Efficiency of Constraint Forms

The Solver recognizes the case where the constraint left hand side is a decision variable, or a set of decision variables. As long as the corresponding right hand sides are constant (i.e. not dependent on any of the variables), these constraints are specially treated as bounds on the variables. The most common instance of a bound on a variable is a non-negativity constraint such as A1 >= 0, but any sort of constant bounds are efficiently handled by both the linear and nonlinear Solvers.

There is no difference in terms of efficiency between a constraint entered (for example) as A1 <= 100 or as A1 <= B1 where B1 contains 100; the Solver recognizes that B1 is equivalent to a constant. The form A1 <= B1 is usually better from the standpoint of maintainability of your optimization model.

On the other hand, a constraint right hand side which is a formula -- even a simple one like 2+2 -- will incrementally increase the solution time for the model. Because the Solver doesn't have the facilities to recognize the right hand side "on the fly," it treats any formula as a RHS potentially dependent on the variables, and internally creates a constraint "LHS - RHS >= 0" -- even if the formula really was a constant bound on a variable. It is better to place whatever formula you need into a cell, and reference that cell as the constraint right hand side.