Unravel the impact of COVID-19 on the spatio-temporal mobility patterns of microtransit
Unravel the impact of COVID-19 on the spatio-temporal mobility patterns of microtransit
复制标题
DOI：
10.1016/j.jtrangeo.2021.103226
复制
发表时间：
2021-12
影响因子：
6.1
通讯作者：
Yirong Zhou;X. Liu;T. Grubesic
中科院分区：
文献类型：
--
作者：
Yirong Zhou;X. Liu;T. Grubesic
共享出行是更大的共享经济的重要组成部分。网约车、共享单车、电动滑板车和其他类型的共享出行在全球范围内持续增长。这些服务包括微型运输，这是一种新的运输模式，可扩大区域内的运输覆盖面。移动的设备使微交通服务，聚集乘客和使用实时路由算法组客户在类似的方向旅行。与此同时，新出现的新型冠状病毒COVID-19从根本上改变了包括微型交通在内的所有交通服务的乘客行为。虽然现有研究评估了疫情前微型交通试点项目的表现，但没有关于新型冠状病毒影响下微型交通活动时空模式的信息。本文的目的是应用特征分解和k-团渗流方法来揭示微交通出行的时空模式。此外，我们使用这些方法来确定潜在的社区使用的数据从湖城，犹他州的试点计划。由此产生的研究提供了对COVID-19如何改变旅行行为的深入了解。具体而言，特征分解划定了跨时间维度的旅游模式的同质性和异质性。我们发现，第一英里/最后一英里行程是COVID前后期间差异的主要来源，尽管受到COVID-19的威胁，依赖交通工具的用户仍缺乏弹性。k-clique渗滤方法检测出可能的群落形成，并追踪这些群落在大流行期间如何演变。此外，我们系统地分析了重叠社区和共享节点周围的网络结构，通过使用聚类系数。在这项研究中广泛开发的工作流程是普遍的和有价值的理解独特的时空模式的微交通。该框架还可以帮助运输机构进行绩效评估，区域运输战略和最佳车辆调度。
Shared mobility is an essential component of the larger sharing economy. Ride-hailing, bike-sharing, e-scooters, and other types of shared mobility continue to grow worldwide. Among these services is microtransit, a new transport mode that extends transit coverage within a region. Mobile devices enable microtransit services, aggregating riders and using real-time routing algorithms to group customers traveling in similar directions. Meanwhile, the newly emerged coronavirus, COVID-19, has radically reshaped the ridership behavior of all transit services, including microtransit. While existing research evaluates the performance of microtransit pilot programs before the pandemic, there is no information concerning the spatio-temporal pattern of microtransit activities under the impact of COVID-19. The purpose of this paper is to apply eigendecomposition andk-clique percolation methods to uncover the spatio-temporal patterns of microtransit trips. Further, we used these approaches to identify underlying communities using data from a pilot program in Salt Lake City, Utah. The resulting research offers insight into how COVID-19 altered travel behavior. Specifically, eigendecomposition delineated the homogeneity and heterogeneity of travel patterns across temporal dimensions. We identified first mile/last mile trips as a major source of variance in both pre- and post-COVID periods and that transit-dependent users prove to be inelastic despite the threat of COVID-19. The k-clique percolation method detected possible community formations and tracked how these communities evolved during the pandemic. In addition, we systematically analyzed overlapping communities and the network structure around shared nodes by using a clustering coefficient. The workflow developed in this research broadly is generalizable and valuable for understanding the unique spatio-temporal patterns of microtransit. The framework can also help transit agencies with performance evaluation, regional transport strategies, and optimal vehicle dispatching.