This paper proposes a new framework for the optimization of logistic processes using ant colonies. The application of the method to real data does not allow to test different parameter settings on a trial and error basis. Therefore, a sensitive analysis of the algorithm parameters is done in a simulation environment, in order to provide a correlation between the different coefficients. The proposed algorithm was applied to a real logistic process at Fujitsu-Siemens Computers, using the set of parameters defined by the analysis. The presented results show that the ant colonies provide a good scheduling methodology to logistic processes.