This paper presents an analytical model for the prediction of cooking time of meat products in industrial steam ovens. To achieve this aim, the paper first develops a mathematical model for cooking meat, which is numerically solved and validated on the basis of the outcomes of an appropriate experimental campaign. Numerical simulations are then performed setting different values of sample sizes, with the aim to derive a parameterised model able to analytically reproduce the time-temperature curves of meat samples. As input the model developed requires an appropriate "translation" parameter, describing the shape of the time-temperature curve as a function of the sample size and diameter; as output it gives an estimate of the cooking time. The "translation" parameter is provided as a result of the numerical simulations for a wide range of sample size and length. The analytical model is validated by comparing the predicted cooking time with experimental cooking data related to time-temperature curves of seven meat samples. The comparison shows that the average percentage deviation between experimental results and model predictions is about 4.6%, proving good performance of the model developed. The model can be successfully used to estimate the meat cooking time starting from typical values of meat parameters, and, due to its simplicity, it appears to be suitable for direct implementation as a tool to monitor and automate the industrial meat cooking treatments by means of computer control. (c) 2008 Elsevier Ltd. All rights reserved.