The hotel industry has grown rapidly over the past decade, placing higher demands on managers to outperform their competitors in the industry. Planning and allocating sufficient resources to hotel operations, investments, finance and marketing requires accurate forecasts. This study proposes an approach to predict international hotel revenue in Taiwan. First, the SARIMA model is used to predict the number of occupied rooms and average room price. Next, genetic algorithm technology is used to establish an artificial intelligence neural network model for hotel revenue prediction. The hospitality industry can benefit from the suggested approach.