From: Ian Gibson (igibson@hkucc.hku.hk)
Date: Sat Nov 01 2003 - 04:21:20 EET
Phil et. al.
I have looked at the idea of a decision support system (DSS) for RP
selection off and on for the last 7 or 8 years. During that time I carried
out 4 or 5 projects to investigate the issues.
Here are some of my thoughts/findings on this: -
- we felt that there were 2 kinds of DSS systems likely to be of benefit.
- The first DSS would be used to assist the novice in understanding a
little more about the technology. This could help in providing an initial
selection list for those wishing to invest in RP technology or for those
wanting independent information to assist with quotes from service
providers, etc.. This would work best in the form of a web site where
people logged in, answered a few questions and then collected the answers.
They could then try adjusting some of the parameters to see the effect from
changes. There are a lot of machines out there now and it is very difficult
to keep track of them. This is why a web-based system would be most
suitable with the flexibility to be updated as new technology becomes
available.
- The second kind of DSS would be a more detailed and probably specific to
a technology or a company setup. This kind of system would be used to
assist a bureau in making quotes, planning, etc. I saw someone use a
simplified version of this at a bureau I visited, but this system took a
lot of skill and knowledge in order to create meaningful data. The idea
here is to try and distill the expertise of the chief quoter/planner by
using Expert System shells. This would have to be tailor made for an
operation, probably made up from basic building blocks.
- Both of these would be decision support systems because, as is clear from
the recent discussion, benchmarks are very subjective and can only be used
in conjunction with other information to build a complete picture. I would
say Todd's study, along with others, is only part of a large jigsaw puzzle
that can only be put together by someone who has seen the picture on the
box. I would also say that John is perfectly at liberty not to use it, but
it may be helpful to reference such studies if you are an expert trying to
put a particular point across to a client (in the country of the blind, the
one-eyed man is king). Such studies always include caveats and I'm sure
Todd has included a number in his report (which I admit I have not seen).
No one ever makes a study like this without addressing the limitations.
- My first idea for a web based DSS seemed to me to be a potential money
making enterprise. I envisaged users logging on and paying, say, $5 for
each set of data they produced. I even developed a couple of experimental
systems that showed promise. However, when I tried to get funding to
finance the development of a proper system, I found that the funding
agencies thought that RP was sufficiently well funded already and the
industry should be able to finance itself. When I contacted the companies,
they said there was not enough money in it to make it worthwhile. However,
I did get agreement from all the major vendors to provide the basic system
information necessary for the development of the DSS. This was about 3
years ago when I looked into this seriously.
Sorry, got carried away a bit here. I suppose I'm saying that maybe it is
time to review the situation. Would anyone be interested in collaborating
with me to set up a (semi)commercial web-based DSS for RP technology selection?
IG
At 12:22 AM 11/1/2003, you wrote:
>Todd,
>
>Sorry for the delay in my response.
>
>Your offer is most kind and much of your rebuttal I agree with.
>
>Some of my points are the following, although they may ramble around a
>bit. ( it's been a long week ).
>
>The effort of disseminating information on RP, how to present the
>information, and the benefits of each technology, has been an ongoing
>quagmire when viewed as an overall process. It is obvious that as new
>technologies appear, users outside of the service providers need to be
>made aware of what their latest options and potential / real benefits are.
>Some form of helping guide decisions has been needed, and that need is
>growing.
>
>A simplistic view about the way most users approach the RP industry as a
>whole, is that much of everyday allocation of work to different
>technologies by outside designers is a seat of the pants process, or
>directly cost driven with many blind spots. As with any industry, there
>are many who have done their homework and use it wisely.
>
>My point about any statistical approach is that they are often lacking the
>needed numerous footnotes about the exceptions, exclusions, inclusions,
>explanations, deviations, etc., about each data source and what a huge
>impact, a small deviation in the test model can make, sometimes rendering
>an examined system useless for the task.
>
>This confirms your position that the amount of data to present can quickly
>become very unwieldy and confusing. To boil it down to a common
>denominator that covers the systems and reduces that data load also
>creates a tunnel vision view of the various available processes. Again the
>only way around this is lengthy explanations such as: "When evaluating
>systems, use this benchmark data as an initial selection guide. Then
>define the application and the types of parts used in the product
>development process. Evaluate the systems with the operational and output
>requirements that are important to the success of the prototyping effort.
>Finally, add the evaluation of two important criteria that were not
>reviewed in the benchmark, material properties and finishing time."
>
>The need to be objective and clear in a relatively small data set
>precludes the ability to examine the endless options available.
>Who would want to read a 20 pound book just to get at a few small details,
>albeit the importance of them?
>
>I have to applaud any individuals attempt to tackle the task of trying to
>place a common thread through so many diverse systems.
>
>A note about my experiences from both sides of the user/supplier points of
>view.
>I often find that the people who are likely to use RP have the long term
>view for their business. They will invest in using the technology to
>realize gains further down the line.
>Others (an increasing majority) who are more concerned about "what is our
>bottom line" at 5:00 on Friday, will not spend the money, time, or
>resources to save themselves money and effort down the line. Thus they
>lose money AND market.
>With business's increasing fascination with "the bottom line" and the
>increasing management that lacks planning for the long haul, this makes
>the task of getting RP to the front in manufacturing, R&D, and new venues
>a seemingly daunting task.
>
>I am not all gloom.. Many inspired people are finding unique and novel
>applications for RP that are blossoming. I applaud them all.
>
>
>Regards,
>
>Phil Iehle
>Versadyne LLC.
>
>
>-----Original Message-----
>From: Todd Grimm [mailto:tgrimm1@insightbb.com]
>Sent: Thursday, October 30, 2003 8:02 AM
>To: Phil@Versadyne.net
>Cc: rp-ml@rapid.lpt.fi
>Subject: RE: New Rapid Prototyping Benchmark Study Examines 3D Printers
>
>Phil:
>
>You make some very good points. I would like to address them.
>
>"I have to agree with John on some points. A Viper is a high end RP system
>and not a "3D printer". Results on the 0.006" build style without mention
>of the flexibility of building at 0.002" layers and 0.003" spot are
>misleading. I also believe it unfair that as a high end flexible system,
>that his cost per hour takes account of the acquisition price which is of
>course much higher. I quote - "3D printers are affordable and easy to use
>rapid prototyping systems for personal, office or departmental use." Why
>not compare the InVision printer or at least draw the line at a dollar amount.
>I feel this is a case of measuring apples and oranges. "
>
>Yes, I agree that the 0.006 in. could be misleading. But, if the report
>were to document all of the build options for all of the systems, it would
>have become unwieldy. I did consider including the Viper with 0.002 in./
>0.003 in. build parameters, but that didn't fit the criteria of
>concept/form & fit models. I also felt that the reported time for these
>build parameters would have given the wrong impression to the reader. The
>other consideration is that many Viper users would not consider
>0.002/0.003 for dense/large parts like the fan and trackball housing .
>
>If the InVision would have been commercially available when the testing
>began, I could have consider it. However, the system was not commercially
>available until 2 1/2 months after the building and measurement phases
>were complete.
>
>"And what about build material choice?
>
>The fact that the study is aimed at steering people to various methods by
>comparison can be misleading without understanding the very real impact
>that geometry, accuracy, and end use / purpose has on making the right
>choice................"
>
>Both of these statements are absolutely true. I hope that the conclusion
>to the Executive Summary/Benchmark report addresses that. It states,
>"There are a vast number of combinations of build parameters, prototyping
>materials, part definitions and operating conditions. Testing of all
>scenarios in the benchmark study is impractical and unreasonable.
>Therefore, the results presented in the benchmark are best suited for the
>relative positioning of the rapid prototyping systems when similar parts
>are constructed with similar build parameters.
>
>When evaluating systems, use this benchmark data as an initial selection
>guide. Then define the application and the types of parts used in the
>product development process. Evaluate the systems with the operational and
>output requirements that are important to the success of the prototyping
>effort. Finally, add the evaluation of two important criteria that were
>not reviewed in the benchmark, material properties and finishing time."
>
>"I agree that does need addressing. Lets do it right."
>
>I agree. If you would like to discuss revisions to the benchmarking
>procedures, please give me a call. My goal is to help potential users make
>the right choice, and any assistance that you could offer would be appreciated.
>
>Regards.
>
>T.A. Grimm & Associates, Inc.
>3028 Beth Ct.
>Edgewood, KY 41017
>Phone: (859) 331-5340
>Fax: (859) 331-5342
>Cell: (859) 240-0574
>Email: tgrimm@tagrimm.com
>Web: www.tagrimm.com
>
>Marketing Engineered for Sales Results (sm)
>-----Original Message-----
>From: owner-rp-ml@rapid.lpt.fi [mailto:owner-rp-ml@rapid.lpt.fi] On Behalf
>Of Phil Iehle
>Sent: Wednesday, October 29, 2003 3:21 PM
>To: Rp-Ml
>Subject: RE: New Rapid Prototyping Benchmark Study Examines 3D Printers
>
>Hi List
>
>I have to agree with John on some points. A Viper is a high end RP system
>and not a "3D printer". Results on the 0.006" build style without mention
>of the flexibility of building at 0.002" layers and 0.003" spot are
>misleading. I also believe it unfair that as a high end flexible system,
>that his cost per hour takes account of the acquisition price which is of
>course much higher. I quote - "3D printers are affordable and easy to use
>rapid prototyping systems for personal, office or departmental use." Why
>not compare the InVision printer or at least draw the line at a dollar amount.
>I feel this is a case of measuring apples and oranges.
>
>I have to reiterate, Fine Line Prototyping is a prime example. You can't,
>and wouldn't think of, getting the same type of part and quality from "low
>cost" 3D printing systems. They are miles apart.
>And what about build material choice?
>
>The fact that the study is aimed at steering people to various methods by
>comparison can be misleading without understanding the very real impact
>that geometry, accuracy, and end use / purpose has on making the right
>choice................
>
> "I produced this benchmark study specifically for design engineers who
> are trying to select the rapid prototyping system right for them,"
>
>We are a bit of a 'not-that-well-known' and cryptic industry. I agree that
>does need addressing. Lets do it right.
>
>Phil Iehle
>Versadyne LLC
>
>-----Original Message-----
>From: owner-rp-ml@rapid.lpt.fi [mailto:owner-rp-ml@rapid.lpt.fi]On Behalf
>Of Neil Hopkinson
>Sent: Wednesday, October 29, 2003 11:47 AM
>To: rp-ml@rapid.lpt.fi
>Subject: RE: New Rapid Prototyping Benchmark Study Examines 3D Printers
>
>List,
>A few years back I performed a similar sounding study for Volvo Cars as
>part of our Rapid Manufacturing Consortium.
>A couple of key points to make with this kind of benchmark include:
>- the findings are often geometry/purpose specific
>- the findings are often out of date by the time they go to print.
>
>I have not read the report but would expect that Todd has covered the
>first point, previous work by him appeared to be well balanced and fair.
>The second point is very important to remember however it does not mean
>there is no value in the excercise, it's just that the reader should be
>aware of its limitations (ideally the author will make this clear).
>The second point is in fact a credit to system/material developers -
>paradoxically it is good news if the report is out of date quickly as it
>(usually) means significant improvements have been made.
>
>I look forward to watching a lively debate!
>
>Neil
>
>
>
>
>-----Original Message-----
>From: owner-rp-ml@rapid.lpt.fi [mailto:owner-rp-ml@rapid.lpt.fi]On Behalf
>Of John Kerr
>Sent: 29 October 2003 14:20
>To: tgrimm@tagrimm.com; rp-ml@rapid.lpt.fi
>Subject: Re: New Rapid Prototyping Benchmark Study Examines 3D Printers
>
>Dear List,
>
>Went thru the report !
>
>Some questions :
>
>1. Is this site marketing site for selling reports ?
>
>2. This report is definitly not made by any RP Veteran but some novice
>individual.
>
>3. Everybody knows that Viper is serious high end RP system not 3D
>Printer. Why include it in the comparisons ? Viper accuracy same as
>Dimension printer ? Ha Ha , who are they fooling ? Ask Fine Line
>Prototyping or experts from 3D Systems, please wake up, or these fools
>will confuse the market ! I doubt the credibility of of the publishers
>
>Why Invison, Eden is missing my friend ?
>
>Johnn
>
>Todd Grimm <tgrimm@tagrimm.com> wrote:
>FOR IMMEDIATE RELEASE
>
>Edgewood, Ky., October 27 -- T. A. Grimm & Associates, Inc. today
>announced the release of a new rapid prototyping benchmark study that
>examines 3D printers. This extensive study reports the performance of
>seven rapid prototyping systems in the areas of time, cost and quality.
>
>3D printers are affordable and easy to use rapid prototyping systems for
>personal, office or departmental use. The benchmark study finds that
>five of the seven systems qualify as 3D printers, as illustrated by the
>new Rapid Prototyping Index. A ranking methodology that incorporates 13
>performance measures, the index rates the systems in four applications
>areas: concept models, form and fit prototypes, functional prototypes
>and patterns.
>
>Todd Grimm, president of T. A. Grimm & Associates and a veteran of the
>rapid prototyping industry, conducted the five month project to assist
>companies in their evaluation and selection of 3D printers. Using
>independent organizations for all prototype construction and quality
>measurement, the benchmark offers an unbiased comparison of the seven
>rapid prototyping systems.
>
>"I produced this benchmark study specifically for design engineers who
>are trying to select the rapid prototyping system right for them," said
>Grimm. "Beyond vendor supplied data, there has been little information
>to help companies discover the true performance capabilities of rapid
>prototyping devices. Without real-world, user- supplied data, this can
>lead to poor decisions when selecting a system."
>
>The 55-page benchmark, which contains 34 charts, 5 tables and 36 images,
>reports results for three prototypes that vary in size, volume and
>complexity. The benchmark study discovered that there is a large
>variance in prototype costs, ranging from $60.53 to $267.85. The expense
>measures include system and accessories costs, annual costs, labor and
>materials.
>
>The report is available for purchase at
>http://www.tagrimm.com/benchmark/. T. A. Grimm also offers a free
>executive summary of the benchmark results at the same Web address.
>
>About T. A. Grimm & Associates
>Founded by Todd Grimm, a 13-year veteran of the rapid prototyping
>industry, T. A. Grimm & Associates, Inc. offers consulting services on
>rapid prototyping and related technologies, including competitive
>analysis, benchmarking and educational programs. The company also offers
>outsourced marketing services that include marketing plan development,
>Web optimization, copywriting and lead generation. Grimm combines his
>engineering background and technical knowledge with years of sales,
>management and marketing experience to create and implement strategic
>and tactical plans. For more information, visit the T. A. Grimm &
>Associates Web site at http://www.tagrimm.com.
>
>Todd Grimm
>T.A. Grimm & Associates, Inc.
>3028 Beth Ct.
>Edgewood, KY 41017
>Phone: (859) 331-5340
>Fax: (859) 331-5342
>Cell: (859) 240-0574
>Email: tgrimm@tagrimm.com
>Web: www.tagrimm.com
>
>Marketing Engineered for Sales Results (sm)
>
>
>[]
> <http://sg.rd.yahoo.com/mail/tagline/?http://sg.search.yahoo.com>The New
> Yahoo! Search
>- Now with
><http://sg.rd.yahoo.com/mail/tagline/?http://sg.search.yahoo.com/images>image
>search!
Dr. Ian Gibson
Currently on study leave at
National University of Singapore,
Dept. Mechanical Engineering
9 Engineering Drive 1
Singapore 117576
Tel: +65 6874 1917
Mob: +65 9087 3512
"Everything really is stupidly simple, and yet all around is utter confusion,
don't look around to find the sound that's right beneath your feet"
This archive was generated by hypermail 2.1.7 : Sat Jan 17 2004 - 15:18:19 EET