This work presents a hybrid strategy for optimization of Capacitated Vehicle Routing Problem (CVRP) that employs Genetic Algorithms (GA) combined with the heuristics of Gillett & Miller (GM) and Hill Climbing (HC). The first heuristic is used to incorporate feasible solutions in the initial population of the GA while the second is responsible for the refinement of solutions after a certain number of generations without improvements. The computational experiments showed that the proposed strategy presented good results for the optimization of CVRP with respect to the quality of solutions well as the computational cost.