Abstract: In view of the features of wind speed forecasting and the innate uncertainty of judgments, a combined model based on the fuzzy Analytic Hierarchy Process (AHP) was presented in the paper to improve the wind speed forecasting accuracy. The criteria proposed in this paper are not only the wind speed forecast cycle but also the fluctuation of the wind speed and the reliability of the forecast models. The combined model adopts triangular fuzzy number to indicate expert judgment information according to its fuzziness, and uses AHP to deal with optimizing weight of each scheme as to expert judgment results. In the combined model, multiple uncertain factors that influence wind speed forecasting are taken into consideration, and expert experience is introduced based on different model forecasting results. Example results show that the combined model can effectively improve the accuracy of wind speed forecasting.