基于人工神经网络的物流中心选址决策模型研究

物流中心选址在整个物流系统规划中占有十分重要的位置,需要决策者综合考虑多个因素。本文根据物流环境、生产能力和经济效益三方面组成的10个评价指标和单因素评价标准,应用BP神经网络方法对某市物流中心选址进行决策研究。根据单因素评价标准生成足够多用于神经网络建模用的训练样本、检验样本和测试样本,用检验样本实时监控训练过程以避免出现“过训练”现象,建立可靠的BP模型。示例计算结果表明：与多级模糊综合评价方法相比,本文建立的BP神经网络决策模型具有定量分析和比较精细、客观的特点,在物流中心选址决策研究中具有较好的应用价值。

Decision Model on Site Selection of Logistics Center Applying Artificial Neural Networks

Distribution center's site selection,influenced by various indictors,is crucial to logistics planning.According to the evaluation standard of logistics environment,production capacity and economic efficiency,a decision model on site selection of logistic center is established applying artificial neural networks.According to the critical standard of every indicator for various degrees,efficient training set data,verification set data and testing set data for establishing robust NN-based model for site selection of logistics center.As a case study,the decision results show that,compared with multi-class fuzzy comprehensive evaluation,BP neural network-based model possesses the characteristics such as quantity analysis and accuracy etc.,and be suitable for decision site selection of logistics center.