Distributed Evolutionary Algorithms to TSP with Ring Topology

Distributed Evolutionary Algorithms (dEAs) are stochastic search methods applied successfully in many searches, optimization, and machine learning problems. This paper proposes a dEA based on IGT. The dEA uses a ring topology ,with each population choosing a random individual from it in a migration interval. The individual will be sent to next population and replace the worst individual of the population. It can be seen from the experiment that dEA with Ring transmission can produce better solutions when the population size of a TSP problem is large enough. The numerical results show the utility, versatility, efficiency and potential value of the proposed distributed evolutionary algorithm.