General Variable Neighborhood Search for the Minimum Stretch Spanning Tree Problem

Abstract

For a given graph G, minimum stretch spanning tree problem (MSSTP) seeks for a spanning tree of G such that the distance between the farthest pair of adjacent vertices of G in tree is minimized. It is an NP-hard problem with applications in communication networks. In this paper, a general variable neighborhood search (GVNS) algorithm is developed for MSSTP in which initial solution is generated using four well-known heuristics and a problem-specific construction heuristic. Six neighborhood strategies are designed to explore the search space. The experiments are conducted on various classes of graphs for which optimal results are known. Computational results show that the proposed algorithm is better than the artificial bee colony (ABC) algorithm which is adapted by us for MSSTP.