A new numeric method to simulate stochastic dispersion (a natural phenomenon that occurs when a magnitude cannot be associated to a specific value, but to the probability of being within a range of values) is proposed and applied to predict Radio Frequency Vacuum (RFV) drying of timber. A theoretical formulation of the method is described and complemented to take into account the frequency distribution of the timber initial moisture content, so that it can be applied to industrial runs. Experimental data obtained from mixed western hemlock and amabilis fir dried in a commercial RFV kiln are used to validate the stochastic model, and the results are compared through moisture content histograms and probability charts. A numerical example is shown in order to provide an idea of the movement of the moisture profiles during RFV drying.