This study aims to evaluate the ability of the Near Infrared Reflectance spectroscopy (NIRs) technique to analyse texture-related characteristics of sous-vide pork loins at different times of cooking. For that, pork loins were sous-vide at 70 degrees C for 1, 2, 4, 6 and 8 h. Cooked samples were analysed by means of NIRs, instrumental (cooking loss, pH, moisture, hydrolysed collagen and texture profile analysis) and sensory analysis. Classification and predictive techniques of data mining were applied on the obtained data. Sous-vide loins were correctly classified as a function of time of cooking and their texture-related characteristics were predicted accurately, achieving correlation coefficients (R) higher than 0.5 and Mean Absolute Scaled Errors lower than 1 for most parameters. Thus, it is demonstrated the capability of NIRs to analyse most texture-related parameters of warm loin samples, and it may be recommended as a rapid and automatic techniques to stablish optimal cooking conditions of food.