Based on the extreme value data of the maximum wind speed when typhoon makes landfall over South China from 1995 to 2017, a prediction model of extreme values based on fuzzy time series is constructed, and the model is compared with the traditional time series ARIMA model. The prediction results show that the mean absolute error, mean relative error, and root mean square error of the fuzzy time series are 2.621 m·s-1, 0.066, and 2.727 m·s-1, respectively. The accuracy of the model is obviously higher than that of the traditional time series ARIMA model. At the same time, it shows that the application of fuzzy time series to the prediction of extreme values of the maximum wind speed when typhoon makes landfall can obtain better prediction results.