More information: I just found some information in the help system stating
"When the relative change in the target cell value is less than the number in
the Convergence box for the last five iterations, Solver stops."
This seems to be what's happening. By selecting "Show Iteration Results" in
the Options window I can see the five iterations. And nothing is changing
from one attempt to the next. I'm even looking at the numbers out 25 places.

Why isn't at least one variable changed?

Rick

"Rick" wrote:

Tushar,

Well, I've tried several versions of this model without success.
First of all, maybe I didn't understand your reference to linear
optimization. Were you telling me to select "Assume Linear Model" in the
Options window? That didn't seem to make a difference.

Is there any way I can send you the worksheet I'm working with to show you
all of the constraints and my latest attempt?

The solver only makes a half dozen trials which all seem to be the same and
returns a solution that does not fit the constraints. And, it doesn't matter
what values I start with. Nothing changes between the initial setpoints and
those returned by the solver.

Rick

"Rick" wrote:

Thanks. I'll give this a try. I didn't realize I could add constraints to a
cell containing a formula. It will be interesting as there are four gears in
the system with several levers allowing additional multipliers for the final
ratio. I also have to be sure the same gear isn't selected for each position
as I only have one of each gear.

What are my limits on constraints?

"Tushar Mehta" wrote:

Assuming you have created a linear optimization (much faster than a
non-linear problem and guaranteed to give you a global solution), you
can maintain the linear nature of the problem with the following
approach:

Suppose you have the gears (each with an integer number of teeth) in
column A. For my test I used A1:A6, which had the values 1,2,3,5,8,
and 13.

Then, designate a corresponding range in some other column as the
'selector' range. I picked C1:C6. In C7 enter the formula =SUM
(C1:C6). Designate some other cell as the 'selected gear' cell. I
picked D1. D1 should contain the formula =SUMPRODUCT(A1:A6,C1:C6)

In Solver, add the contraints: C7<=1 and C1:C6 are binary.

The way the above works is as follows: Because of C7, only one of
C1:C6 can be 1 (the others will be zero). Then, the SUMPRODUCT will
multiply the selected gear by 1 and all others by zero. The resulting
'sum' will be the selected gear.

--
Regards,

Tushar Mehta
www.tushar-mehta.com
Excel, PowerPoint, and VBA add-ins, tutorials
Custom MS Office productivity solutions

In article ,
says...
I€=3Fm trying to use Solver to determine gears in a gear train to achieve a
specified ratio input to output. My problem is I don€=3Ft have a continuous
range of gears. Of course, the gears have to have an integer as the number of
teeth, but how do I set the constraint to be a listing of the available gears?

I€=3Fve been all over the help system furnished with Excel and find notes about
setting the constraint as a range reference, but I get an error message when
I try it. €œUnequal number of cells in Cell Reference and Constraint.€�

Any ideas?