Algoritmo evolutivo com metodologia híbrida para elaboração de rotas de distribuição

Abstract

In the logistical chain of parcel delivery in general, the difficulties of managing the routes of distribution are remarkable. The present work has the objective of developing a software prototype capable of optimizing the order distribution route, using as a case study the registered orders of the post office in the city of Coromandel / MG. It is highlighted in this paper the option of elaborating routes that facilitate their use according to the type of vehicle used by the delivery of orders. An OpenStreetMap database was used to locate the geographical coordinates of residences and streets, and the routes are defined through a hybrid genetic algorithm based on travelling salesman problem and also with the Dijkstra algorithm to define the minimum cost path between the vertices of the problem. This work is also described as elaboration stages, as well as the results obtained experimentally. Finally, it is concluded through this work that the use of hybrid algorithms presents a great potential for the resolution of routing problems, including those in which the type of vehicle to be used in the deliveries is considered.