Point to ponder.... Are we looking for a 100% solution to the selection
process? I'm not. I want to be able to scan the field, find "x" number of
processes that would work and go out and do my own due diligence on those
processes.
I can't imagine a situation where we would be able to hone in on a "best
practice" process. Heck, if we did that process would probably be making
lot's of $$$ for someone without much work. The answer would be so obvious.
Why can't we come up with some "industry standard" way of defining parts
.... say a Group Technology approach ... or Design for
Manufacturability/Assembly approach (Boothroyd and Dewhurst) and use that as
initial input to the selection process. This would generate a range of
processes that would acheive some percentage of the users goals.
Additionally, it would be nice to factor in things like..... How much space
does this process require? Utilities? Safety Requirements? Machine cost ?
Maintenance cost? Training? Learning Curve? Material cost per model? Startup
Issues? Other changes required or desired to existing processes? Integration
of existing processes?
This is a decision making and managerial tool. I hope no one expects the
process to be perfect... heck the RP systems are far from it!
Paul Finelt
Director, Product Design & Product Engineering
Swarovski North America
1 Kenney Drive
Cranston, RI 02920-4400
voice 401.463.2757
fax 401.463.4757
> -----Original Message-----
> From: Roger Spielman [SMTP:RSpielman@regale.com]
> Sent: Monday, December 17, 2001 12:11 PM
> To: 'Marshall Burns'; 'RP Mailing List'
> Subject: RE: STONE SOUP
>
> Hello all:
>
> Good point Marshall. This really is inevitable, and necessary to the
> evolution of a "neural network" for Rapid Manufacturing. You touched on
> the
> cornerstone when you mentioned material properties. Unfortunately, very
> little work has been done (that is available) on this subject for many
> processes other than SLS and SLA. By that I mean information that can be
> used in a materials database or selector. Another aspect will be the
> ability
> to determine which process (including turning and high speed machining)
> would be most appropriate for a given geometry. With all the great work
> that
> has been accomplished in the past ten years or so we really are still on
> the
> ground floor, and in my opinion, looking at a pretty bright future.
>
> Regards
>
> Roger Spielman
> Regale, Inc
>
>
> -----Original Message-----
> From: Marshall Burns [mailto:Marshall@Ennex.com]
> Sent: Friday, December 14, 2001 9:48 AM
> To: 'RP Mailing List'
> Subject: Re: STONE SOUP
>
>
>
>
> From: "Bibb, Richard (PDR)" <RBibb-pdr@uwic.ac.uk>
> > The problem with trying to come up with a system that evaluates RP
> systems
> > is that it is in danger of comparing apples and bananas and not apples
> and
> > apples. I suggest that a practical way forward is to compare RP system
> > capabilities to product development objectives rather than comparing all
> RP
> > systems to each other.
>
>
> One of the exciting things about this whole field of technology is
> that
> there are so many different ways to accomplish its objective (to make a
> solid object from a digital description and raw materials). I believe the
> issue proposed by Paul Finelt in his "Stone Soup" e-mail has an even
> greater
> long-term importance than deciding which *machine* to buy or use today.
> Future machines will incorporate a variety of processes and the software
> will have the challenge of determining which *process* to apply to
> creating
> any particular shape and structure found in the CAD data.
>
> Perhaps a step forward for working on this problem from where we are
> today is to develop a database of technology capabilities in terms of a
> multi-dimensional matrix of materials, material properties, resolution
> (in-layer and z), accuracy, solid formation rate, results repeatability vs
> need for experienced operator, etc., etc. There's a wonderful book by M.
> F.
> Ashby called "Materials Selection in Mechanical Design" that gives dozens
> of
> two dimensional selection grids on the basis of many pairs of
> characteristics. A computerized, multi-dimensional version of this
> technique
> could help a user zero in on a desired domain of capabilities and the
> processes/machines capable of working in it.
>
> Best regards,
> Marshall Burns
> President, Ennex Corporation
>
> Marshall@Ennex.com
> Los Angeles, USA, (310) 397-1314
> http://www.Ennex.com
>
>
> For more information about the rp-ml, see http://rapid.lpt.fi/rp-ml/
>
> For more information about the rp-ml, see http://rapid.lpt.fi/rp-ml/
For more information about the rp-ml, see http://rapid.lpt.fi/rp-ml/
This archive was generated by hypermail 2.1.2 : Fri Jan 04 2002 - 09:58:19 EET