Genetic algorithm and variable neighborhood search for point to multipoint routing problem

Abstract

Routing of data in a telecommunication network is important due to the vast amount of data flow in the network. Message Scheduling Problem (MSP) is an important aspect of data routing. It is the process of scheduling a set of requests where each set has a single source and multiple destinations. In this paper, we propose a hybrid Genetic Algorithm to determine the order of request to be scheduled. The proposed algorithm embeds a different Steiner Tree algorithm than those reported in [3]. The second part of the study involves designing three algorithms based on the Variable Neighborhood Search. These algorithms differ in the local search heuristics employed in searching the neighborhood. The performance of these algorithms are compared on a modified set of data taken from the OR library. © Springer-Verlag Berlin Heidelberg 2008.