摘要
研究物流需求预测准确度问题。物流需求预测中存在数据小以及非线性特点,使预测系统存在不确定性。为解决上述问题,提出了一种泊松分布的神经网络需求预测算法,采用泊松分布算法对物流的整体需求进行分类,然后采用灰色理论算法选择物流需求影响因子,对物流的需求进行实时预测,仿真结果表明,改进物流需求预测方法比传统的灰色理论预测模型以及BP神经网络具有更高的预测精确度,有效地提高了区域物流需求的预测准确度,具有一定的实际应用价值。
This article mainly aims at the requirement forecasting of logistics data and nonlinear characteristic.Aiming at this problem,this paper put forward a new Poisson distribution based neural network forecasting algorithm.The method first employed a Poisson distribution algorithm for the logistics of the overall demand for classification,and then used the gray theory algorithm for the selection of logistics demand influence factor to predict the demand for logistics in real-tim.Simulation and experimental results show that the logistics demand forecast method is better than the traditional gray theory,and BP neural network prediction model has higher prediction accuracy,effectively improves the regional logistics demand forecast accuracy,and has a certain practical value.
出处
《计算机仿真》
CSCD
北大核心
2012年第4期229-233,共5页
Computer Simulation
基金
河南省教育厅科学技术研究重点项目(12A520029)
关键词
泊松分布
区域物流
灰色理论
神经网络
预测精度
Poisson distribution
Regional logistics
Grey theory
Neural network
Forecast precision
