The vehicle routing problem consists in determining a group of optimal delivery routes that allow satisfaction of the demand from customers spread along a network, which can be represented as a graph type structure. This paper presents two GRASP metaheuristic algorithms for the above mentioned problem, considering the capacity and shared demand of the customers, solving besides, the 3D bin packing problem within of such vehicles. In this paper the solution obtained is compared with a greedy solution and two hybrid solutions (greedy and random). The results obtained show that the GRASP algorithm obtains a better quality solution for this kind of problem.