Stochastic Dynamic Game-Theoretic Models of Urban Freight Systems

城市货运系统的随机动态博弈论模型

• 批准号: 0900040
• 负责人: Terry Friesz
• 金额: $ 35万
• 依托单位: Pennsylvania State Univ University Park
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2009
• 资助国家: 美国
• 起止时间: 2009-09-01 至 2015-05-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=0900040&HistoricalAwards=false
• 关键词: Stochastic; Dynamic; Game; Theoretic; Models

The efficient and timely delivery of urban freight is a key factor in determining the economic health and sustainability of urban centers and their surrounds. Urban freight consists of both large shipments needed to replenish retail inventories and small parcels that fulfill electronic and telephonic commerce initiated by individual consumers. This study will develop new urban freight transportation planning and pricing tools that recognize the intrinsic uncertainty of freight demands, costs and delivery times. A family of mathematical models in the form of stochastic dynamic games, and their robust counterparts, that exploit recent advances in revenue management and nonlinear pricing theory will be created to describe a variety of urban settings. Fixed point algorithms, descent in Hilbert space and agent based simulation will be employed to solve these models for hypothetical data and, thereby, develop rules-of-thumb to guide urban planners and freight company operators in configuring urban freight networks and pricing urban freight services.This research will provide models and algorithms that can be integrated with evolving information technology and decision support systems for influencing urban and suburban traffic congestion. Urban freight planning that directly considers the risk to shippers of untimely deliveries, the risks to carriers of penalties for violating quality of service guaranties, and the risks to private vehicles of excessive congestion will be possible, for the first time. Moreover, the results of this research will allow planners to quantify the advantages of restricting truck-based deliveries to particular times of day or particular routes, including the possible assignment of a unique time and route to each specific truck. Additionally, time-of-day tolls and auction-based strategies that charge delivery vehicles for access to congested areas will be facilitated by tools developed in this study.

城市货运的高效和及时交付是决定城市中心及其周边地区经济健康和可持续性的关键因素。城市货运包括补充零售库存所需的大型货物和满足个人消费者发起的电子和电话商务的小型包裹。这项研究将开发新的城市货运规划和定价工具，认识到货运需求，成本和交货时间的内在不确定性。一个家庭的数学模型的形式随机动态游戏，和他们的强大的对手，利用收入管理和非线性定价理论的最新进展将创建描述各种城市环境。固定点算法，希尔伯特空间的下降和基于代理的模拟将被用来解决这些模型的假设数据，从而，制定规则，大拇指指导城市规划者和货运公司运营商在配置城市货运网络和定价城市货运服务。这项研究将提供模型和算法，可以与不断发展的信息技术和决策支持系统，以影响城市和郊区交通拥挤。城市货运规划将首次直接考虑发货人不及时交货的风险、承运人违反服务质量规定的处罚风险以及私人车辆过度拥堵的风险。此外，这项研究的结果将使规划人员能够量化将卡车交付限制在一天中特定时间或特定路线的优势，包括可能为每辆特定卡车分配独特的时间和路线。此外，本研究中开发的工具将促进对进入拥堵地区的送货车辆收取当日通行费和基于拍卖的策略。