Search found 110 matches

by bussieck
2 weeks ago
Forum: Announcements
Topic: GAMS Courses and Workshops
Replies: 0
Views: 149

GAMS Courses and Workshops

Dear GAMS users, Please find below a list of forthcoming GAMS courses: --------------------- Courses 2018 ------------------- Jun 11–Jun 13: Basic GAMS Mode​​ling - An Introductory Class Jun 11–Jun 15: Basic to Advanced GAMS Modeling Jun 12–Jun 15: Energy and Power System Modeling in GAMS Jun 13–Jun...
by bussieck
3 weeks ago
Forum: Syntax
Topic: Define set with integer scalar
Replies: 4
Views: 141

Re: Define set with integer scalar

You can't. If you read what Renger and I wrote you will understand the reasons. What users often do in such a situation is to a) estimate the size of the set and build a super set and then b) use a dynamic set with the actual size: set super / 1*10000 /; scalar num; num = uniformInt(1,card(super)); ...
by bussieck
3 weeks ago
Forum: Syntax
Topic: Define set with integer scalar
Replies: 4
Views: 141

Re: Define set with integer scalar

You can even use the value of a scalar with a data statement to do this: Scalar test1 / 10 /; $eval TEST1 test1 set test / 1*%TEST1% / This gives you set elements 1*10. You cannot use the value of test1 after it has been assigned during execution (before the set declaration happens before the execut...
by bussieck
3 weeks ago
Forum: Modeling
Topic: Linearization of the SMIN function?
Replies: 5
Views: 131

Re: Linearization of the SMIN function?

If you read Fred's code carefully, he is using x.up in the smax operator. The upper bound of the variable (not the variable itself). x.up can be replaced by any number that is to be guaranteed an upper bound on the x variable. We have seen models where users just use 1e9 or other crazy numbers. Whil...
by bussieck
4 weeks ago
Forum: Announcements
Topic: GAMS Distribution 25.1.1 Announcement
Replies: 0
Views: 61

GAMS Distribution 25.1.1 Announcement

Hi GAMS Users, We started distributing GAMS Release 25.1.1 on May 19, 2018. This release contains the first preview version of GAMS Studio - a completely new integrated development environment for GAMS. GAMS Studio is still in a very early stage of development, but we believe it is already mature en...
by bussieck
1 month ago
Forum: Syntax
Topic: how a parameter used as set in equation
Replies: 2
Views: 87

Re: how a parameter used as set in equation

I suggest investing a little more time learning before you seek help here, see the rules of the forum ( https://forum.gamsworld.org/app.php/rules ). There is good tutorial material out there. As a starter: https://www.gams.com/latest/docs/UG_MAIN.html#UG_Tutorial_Examples and https://www.youtube.com...
by bussieck
1 month ago
Forum: Modeling
Topic: Error in MIQCP sample model
Replies: 3
Views: 96

Re: Error in MIQCP sample model

Sean,

**2 is real-power (rpower, see https://www.gams.com/latest/docs/UG_Par ... s_dmrpower) and requires NLP model type. For quadratic models you need to use power or sqr.

-Michael
by bussieck
1 month ago
Forum: Modeling
Topic: Error: integer infeasible No solution exists
Replies: 1
Views: 63

Re: Error: integer infeasible No solution exists

That's not an error. Optimal, infeasible and unbounded are the three acceptable states of a mathematical program. You might not want to have an infeasible model but that's your issue. There are lots of good write-ups about how to deal with infeasible model, e.g. Bruce McCarl's excerpt which you find...
by bussieck
1 month ago
Forum: Solvers
Topic: normal compilation but Constraints not satisfied
Replies: 1
Views: 62

Re: normal compilation but Constraints not satisfied

"Normal completion" is the solver status code. You need to look at the model status. Infeasible or unbounded are also very fine model status codes in optimization. Perhaps BARON found your model to be infeasible.

-Michael
by bussieck
1 month ago
Forum: Modeling
Topic: MIP - CPLEX solver does not give optimal solution in a facility location problem
Replies: 4
Views: 126

Re: MIP - CPLEX solver does not give optimal solution in a facility location problem

I said optCR (not scaleOpt)! If I set optCR to 0 both models return the same objective: max7: Proven optimal solution. MIP Solution: 25415486.875000 (55 iterations, 0 nodes) Final Solve: 25415486.875000 (20 iterations) Best possible: 25415486.875000 Absolute gap: 0.000000 Relative gap: 0.000000 all1...