航空货运量预测是航空公司运力安排的重要参考。本文根据分形拼贴定理,由分形插值方法求取一个吸引子与航空货运量历史数据相近的迭代函数系统,建立分形预测模型,对航空货运量进行预测。预测结果表明,该方法不存在收敛性问题,数据收集简单,具有较好的实用性。货运量时间序列R/S分析进一步表明了预测值的合理性。 Air cargo forecasting is an important reference for airlines’ capacity allocation. According to the fractal collage theorem, a fractal interpolation method is used to find an iterative function system with attractor and historical data of air cargo. The fractal prediction model is established to predict the air cargo volume. The prediction results show that this method does not have the problem of convergence, the data collection is simple and has good practicability. Freight time series R / S analysis further shows the rationality of the forecast value.