The propagation of parameter uncertainties which fluctuate in space and time through a thermal sterilization process of a conductive heated food was calculated by means of a Monte Carlo and variance propagation algorithm. The parameters were modelled as autoregressive random waves. A good agreement between both methods was obtained for the mean and variances of the temperature field inside the can. The variance propagation algorithm was advantageous in terms of computer time. It is shown that high frequency fluctuations in space and time of the retort temperature and the surface heat transfer coefficient are attenuated because of the thermal inertia of the food.