Originally posted by: Sothea

Dear,

I just found that my model is not sufficient bounded like:

Min x11+x12+x21+x22
s.t
x11+x12>=10
x21+x22>=10
x>0,y>0

so the objective value is 20 with the s1(5,5,5,5) or s2(0 10,0 10) or...

Anyone knows which solution Cplex find? how? If kind of this problem have big size, it can make infeasible solution? or problem of numerical stability?

thanks,
Sothea HONG
#CPLEXOptimizers
#DecisionOptimization

Originally posted by: SystemAdmin

> Sothea wrote:
> I just found that my model is not sufficient bounded like:

I'm not sure what you mean by "not sufficiently bounded". Your sample problem has multiple optima, which has nothing to do with bounds.
>
> Min x11+x12+x21+x22
> s.t
> x11+x12>=10
> x21+x22>=10
> x>0,y>0
>
> so the objective value is 20 with the s1(5,5,5,5) or s2(0 10,0 10) or...
>
> Anyone knows which solution Cplex find?

There is no way to predict that in advance.

> how? If kind of this problem have big size, it can make infeasible solution?

Having multiple solutions cannot make the problem infeasible, because infeasible means there are no solutions, hence no optimal solutions.

> or problem of numerical stability?

The existence of multiple optima is not generally tied to numerical stability.

Paul

Mathematicians are like Frenchmen: whenever you say something to them, they translate it into their own language, and at once it is something entirely different. (Goethe)
#CPLEXOptimizers
#DecisionOptimization