The application of a genetic algorithm to trunk network routing table optimisation

An approach to routing table optimisation is taken, based on combining a genetic algorithm with Sinclair´s (1991) analysis method. Genetic algorithms are search algorithms based on the mechanics of natural selection and natural genetics. They combine survival of the fittest [amongst a population of] string structures with a structured yet randomised information exchange [between population members] to form a search algorithm with some of the innovative flair of human search. In every generation, a new set of artificial creatures (strings) is created using bits and pieces of the fittest of the old; an occasional new part is tried for good measure. The paper begins by describing the network model used and, briefly, both the underlying analysis method and the earlier heuristic optimisation methods. The basic principles of genetic algorithms are presented, and the details of the particular genetic algorithm used are described. The paper concludes with a comparison of the execution times and results obtained for an example network with those from Sinclair´s heuristic method, and some suggestions for possible future work