Closest assignment constraints in discrete location problems

The objective of this paper is to identify the most promising sets of closest assignment constraints in the literature of Discrete Location Theory, helping the authors in the field to model their problems when clients must be assigned to the closest plant inside an Integer Programming formulation. In particular, constraints leading to weak Linear Programming relaxations should be avoided if no other good property supports their use. We also propose a new set of constraints with good theoretical properties.

► Complete classification from the point of view of the dominance. ► Some constraints imply integrality of some variables, reducing the number variables. ► Some constraints imply classical constraints, reducing the size of the formulation.