Thawing is one of the important processes of the food industry. Its speed and conditions can influence the quality and initial concentration of the microbial population in raw materials used for further processing. Experimental prediction of the optimum conditions of the thawing process is very time consuming and expensive. Therefore, a mathematical model of the process can help in the design of efficient thawing apparatus and process conditions. This paper compares experimental data of different thawing procedures with numerically calculated predictions using the software package food product modeller (FPM). The model calculations agree very well with experimental data on three different thawing process arrangements with reasonable accuracy. The verified mathematical model for convective thawing was selected for sensitivity analysis of the process parameters. The time of thawing was calculated as a function of different parameters for three different points in the thawed food piece. Graphs of the relative change of thawing time as a function of the relative change of the process parameters have been constructed. From these graphs one can read that the air temperature, surface heat transfer coefficient (SHTC), thickness of the layer, water content and enthalpy are the parameters with the greatest influence on thawing time. On the other hand the thermal conductivity or initial freezing point does not influence the thawing time as much. (C) 2002 Elsevier Science Ltd. All rights reserved.