Should I process the data or add a new constraint to achieve the target?

Question

I have an MILP as below

$\begin{equation} \begin{array}{*{35}{l}} \underset{d_{u,c}}{\max}\hspace{1mm}\hspace{1mm}\sum_{u=1}^{U}\sum_{c=1}^{C}d_{u,c}\omega_{u,c}\\ \text{}\text{subject to }\text{ C1:} \hspace{2mm}1\le \sum_{c=1}^Cd_{u,c}\le 10,\forall u, u=1,\cdots, U, \\ \text{}\hspace{16.5mm}\text{ C2:} \hspace{2mm}\sum_{u=1}^U d_{u,c}\le 50,\forall c, c=1,\cdots, C, \\ \end{array} \end{equation}$

In addition, I want to enforce that for $\omega_{u,c}<t_{\min}$, $d_{u,c}=0$.

Should I replace the entries of $\omega_{u,c}$ that are less than $<t_{\min}$ to 0 or negative value, or add a new constraint?

Answer 1

Neither. You should delete $d_{u,c}$ from the model whenever $\omega_{u,c} < t_\min$.

Answer 2

To add to @prubin's answer (which is the correct answer) if you somehow need to choose between the two obtions, my advice would be to add the constraints (or better, change the bounds on the variables). Obviously, a good solver will immediately identify the variables with a negativ profit and set their value to zero, but a bad/homebrew solver might not. Thus, you help the solver more by giving the values of the fixed variables explicitly, compared to indicating the values.