This paper presents an experimental study of the thermal inertia of a pallet loaded with returnable plastic crates containing primary packages of smoked ham. Based on this, food quality variations within the pallet were also investigated. Thermal time constants from 83 sensor locations were identified by studying the temperature changes when the pallet was exposed to instant temperature drops (16 C - 2 C) and temperature elevations (2 C - 16 C). The thermal time constants were used in microbiological prediction models to calculate the maximum difference in shelf life between packages at the two most extreme spots in the pallet unit load, when temperature elevated from 4 C to a higher temperature (ranging from 4.5 C to 12 C), during different periods of time (ranging from 0.5 h to 200 h). The results showed a maximum difference in shelf life of approximately 1.8 days. The identified thermal time constants were also used to calculate the maximum difference in shelf life between packages at the two most extreme spots of a pallet unit load, in a real chilled food supply chain lasting for about 2.5 days. This resulted in a maximum difference of 0.1 days. The results imply that the location of a product in a pallet has a relatively low influence on the product shelf life. This means that a temperature sensor used for calculating the predicted shelf life of a product, can be placed relatively far from the product itself (e.g. on the secondary package or even on the pallet) without jeopardizing the reliability of the resulting shelflife prediction. However, the results also emphasize the importance of continuous temperature monitoring along the entire chilled food supply chains.