SIMBA - POP is a model developed in Martinique (France), which uses the concepts of cohorts and cohort chains to simulate the population dynamics of a commercial plantation of bananas. The original model was slightly modified by Ochoa (2010) in order to correct for an error in the flowering rate curve. The model was programmed in EVOLUCION 4.0. SIMBA - POP was then calibrated and validated for the normal climatic pattern in Uraba (Antioquia). Care was taken that the model adequately represent bagging and harvesting cycles, and the time at which maximum values of bagging (flowering) were reached. To evaluate the calibration we used average deviation (DM), the average error of prediction (EMP), the error mean square of prediction (ECMP), the root of the error mean square (RECMP), mass curves and correlation coefficient. Results showed that SIMBA-POP is a suitable tool for simulating banana crops in Uraba; the model reproduced quite well the times of the peaks and the valleys of production, although the magnitude of these peaks and valleys is considerably sensitive to model parameters.