today per
> >model. The bolt locations would be fixed per model or model type
> while the styling
> >would change. The biggest issues are where the parts interfacce with
> other
> >parts so they would need to come to a common plane or arc.
> >If I am correct in the feasibility of smaller batches then RP may
> enable
> >smaller boutique type auto manufacturers to flourish.
> >Steven
> >Brock Hinzmann wrote:
> >Terry's note reminds me of a couple virtual car companies that some
> >former big3 execs tried to form a couple years back. They may still
> be in
> >business. The idea was to allow people to log in and design their
> custom car
> >on-line. The virtual car company would then contract out all of the
> >components to suppliers or even to the OEMs. I guess it's not much
> >different than what the OEMs do now. I just didn't see how you could
> rely on
> >the current supplier system to make sure a Mercedes hood would fit on
> a
> >PT Cruiser (to use a really bad example). To take it further, how
> flexible are
> >suppliers in responding to someone requesting, say, a shower and a
> bath
> >tub in the back of their van? Still further out is the notion of
> using RP to
> >make completely original designs and parts that will fit the standard
> model?
> >Brock HInzmann
> >
For more information about the rp-ml, see http://rapid.lpt.fi/rp-ml/
This archive was generated by hypermail 2.1.4 : Tue Jan 21 2003 - 20:13:18 EET