The paper presents a way of systematic construction of recombination operators for the vehicle routing problem by global convexity tests. It describes the vehicle routing problem and defines similarity measures for its solutions, and presents the results of global convexity tests; it has been demonstrated here that strong global convexity exists in almost all instances of the problem. Based on results of the tests some distance preserving recombination operators are defined. Computational experiments with a genetic local search indicate that these operators significantly ameliorate the search process and generate very good solutions. It is also shown that preservation, by an operator, of both of important features of solutions found in global convexity tests speeds up computation. The paper is, therefore, another evidence that systematic construction of recombination operators is a good means of adaptation of the genetic local search to a problem.
2
Dostęp do pełnego tekstu na zewnętrznej witrynie WWW
We use an evolutionary algorithm with a modified version of edge recombination crossover to search near optimal tours in the Traveling Salesman Problem (TSP). We introduce so called edge sensitivity to estimate the importance of having a given edge within an optimal solution. In order to test this approach we used the data files gr24, bays29 and gr48. Results we obtained show that edge sensitivities allow us to find solutions quicker than other similar algorithms with an edge recombination crossover.
JavaScript jest wyłączony w Twojej przeglądarce internetowej. Włącz go, a następnie odśwież stronę, aby móc w pełni z niej korzystać.