Modelling and solving the perfect edge domination problem

Abstract

A formulation is proposed for the Perfect Edge Domination Problem and some exact algorithms based on it are designed and tested. So far, perfect edge domination has been investigated mostly in computational complexity terms. Indeed, we could find no previous explicit mathematical formulation or exact algorithm for the problem. Furthermore, testing our algorithms also represented a challenge. Standard randomly generated graphs tend to contain a single perfect edge dominating solution, i.e., the trivial one, containing all edges in the graph. Accordingly, some quite elaborated procedures had to be devised to have access to more challenging instances. A total of 736 graphs were thus generated, all of them containing feasible solutions other than the trivial ones. Every graph giving rise to a weighted and a non weighted instance, all instances solved to proven optimality by two of the algorithms tested.

Some New Results on Efficient and Perfect Edge Domination of Graphs

“…It includes [2,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19]. Papers on perfect edge domination are less frequent, The reader can refer to [1,20,21,22]. In the latter paper, the authors describe ILP formulations for the PED problem, together with some experimental results.…”