In this study, we report on the rubber compounds in a model passenger tire selected for modeling their cure curves at different temperatures using Adaptive Neuro-Fuzzy Inference Systems (ANFIS). Optimum cure time of the rubber compounds are predicted using ANFIS model. Equivalent cure concept (ECC), that is traditionally used in rubber and tire industries and Artificial Neural Networks (ANN) were also used to predict optimum cure time of the same rubber compounds, in our previous study. The results of three techniques, i.e. ANFIS, ANN and ECC were compared in view of prediction error criteria. The effects of fuzzy membership functions and the number of rules were investigated for ANFIS model. The best ANFIS architecture that provides minimum percentage error was determined for a selected compound. Then, the best ANFIS architecture was also used for predicting optimum cure time of the other 10 rubber compounds. The performances of ANFIS, ANN and ECC were compared by means of prediction errors. For overall evaluation of the compounds, while average percentage error calculated by ANFIS was found as 3.89% that of was found as 4.23% for ANN model and as 7.15% for ECC. These results show that ANFIS could be used as a more powerful technique than traditional ECC and ANN to predict optimum cure time of rubber compounds. (C) 2011 Elsevier Ltd. All rights reserved.