Influenced by many factors, the characteristics of airport energy consumption are stochastic, nonlinear and dynamic. In order to predict the airport energy consumption and its trend, an unbiased grey markov prediction model was proposed. To weaken the random fluctuations of original energy consumption data sequence, accelerate its translation transformation and geometric mean transformation firstly. The proposed model makes use of the advantages of unbiased GM (1,1) model and markov prediction model. Using the measured energy consumption data from five airports, we analyzed and compared the prediction results of the proposed prediction model with that of traditional GM (1,1) model and unbiased GM (1,1) model. The comparison result shows that unbiased grey markov prediction model has a better accurate prediction.