Stuart Altmann on thu 12 aug 99
------------------
David Hendley wrote: =22If you are comfortable with glaze calculations, and
with the advent of computers, it=92s a quick and simple matter, when you see
a recipe you=92d like to try, to make these =5Bingredient=5D substitutions.
Many, many potters, however, don=92t know this, so the old recipes that
include needlessly expensive ingredients live on.=22
David=92s comment and my own work in an area called linear optimization
stimulate me to ask a question of those of you who know various glaze
calculation programs. Does any glaze calculation program enable one to
calculate the =2Aleast cost=2A recipe for a given glaze composition? As =
David
points out, dolomite is a less expensive source of magnesium than magnesium
carbonate, and talc is cheaper still. How does one simultaneously take
into account all the other glaze components and their costs, in such a way
as to minimize the cost of the glaze without changing its composition?
This problem seems to be exactly analogous to the first problem proposed in
linear optimization, the least-cost diet problem, which was formulated by
George Stigler and solved independently by Dantzig (1963) and by Steve
Wagner and me (1976, 1978). The least cost diet problem is this: What is
the cheapest combination of available foods that enables one to fulfill all
of one=92s nutrient requirements but not get too much of any diet component,
e.g., because beyond a certain point it is toxic or otherwise unacceptable.
Today, efficient programs for solving all sorts of linear and non-linear
optimization problems are available. (My favorite is LINDO, by Linus
Schrage.)
Has linear optimization (linear programming) been incorporated into any
glaze program? If not, we are often paying too much to make our glazes.
Stuart Altmann
email: salt=40princeton.edu office 'phone: 609/258-4520
David Hendley on mon 16 aug 99
I don't think there is a glaze calculation program that will do what
you are asking: hit the 'least-cost key' and the cheapest recipe
for the glaze formula in question will appear.
I don't think anyone reading Clayart would want or use this
feature, anyway. Just as most (all?) of us are not concerned with
calculating our own personal 'least-cost diet' (anyone here eat that
horribably inefficient source of nutrition, meat, occasionally?) we are
not concerned with minimizing the cost of our glazes.
It's just too insignificant as a cost to us.
Now, just as determining a least-cost subsistance diet is important in
areas of the world with chronic stravation, there could be instances
where the most important characteristic of a glaze would be its cost.
If cost were of paramount importance, however, I would suggest
that not only should the recipe be calculated for low cost, but the
glaze formula itself could be altered, to achieve a similar glaze for
even less cost.
In other words, even if a program could do what you are asking,
you would be over-looking another valid road to low cost.
Glaze programs WILL automatically calculate the cost of a given
recipe, so it is a simple matter to compare a recipe with variations
using different ingredients. To make an accurate comparison, I think
fired weight would have to be figured in the calculation. For example,
'X oxide' might cost more than 'X carbonate', but it might actually
be cheaper because more of the material is still left after firing, as a
glaze component. It would also be easy to take this into account, as
the glaze programs also show the fired weight of a recipe.
Personally, I am concerned about the cost of a glaze, but this is far
from my my most important concern in glaze design. Only a gross
difference in the cost of different ingredients, such as magnesium
carbonate ($3) compared to talc (12 cents) would affect my decision
as to which to use in a glaze.
One more point about materials substitution: a substitution will
often slightly change a glaze, even if the theoretical formula stays
the same.
Usually the cheapest glaze ingredients are naturally occuring
un-refined substances that often have traces of many different
minerals and also suffer from variability.
David Hendley
Maydelle, Texas
hendley@tyler.net
http://www.farmpots.com
At 03:32 PM 8/12/99 EDT, you wrote:
>----------------------------Original message----------------------------
>------------------
>David Hendley wrote: "If you are comfortable with glaze calculations, and
>with the advent of computers, it s a quick and simple matter, when you see
>a recipe you d like to try, to make these [ingredient] substitutions.
>Many, many potters, however, don t know this, so the old recipes that
>include needlessly expensive ingredients live on."
>
>David s comment and my own work in an area called linear optimization
>stimulate me to ask a question of those of you who know various glaze
>calculation programs. Does any glaze calculation program enable one to
>calculate the *least cost* recipe for a given glaze composition? As David
>points out, dolomite is a less expensive source of magnesium than magnesium
>carbonate, and talc is cheaper still. How does one simultaneously take
>into account all the other glaze components and their costs, in such a way
>as to minimize the cost of the glaze without changing its composition?
>
>This problem seems to be exactly analogous to the first problem proposed in
>linear optimization, the least-cost diet problem, which was formulated by
>George Stigler and solved independently by Dantzig (1963) and by Steve
>Wagner and me (1976, 1978). The least cost diet problem is this: What is
>the cheapest combination of available foods that enables one to fulfill all
>of one s nutrient requirements but not get too much of any diet component,
>e.g., because beyond a certain point it is toxic or otherwise unacceptable.
>Today, efficient programs for solving all sorts of linear and non-linear
>optimization problems are available. (My favorite is LINDO, by Linus
>Schrage.)
>
>Has linear optimization (linear programming) been incorporated into any
>glaze program? If not, we are often paying too much to make our glazes.
>
>Stuart Altmann
>
>email: salt@princeton.edu office 'phone: 609/258-4520
>
Tony Hansen on tue 17 aug 99
> Does any glaze calculation program enable one to
> calculate the least cost recipe for a given glaze composition? As David
> points out, dolomite is a less expensive source of magnesium than magnesium
> carbonate, and talc is cheaper still. How does one simultaneously take
> into account all the other glaze components and their costs, in such a way
> as to minimize the cost of the glaze without changing its composition?
I'd be cautious about this one. The selection of materials to supply oxides is
potentially very complex, there are so many more issues as stake than just
cost. For example, mag carb is refractory and isn't easily dissolved by the
melt. It contributes to crawling and is a matting agent in low fire. It is usual
employed for is mineralogy, not its chemistry.
--
T o n y H a n s e n thansen@digitalfire.com
Don't fight the dragon alone http://digitalfire.com
Calculation/Database Software for Ceramic Industry
--
| |
|
