基于BP神经网络的货运量预测方法

Forecasting Method of Fright Volume Based on BP Networks

总结了常用的几种定量预测方法,并指出其在实际应用中的不足。而人工神经网络自身具有鲁棒性、容错性、有表示任意非线性关系的能力和学习能力等特性,为预测技术提供了一种新的思想和方法。最后把这种预测方法应用于物流预测取得了满意的结果。

The paper summarizes some common forecasting methods, points out their shortcomings and applies Artificial Neural Network in logistics transportation forecast because of its strong robustness, fault-tolerance quality, showing random nonlinear relations and learning.