The discussion revolves around the formulation of linear programming (LP) constraints, focusing on whether variables must be isolated on one side of the inequality or equation. Participants are exploring the general principles governing LP constraints and their structure.
The discussion is ongoing, with participants providing insights and clarifications regarding the formulation of constraints. Some guidance has been offered about the implications of having variables on both sides, and there is an acknowledgment of differing practices in constraint formulation.
There is mention of a test where constraints were set equal to zero and variables isolated, raising concerns about potential grading implications for not following this format. Participants are reflecting on their understanding of standard practices in linear programming.
This would normally be written as x1 > 1. Did you mean x1 + 2x2 > 3?bobbo7410 said:Homework Statement
I don't exactly have a specific example, I'm looking for the general idea behind LP constraints.
Homework Equations
Within the constraints, do the variables have to be all to one side?
For instance, X1 + 2 > 3 is a standard constraint
No. This is equivalent to 2 > 1, which is always true, so doesn't constrain things in any way. Did you mean x1 + 2 > x2 + 1?bobbo7410 said:would for example
X1 + 2 > X1 + 1
still be acceptable as a constraint?
Mark44 said:You can always move variables from one side of an equation/inequality to the other.
Mark44 said:To use matrix methods (e.g., Simplex tableau), you need to have all the variables on one side, and the constant on the other. I'm pretty sure this is correct, although I haven't done one of these problems for a good long while.