Abstarct: Mankind is currently trying to identify natural disasters and take the necessary measures to reduce the damage. One of these natural disasters that is very important for human beings is earthquakes. Knowing more about the earthquake and taking the necessary measures to deal with it will minimize the loss of life and property. One of the measures taken to predict an earthquake is to prepare an earthquake risk zoning map.Today, the hazard zoning map, which determines the amount of risk from very low to very high for the target area, is baxsed on GIS remote sensing software. Accepts. For this purpose, laxyers; the distance from the fault, the slope of the area, the direction of the slope and the population are considered to be six and GIS, ENVI and Matlab software have been used. In this research, supervised classification methods have been used, which, according to experts, weigh the laxyers. Finally, by combining laxyers and expert opinions and using supervised classification methods, a zoning map is obtained. In this research of methods; Maximum probability (ML), support vector machine (SVM) and neural networks (NN) are used with 63%, 76% and 77% accuracy, respectively. Finally, with the help of the theory of data integration by voting method (MV) which is done in MATLAB software, the final zoning map can be achieved, which has a high accuracy. The accuracy of the final zoning map is 90%, which is higher than the accuracy of any of the monitored classification methods, which indicates an increase in the accuracy of the zoning map. And the accuracy of the set of methods can be increased with the help of the integration method. According to this zoning map, the metropolis of Tehran can be classified from very low risk to very high risk.