Multi-scale Modeling of Crowdshipping as a New Form of Urban Delivery

众包作为城市配送新形式的多尺度建模

• 批准号: 1663411
• 负责人: Bo Zou
• 金额: $ 34.93万
• 依托单位: University of Illinois at Chicago
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2017
• 资助国家: 美国
• 起止时间: 2017-08-15 至 2021-07-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1663411&HistoricalAwards=false
• 关键词: Multi; scale; Modeling; Crowdshipping; New

E-commerce in the US has experienced explosive growth in the past decade, resulting in a precipitous increase in truck traffic. The surge of truck traffic has caused many negative consequences to the urban environment such as greater traffic congestion and emissions, faster wear-and-tear on road infrastructure, and more acute shortage of truck parking space. These consequences are increasingly at odds with the need to develop livable urban communities which requires significant reduction of truck traffic. Crowdshipping has emerged recently as an alternative to truck-based delivery. Using a crowd of ordinary individuals who walk, bike, or drive to perform delivery in urban areas, crowdshipping presents considerable promise to reconcile the strong need for livable community development with the rapid growth of urban freight demand. This award develops theoretic foundations for modeling and improving the efficiency of crowdshipping for urban delivery. Results from the research will lead to a better understanding of the traffic, economic, and environmental impact of crowdshipping and support urban freight policy making. Knowledge gained from this project will also provide generic insights for other areas in the increasingly crowdsourced and on-demand society, such as ride sourcing and mobile sensing. By designing, delivering, and evaluating a pipeline of engineering education activities, this award will further enhance the awareness and understanding of freight transportation and urban delivery innovations among diverse student groups. The research activities are organized around two thrusts. The first thrust establishes static and dynamic mechanisms to address individual solicitation and shipment assignment problems at the microscale. The properties of the mechanisms in incentivizing individual participation in crowdsourced delivery and the asymptotics of an approximation technique in shipment assignment will be investigated. The second thrust creates a queueing network-based framework to characterize and improve system performance of crowdshipping at the macroscale. Conditions for crowdshipping system equilibrium existence and equilibrium flows of crowdsourced individuals will be identified. Analytical capabilities will be developed to determine the optimal pool size of crowdsourced individuals to meet given shipping demand and improve delivery service fairness. The developed mechanisms, techniques, and models will be evaluated using case studies informed by real world data. If successful, this research will advance the knowledge and analysis capabilities of crowdshipping for urban delivery, and enrich the literature of transportation systems analysis and city logistics. Additionally, the PI will engage diverse student groups (high school and community college students) to raise awareness in understanding of freight transportation and urban delivery innovations among diverse student groups.

在过去的十年里，美国的电子商务经历了爆炸式的增长，导致卡车运输量急剧增加。卡车交通的激增对城市环境造成了许多负面影响，如更严重的交通拥堵和排放，道路基础设施的加速磨损，卡车停车位的严重短缺。这些后果与发展需要大量减少卡车交通的宜居城市社区的需要越来越不一致。最近，众筹成为卡车配送的替代方案。利用一群步行、骑自行车或开车的普通人在城市地区进行递送，众包运输在调和对宜居社区发展的强烈需求与城市货运需求的快速增长方面表现出了相当大的前景。该奖项为建模和提高城市配送的众筹效率奠定了理论基础。研究结果将有助于更好地理解众包运输对交通、经济和环境的影响，并为城市货运政策的制定提供支持。从该项目中获得的知识也将为日益增长的众包和按需社会的其他领域提供通用见解，例如乘车资源和移动传感。通过设计、实施和评估一系列工程教育活动，该奖项将进一步提高不同学生群体对货运和城市运输创新的认识和理解。研究活动围绕两个重点组织。第一个重点是建立静态和动态机制，以解决微观尺度上的个别征求和装运分配问题。将研究激励个人参与众包交付的机制的性质，以及运输分配中近似技术的渐近性。第二部分创建了一个基于排队网络的框架，以在宏观尺度上表征和提高众筹系统的性能。确定众包系统均衡存在的条件和众包个体均衡流动的条件。将开发分析能力，以确定众包个人的最佳池大小，以满足给定的运输需求，并提高交付服务的公平性。开发的机制，技术，和模型将评估使用案例研究告知真实世界的数据。如果本研究取得成功，将提升众筹对城市配送的认识和分析能力，丰富交通系统分析和城市物流的文献。此外，PI还将动员不同的学生群体（高中和社区大学学生），提高不同学生群体对货运和城市运输创新的认识。

项目成果

Vehicle ownership models for a sharing economy with autonomous vehicle considerations

考虑自动驾驶汽车的共享经济车辆所有权模型

• DOI: 10.1080/19427867.2021.2007681
• 发表时间: 2021
• 期刊: Transportation letters
• 作者: Stinson, M.

Deep Reinforcement Learning for Crowdsourced Urban Delivery

• DOI: 10.1016/j.trb.2021.08.015
• 发表时间: 2021-09-13
• 期刊: TRANSPORTATION RESEARCH PART B-METHODOLOGICAL
• 影响因子: 6.8
• 作者: Ahamed, Tanvir;Zou, Bo;Tulabandhula, Theja
• 通讯作者: Tulabandhula, Theja