Greedy randomised adaptive search procedures for topological design of MPLS networks

Abstract

In this paper, the IP/MPLS network cost optimisation problem of selecting localisation of nodes and links, combined with link`s dimensioning, is discussed. As the considered problem is hard, we discuss and propose greedy randomised adaptive search procedure (GRASP) based solution method. GRASP is an iterative randomised sampling technique which combines adaptive randomised greedy function in constructing initial solution with local search optimisation. The effectiveness of the method is illustrated by means of a~numerical study. We compare the GRASP results with results for both exact and heuristic methods obtained in previous research concerning topological design problem.