A 3D computational fluid dynamics (CFD) model was developed to predict the airflow and heat transfer in half of a pallet layer of tomatoes. The numerical and experimental results were compared, and a good agreement was obtained between both results using the root-mean-square error (RMSE) and absolute relative deviation (ARD) values as criteria. The validated CFD model was then used to minimise the product temperature heterogeneity by optimising the airflow rate and the crate design. A downward flow of the air along the central parts of the crates and an upward flow close to the lateral walls of the crates were observed. Three different total ventilated areas (TVAs) were tested to study their influence on the product temperature uniformity and cooling rate. The MTD and ATD decreased from 6.8 to 3.5 °C and from 1.5 to 0.7 °C, respectively, when the TVA was increased from 11 to 15%.