View Single Post
  #4   Report Post  
Tushar Mehta
 
Posts: n/a
Default

For non-linear non-convex problems, Solver, like most other
optimization packages, cannot guarantee a global optimum. It might
work depending on the options selected as well as the starting values
for a specific solution attempt.

As far as not changing any setting from when solver used to work, how
could you possibly do that? I used Solver a few times yesterday, and I
couldn't tell you what my option settings were! It is hard to imagine
someone being in the habit of noting or otherwise memorizing the
settings for max.time, max.iterations, precision, tolerance,
convergence, estimates, derivaties, search, etc.

On your colleague's computer and your own, open the workbook you are
having problems with. Compare the values of *all* the cells in the
model. Open the solver dialog box, click Options. Note the values of
*all* 12 settings.

Are there any differences between the two?

Now, what happens if you ask Solver to solve the problem?

--
Regards,

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

In article ,
says...
H Mike, Thanks for your response, but I have already looked at that and I
have not changed any of my settings from when the solver used to work for me
with no problems. Any other suggestions?
Thanks in advance.