电商集中促销期间物流峰值统计模型仿真

发布时间：2019-06-24 17:46

【摘要】：电商集中促销期间会产生物流峰值,物流峰值的变化受到消费区域、经济水平、人员结构等复杂因素的影响,峰值出现的时间和大小的估计过程存在众多因素干扰。传统的估计方法在进行电商集中促销期间物流峰值统计的过程中,受到诸多印象的影响,物流峰值估计的相关参数选择缺少准确约束,使得物流峰值统计模型的可信度降低。提出基于K均值粒子群滤波的电商集中促销期间物流峰值统计方法。根据K均值算法对全部类型的电商进行分类,获取不同类型的电商对于物流峰值的影响程度。根据粒子群滤波算法的原理,将不同类型的电商看作一个个独立的粒子进行训练。根据训练结果,对当前全部电商集中促销期间的促销行为对物流峰值的影响进行估计。实验结果表明,改进算法能够对电商集中促销期间物流峰值进行精确统计。
[Abstract]:The peak value of logistics will occur during the centralized promotion of e-commerce. The change of peak value of logistics is affected by the complex factors such as consumption area, economic level, personnel structure and so on. There are many factors interfering with the estimation process of the peak value. The traditional estimation method is affected by many impressions in the process of logistics peak value statistics during the centralized promotion of e-commerce. The selection of related parameters of logistics peak estimation is lack of accurate constraints, which reduces the credibility of the logistics peak statistical model. A statistical method of logistics peak value during centralized promotion of e-commerce based on K-means particle swarm optimization is proposed. According to the K-means algorithm, all types of e-commerce are classified, and the influence degree of different types of e-commerce on logistics peak value is obtained. According to the principle of particle swarm optimization algorithm, different types of e-quotient are trained as independent particles. According to the training results, the influence of the promotion behavior on the logistics peak value during the current centralized promotion of e-commerce is estimated. The experimental results show that the improved algorithm can accurately count the logistics peak value during the centralized promotion of e-commerce.
【作者单位】：南开大学经济学院;
【分类号】：F713.36;TP18

【参考文献】