基于改进灰色神经网络的2022年冬奥会食品冷链物流需求预测  

Forecast of Cold Chain Food Logistics Demand during 2022 Winter Olympics Based on Improved Grey Neural Network

作　　者：王健行 岳帅 兰洪杰[1] 戴梓轩 夏梦圆 Wang Jianxing1, Yue Shuai2, Lan Hongjie1, Dai Zixuan1, Xia Mengyuan1(1. School of Economics ＆ Management, Beijing Jiaotong University, Beijing 100044;2. Department of Automation, University of Science ＆ Technology of China, Hefei 230026, Chin)

机构地区：[1]北京交通大学经济管理学院,北京100044 [2]中国科学技术大学自动化系,安徽合肥230026

出　　处：《物流技术》2018年第7期62-68,共7页Logistics Technology

基　　金：国家自然科学基金(G71390334);北京市社科基金"北京冬奥会食品冷链物流安全管理研究"(16JDGLB012)

摘　　要：冬奥会食品冷链物流需求在实际运作中存在不确定性,为减轻其对冷链物流系统的影响,有必要对其进行预测。建立了一种基于改进混合算法优化的灰色神经网络预测模型,该模型考虑了冬奥会食品冷链物流需求主体贫数据、复杂性等特征,在选择传统灰色神经网络模型的基础上,结合遗传粒子群混合算法对其权值与阈值进行优化,并对粒子交叉与变异的概率进行自适应的改进。仿真结果表明,该模型预测精度优于常规灰色神经网络以及单一算法优化的灰色神经网络模型,证明了该预测模型的准确性。最终根据需求主体的预测数量,结合人均冷冻冷藏食品日消耗量、冷藏设备承载能力预测食品冷链物流需求总量,为2022年冬奥会食品冷链物流的实际运作提供参考。In this paper, a grey neural network forecasting model based on the improved hybrid optimization is established, which takes into account the characteristics of the cold chain food logistics demand during the 2022 Winter Olympics such as insufficient subject data and complexity, etc. Next, based on the traditional grey neural network model, the weight and threshold of the model are optimized with the genetic particle swarm optimization, and adaptively improved against the probability of particle crossover and mutation. The simulation shows that the forecasting accuracy of the model is superior to that of the conventional grey neural network model and the grey neural network model optimized by a single algorithm.

关 键 词：冬奥会 食品冷链物流 改进混合算法 灰色神经网络 需求预测 

分 类 号：F252[经济管理—国民经济] F224