評估 COVID-19 疫情對於都市交通轉運功能的衝擊： 多運具移動網絡模型的階層分群分析

Abstract

至今為止，COVID-19 已對民眾的生活造成巨大的影響，在疫情期間人們可能會改變其旅行的習慣。了解城市內旅行者的行為是迫切的，掌握人流動向有利政府當局在疫情期間採取應對的管制措施。 本文利用了適用多層網絡模型的階層分群演算法，對疫情前與疫情當下時期的多運具智慧卡資料模型進行了分群，產生重疊的空間結構，其顯示人流往來變得稀疏、生活圈範圍擴大。另外，結合橋接性與可及範圍的概念，提出了與以往不同觀點的角度來衡量轉乘站的重要性和風險性。最後，藉土地利用資料推測旅行目的地，發現旅客降低前往商業場所的比例，而更往混合使用住宅移動等，顯示人們會避開人流群聚的場所，傾向待在家中，降低被傳染的風險。 本研究反映了城市內部的公共運輸人流對於疫情爆發的反應。成果能幫助大眾運輸系統的管理與運營，以及公共衛生方面的防疫策略訂定時的參考。

So far, COVID-19 has had a huge impact on people's lives, people may change their travel behaviors during epidemic time. Understanding the behavior of travelers in the city is urgent, which is beneficial for the government authorities to take corresponding control measures during the epidemic. In this paper, a hierarchical clustering algorithm for multi-layer network is used to analyze a pair of multimodal models consisted of smartcard data in the pre-epidemic and epidemic periods, resulting in overlapping spatial structures that indicate sparser footfalls and widening of living circles. In addition, the concepts of bridging and accessibility are introduced to provide a different perspective on the importance and risk of hubs. Finally, land use data were used to infer travel destinations, and it was found that travelers were less likely to go to commercial establishments and more likely to move to mixed-use residences, suggesting that people avoid crowded places and tend to stay at home to reduce the risk of being infected. This study reflects the response of intra-city public transportation flows to an outbreak. The results can be used as a reference for the government and operation of public transportation systems, for developing public health strategies.