---

Dust is a phenomenon that has many environmental effects in various parts of human life, including: agriculture, economy, health and so on. The purpose of this study is to investigate and predict the dust phenomenon in Kermanshah. Meteorological data with a resolution of 3 hours in the statistical period (2020-2000) of Kermanshah station was obtained from the Meteorological Organization. First, the dust data were normalized and then using ANN neural network models to predict dust concentration and ANFIS adaptive neural network to debug and predict the time series of dust occurrence in MATLAB software were debugged and predicted. Findings showed that the maximum predicted dust concentration related to the minimum fenugreek point with the highest Pearson correlation with dust was estimated to be 3451.23 μg / m3. Also, the results of time series prediction using ANFIS model showed that the linear bell membership function with grade 3, in the training and testing stages, has the most desirable input function among other membership functions. According to the forecasting models, the highest probability of maximum dust occurrence in the next 20 years in Kermanshah was 94%.