区域物流需求建模与预测仿真

摘要

研究区域物流需求预测优化问题,区域物流需求与经济结构和资源分布相关,因此存在较强的非线性,属于一种小样本、非线性数据结构.传统线性、大样本预测方法无法进行准确预测,预测精度比较低.为提高了区域物流需求预测精度,提出一种支持向量机物流需求预测方法.首先采用多元回归分析法选择区域物流需求的影响因子,然后将输入样本输入到支持向量机学习,并通过蚁群法对支持向量机参数进行优化,最后建立区域物流需求与影响因子之间复杂的非线性关系模型.采用上海市1978 -2003年物流需求量对模型性能进行测试,结果表明,相对于多元线性回归、BP神经网络模型,支持向量机提高了区域物流需求的预测精度,在区域需要预测中具有广泛的应用前景.%Study the optimization problems of regional logistics demand forecast. The paper put forward a logistics demand forecasting methods based on support vector machine. The multiple regression analysis method was used to select regional logistics demand influencing factors, and ant colony algorithm was used for the optimization of support vector machine parameters. The complex nonlinear relation between regional logistics demands and influence factors was modeled . The model performance was tested with the logistics demands of Shanghai from 1978 to 2003 years. The results show that, compared with the multiple linear regression method and BP neural network model, the support vector machine can improve the forecasting accuracy of regional logistics demand, and has wide application prospect.