My Linear Programmation book tells me to consider the closed and limited Convex Polytope defined as

$$S=\left \{ \right.\bar{x}:A\bar{x}=\bar{b} ,\bar{x}\geq 0\left. \right \}$$

But , does this equation only define the points that belongs to the hyperplanes that contains the convex polytope ? Thanks

  • $\begingroup$ It is a strict equality $A\overline x = \overline b$ and no inequality so any solutions must lie on planes or lines and can't lie in the "volume" or what you would call it. $\endgroup$ – mathreadler Jul 22 '17 at 12:33
  • $\begingroup$ So , am i right ? $\endgroup$ – JDOE Jul 22 '17 at 12:47
  • $\begingroup$ I don't know. I am not sure I understand your question. $\endgroup$ – mathreadler Jul 22 '17 at 13:41
  • $\begingroup$ A polytope is the set of all points included in a system of hyperplanes or just the points in the hyperplanes ? Because my book calls polytope the set i wrote in the main post $\endgroup$ – JDOE Jul 22 '17 at 13:44
  • $\begingroup$ $S$ is the set of all $\overline x$ which fulfil both the equality and the inequality. I don't get the question. It seems you are more confused about semantics. I don't know how your book defines those words. $\endgroup$ – mathreadler Jul 22 '17 at 14:06

This field is not my meat, but let me try to answer according to my understanding.

This is the intersection of the $(+,+,\dots,+)$-cone of your domain space with a (translated) linear subspace thereof. For instance, if the domain space is just the plane, you’re talking about either a single point in the first quadrant, or the first-quadrant part of a line, or all the first quadrant.

If the domain space is $\Bbb R^3$, you’re looking either at a single point in the first octant, or the first-octant part of a line, or the first-octant part of a plane, or the all the first octant.

I don’t know what the book means by “limited”: if they mean compact, that is certainly not the case; if they mean closed, that is all right.

The set defined here turns out to be convex, but you are making a mistake if you are trying to see this definition as a modification of the book’s definition of a convex set. I would say that there are no hyperplanes specifically mentioned here, though if you search for them, you will find them.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.