摘要
高精度的船舶物流配送成本预测可以节省成本,提高船舶物流配送企业的利润,因此具有重要的研究意义。为了获得更高精度的船舶物流配送成本预测结果,提出神经网络的船舶物流配送成本预测方法。首先采集船舶物流配送成本历史数据,通过变换技术建立船舶物流配送成本预测的学习样本,然后引入BP神经网络对船舶物流配送成本学习样本进行训练,拟合船舶物流配送成本的变化特点,从而实现船舶物流配送成本预测,最后与当前经典船舶物流配送成本预测方法进行优越性测试。结果表明,BP神经网络的船舶物流配送成本不仅预测精度平均高于对比方法 5%以上,而且船舶物流配送成本预测稳定性更优,预测结果可以为船舶物流配送企业提供有用的信息。
high precision shipping logistics distribution cost prediction can save costs and improve the profits of shipping logistics distribution enterprises,so it has important research significance.In order to get more accurate prediction results of ship logistics distribution cost,a neural network method of ship logistics distribution cost prediction is proposed.Firstly,the historical data of ship logistics distribution cost is collected,and the learning samples of ship logistics distribution cost prediction are established by transformation technology.Then,BP neural network is introduced to train the learning samples of ship logistics distribution cost to fit the change characteristics of ship logistics distribution cost,so as to realize the prediction of ship logistics distribution cost.Finally,it is compared with the current classic ship logistics distribution cost prediction Methods to test the superiority of BP neural network,the accuracy of ship logistics distribution cost prediction is more than 5%on average,and the stability of ship logistics distribution cost prediction is better,the prediction results can provide useful information for ship logistics distribution enterprises.
作者
冯彦乔
FENG Yan-qiao(Sichuan Vocational and Technical College of Communications,Chengdu 611330,China)
出处
《舰船科学技术》
北大核心
2020年第8期193-195,共3页
Ship Science and Technology
基金
四川省交通运输厅支撑计划资助项目(2017-C-07)
关键词
船舶物流配送系统
成本预测
神经网络
学习样本重构
ship logistics distribution system
cost prediction
neural network
learning sample reconstruction
