Abstract: Based on the resort life cycle theories, 6 prediction models of linear, exponential, GEPTTs, logistic regression, GM (1, 1) and ARIMA (1, 1, 1) was constructed with the statistics of tourists in Potatso National Park from 2007 to 2015. The results indicated that GM (1, 1) model has the best prediction effect, the overall effect were as the follows: GM (1, 1)> logistic regression > linear > exponential > ARIMA (1, 1, 1)> GEPTTs. The simulation equation based on GM (1, 1) grey forecasting model was x(t + 1)= 49.157 119e^{0.159 615t}± 27.302 081. Visitors in Pudacuo National Park may exceed 5 million people in 2025 based on the prediction of tourist number in the next 10 years.