Development of a genetic algorithm for the solution of a load allocation problem

The purpose of the present study case was to determine the best feasible solution for a typical load allocation problem with different constraints associated with the technical conditions of the case. The model proposed intends to maximize the profits when transporting the loads, by an adequate transport plan, according to the capacities of the vehicles, the characteristics and the demand of the loads. Taking into account the magnitude of the problem and its computational complexity, to solve the problem. The Metaheuristic Genetic Algorithms were implemented, which allows solving problems of optimization and search. The algorithm was developed in Visual Basic for Applications, defining and creating an initial feasible population, from which it proceeds to develop the entire evolutionary process associated with the genetic algorithm. Finally, the results were compared with those obtained at the GUSEK software, which offers the optimal solution for the allocation problem studied. From this comparison, it is evident that the method developed in the present research offers feasible solutions very close to the optimal one with a computational cost considerably lower than offered by the GUSEK program.