In multi-objective optimization, the objective functions induce a partial (or often a pre-) order on the search space, by some domination relation in the Pareto sense. i.e., two points are incomparable or indifferent if they differ in two objective functions with a different sign, whereas if one point is consistently not better than the other point in any objective and at least worse in one objective, then the second dominates the first. This partial order can be represented by a graph. I want to know whether there are studies of such graphs, by researchers from maths/graph theory/multi-objective optimization.