Global air transport complex network: multi-scale analysis
Global air transport complex network: multi-scale analysis
复制标题
DOI：
10.1007/s42452-019-0702-2
复制
发表时间：
2019-07-01
影响因子：
2.6
通讯作者：
Krupp, Armin
中科院分区：
文献类型：
--
作者：
Guo, Weisi;Toader, Bogdan;Krupp, Armin
每年世界上几乎一半的人口由航空公司运输，从经济和科学角度理解这种运输方式非常重要。近年来，数据可用性的不断增加导致了复杂的网络和代理交互模型，这些模型试图更好地了解航空运输网络并进行预测。在本案例研究论文中，我们回顾了两种关键方法的现有研究，即：（1）自上而下的多尺度网络科学方法，以及（2）自下而上的熵最大化交互网络方法。使用简单的社会经济指标，我们能够构建一个非常准确的交互模型，可以预测交通量，并且该模型可以预测人口增长或燃料成本的影响。使用网络科学方法，我们能够识别社区结构并将其与经济产出联系起来。我们还看到了中心如何随着时间的推移而演变并变得更具影响力。展望未来，使用随机图论，似乎飞行成本的降低将导致枢纽影响力的增加。本案例研究论文中传播的知识将为学术界和行业从业者提供共同探索有趣的研究领域的前进步骤。
Almost half of the world's population is carried by airlines each year, and understanding this mode of transport is important from economic and scientific perspectives. In recent years, the increasing availability of data has led to complex network and agent interaction models which attempt to gain better understanding of the air transport network and develop forecasts. In this case study paper, we review existing research on two key approaches, namely: (1) a top-down multi-scale network science approach, and (2) a bottom-up entropy-maximization interaction network approach. Using simple socioeconomic indicators, we were able to construct a very accurate interaction model that can predict traffic volume, and the model can forward estimate the impact of population growth or fuel cost. Using network science approaches, we were able to identify community structures and relate them to economic outputs. We also saw how hubs evolved over time to become more influential. Looking into the future, using random graph theory, it seems that reduced flight cost will lead to increased hub influence. The disseminated knowledge in this case study paper will provide both academics and industry practitioners with steps forward to co-explore the interesting research landscape.