Quality of dry peas cooked at 70 degrees C, 80 degrees C, 90 degrees C and 100 degrees C for up to 240 min was assessed using both sensory evaluation and instrumental measurement. The quality changes of the peas cooked at each temperature were modelled using the primary models, i.e. hyperbolic model and its two linear versions. Both Davey＇s modified Arrhenius model and neural network (1-9-16) model were used as secondary models to predict the parameters of the primary models from temperature. Compared to the first order reaction kinetic model the hyperbolic model and its two linear versions significantly improved the fitness to the experimental data. Among the three proposed primary models the performance of the hyperbolic and second linear version models was similar and better than that of the first linear version model. For the full model prediction the performance of the neural network model was better than that of the Davey model.