A multi-scale three-dimensional computational fluid dynamics (CFD) model was developed to predict airflow, heat and mass transfer in a typical full loaded cool storage. In order to reduce the computational costs, the porous media parameters of the bed of the apples inside the vented containers were extracted using a series of wind tunnel CFD simulations and then applied in the cool storage model. The model was validated against experiments by means of velocity, product temperature, and product weight loss measurements in cool storage. The errors of about 23.2 and 9.1% were achieved for velocity magnitude prediction in the cool storage and the product weight loss after 54 days of cooling in the loaded cool storage, respectively. The model over predicted the cooling rate of the products temperature; however, it showed a good trend of cooling rate. About 11°C difference was observed between the hottest and the coldest product temperatures at half cooling time by experiments that were in good agreement with the simulation results with about 10°C. This difference changes versus time of cooling and reached to about 4°C at the end of the cooling time. The product's temperature heterogeneity was predicted 1.9°C between the 7 and 9 hours of cooling and reduced to 0.6°C at the end of the cooling. The multi-scale model was capable of predicting air velocity, product temperature, and weight loss with reasonable accuracy and was reliable enough for numerical studies on larger domain with high reduction in computational costs.