【摘要】 航空物流作为高效、快速的物流方式，在推进长三角一体化发展的过程中发挥着重要作用。随着物流需求的增加，科学、准确地预测未来的航空物流需求量成为地区经济规划和物流战略制定的关键问题。针对当前长三角一体化航空物流的发展现状，基于现有的物流需求预测方法，综合考虑数据的线性与非线性特征，构建了差分整合移动自回归-长短时记忆网络组合模型。以长三角地区2000—2023年的航空货邮吞吐量为数据来源，分别使用差分整合移动自回归模型、长短时记忆网络以及两者的组合模型进行了预测，并对比分析了它们的预测效果。结果表明，该组合模型相较于传统的单一模型具有更高的预测精度，能够有效捕捉数据的复杂特征和趋势变化，对物流需求预测具有更好的效果，并以此模型预测了未来五年长三角航空物流的需求量情况。更多还原

【Abstract】 As an efficient and rapid logistics approach, air logistics assumes a significant role in facilitating the integrated development of the Yangtze River Delta.With the escalation of logistics demands and the imbalance in development, scientifically and accurately predicting future air logistics demands has emerged as a crucial issue in regional economic planning and logistics strategy formulation.Considering the current development status of integrated air logistics in the Yangtze River Delta and the existing logistics demand forecasting methods, this paper comprehensively takes into account the linear and nonlinear characteristics of the data and constructs a autoregressive integrated moving average-long short term memory model.The air cargo throughput in the Yangtze River Delta region from 2000 to 2023 is adopted as the data source.The autoregressive integrated moving average model(ARIMA),the long short-term memory(LSTM),and the combination of the two models were employed for prediction, and their prediction effects were compared and analyzed.The results demonstrate that, in contrast to the traditional single model, the combined model possesses higher prediction accuracy, can effectively capture the complex characteristics and trend variations of the data, and yields a better effect on the logistics demand forecast.This model predicts the demand of air logistics in the Yangtze River Delta for the next five years.更多还原