Improved genetic algorithm for solving optimal communication spanning tree problem

Abstract

Optimal Communication Spanning Tree (OCST) is a well-known NP-hard problem on the graph that seeks for the spanning tree with the lowest cost. The tree cost depends on the communication volume between each pair of nodes. This paper proposed an improved Genetic Algorithm combining with Ahujia and Murty's Tree Improvement Procedure. The proposed algorithm was experimented on known benchmark tests which used in many papers related to OCST problem, and random instances from 200 to 500 vertexes. The experimental results show that the proposed algorithm is better than the heuristic and out-performance the most recent evolutionary algorithm approaches. © Springer-Verlag Berlin Heidelberg 2013.