Intlinprog returns 0 as the best result

I have an optimization problem consisting of 48 variables. When used milp or ga it is returning 0 as the best value eventhough i have mentioned the indices for intcon.Is there some sort of tuning that needs to be done ?

[John D'Errico]

Um, so what is the problem?

You tell us that when you used milp or GA, you got 0 as the result. But now you tried intlinprog, and you got 0. Why is there some confusion here? Why is any tuning that is needed?

Suppose you went to your doctor, who told you you have a cold, that in 5-10 days, you will be over it. Then you want to another doctor, got a second opinion, and they told you exactly the same thing. Then you went to a third doctor, and they too gave you the identical diagnosis. So, now, you are asking us?

You have a cold. ;-)

The point is, unless you give us some valid indication as to why you think those three identical results are wrong, the answer is, you have a cold. That you have "mentioned the indices for intcon" tells us absolutely nothing.

[Mondeep maz]

The solution is not satisfying the constraints. That's the problem. Its like u see 2+2+2 should b equal to 6. But according to the solution it is coming 0+0+0 equal to 6. And i dont know why

[John D'Errico]

So show what you did. Apparently, you did something wrong. We cannot guess what. Or, maybe you are just interpreting the output incorrectly.

If I had to guess, you have formulated the problem incorrectly, since all solvers seem to generate the same solution. You are solving a different problem than the one you think you are solving. But all we have been told is this is a 48 variable problem, and that at least some variables are constrained to be integer. Could you possibly have set both the lower and upper bounds for your variables to 0? Maybe. How can we read your mind, or see into your computer?

[Mondeep maz]

lb =zeros(1,48); ub =zeros(1,48); for j=1:length(lb); if T_in(i)==j ub(j:j+P(i))=1; UB(i,:)= ub; end end for m = 1:length(ub) if ub(m)==1 int(m)=m; end end Intcon= int(int~=0); A = 2*ones(1,48); b(i)= round(Ereq(i)); fog = 2*RTP; [s,fval] = intlinprog(fog,[],[],[],A,Ereq(i),lb,UB(i,:));

Here is a part of my code. Just check it if there is anything wrong. The bounds are getting updated at every iteration. So for example for a certain iteration it would look like [0,0,0,0....,1,1,1,1,1...0,0,0]. Tell me if i can do this because without the Intcon part i am getting solutions.