The objective of this study was to model the extent of improvement in the degradability of phenol dyes by SnO 2 /Fe 3 O 4 nanoparticles by using photo catalytic reactor.The effect of operative parameters including catalyst concentration, initial dye concentration, stirring intensity, and UV radiation intensity on the photocatalytic batch reactor during removal of phenol red was investigated.Fractional factorial design (FFD) and response surface methodology (RSM) were used to design the experiment layout.The SnO 2 /Fe 3 O 4 nanoparticles were synthesized by core-shell method. The results of XRD and TEM showed the successful synthesis of these nanoparticles. The ability of back propagation neural network (BPNN), radial basis function neural network (RBFNN), and adaptivenetwork-based fuzzy inference system (ANFIS) in predicting the performance of photo catalytic reactor was also investigated. It is found that BPNN has better ability in predicting the dye removal (%) than RBFNN, ANFIS, and regression model. The best architecture of BPNN was a network consisted of 3 hidden layers with 20-30-20 neurons.The BPNN predicted the output values with a high determination coefficient (R 2 ) value Downloaded by [University of West Florida] at 22:23 06 October 2014 ACCEPTED MANUSCRIPT ACCEPTED MANUSCRIPT 2 of 0.0.9718, while the predicting R 2 of regression model was 0.8978. And the predicting R 2 of RBFNN and ANFIS were 0.8011 and 0.9023, respectively.