Dynamic Game Theoretic Models for Urban Freight Systems

项目概要：城市货物运输的动态博弈理论模型在运输规划和管理过程中明确考虑货物运输是重要的，原因有很多。从积极的方面来看，一个高效率的货运系统是经济竞争力的必要条件。在消极方面，有大量证据表明，货运活动产生了严重的负外部性，对弱势群体造成了不成比例的影响。经济全球化、用户期望和社区关切的综合压力加剧了这些负面外部性。因此，世纪的货运系统必须覆盖更大的地理区域，对用户的需求和期望做出响应，减少产生的外部效应，并且在提供额外的货运基础设施变得更加困难和昂贵的情况下做到这一切。简而言之，城市货运系统将不得不少花钱多办事。这一巨大的挑战给必须平衡多个利益相关者相互冲突的目标的机构增加了压力。货运活动的复杂性和缺乏适当的货运建模方法只会加剧挑战。以前的城市货运系统建模工作没有考虑到信息技术的全部影响。一个例子是，目前的传闻证据表明，互联网和由互联网促进的电子商务增加了卡车旅行的规模，改变了其性质，社会成本的变化尚未量化。这些净效应不能在现有的理论中解决，这是拟议的研究的主要动机，以创建一个新的家庭的城市货运模型。项目团队设想的新模型将直接处理互联网和电子商务的影响，是明确动态的，并考虑到活跃在城市货运网络上的众多代理人的博弈行为。该提案描述了一项研究计划，可能导致第一个全面的城市货物流动理论。主要研究目标可以描述为城市商品网络的价格、消费水平、出行生成、容量扩展、路线、时刻表和设计的战略和通用战术建模。这将通过创建一系列模型来实现。最详细和一般的模型，这个家庭将动态随机分层网络游戏，其中的货运权的方式与客运交通共享。该项目团队希望展示如何将适合日内时间尺度的城市商品流动和决策与涉及日常，周与周以及年与年时间尺度的网络博弈模型相结合。这将：（a）对商品和商用车辆的流动情况作出更真实的描述，其中考虑到以往不可能做到的运输链和空载运输;（B）在不平衡阶段增加货运供求平衡过程的真实性，从而加强运输模型;（c）提高承运人特有的交通管理效率;以及（d）为未来的研究奠定基础，通过在本研究中开发的动态博弈论模型中嵌入拍卖，协调特定运营商的分布式交通规划，以减少网络范围内的拥塞。

PROJECT SUMMARY: DYNAMIC GAME THEORETIC MODELS FOR URBAN FREIGHT SYSTEMSThe explicit consideration of freight transportation in the transportation planning and management processes is important for many reasons. On the positive side, an efficient freight transportation system is a sine equa non condition for economic competitiveness. On the negative side, there is a wealth of evidence indicating freight activity produces significant negative externalities that disproportionately affect disadvantaged communities. These negativity externalities are compounded by the combined pressures of economic globalization, user expectations and community concerns. As a consequence, the freight transportation systems of the 21st Century will have to cover a larger geographic area, be responsive to user needs and expectations, reduce the externalities produced, and do all of this in a context in which providing additional freight infrastructure will become more difficult and expensive. Simply put, the urban freight transportation system will have to do more with less. This immense challenge adds pressure to agencies that have to balance the conflicting objectives of multiple stakeholders. The complexity of freight movements and the lack of appropriate freight modeling methodologies only serve to heighten the challenge.Prior efforts to model urban freight systems have not considered the full impacts of information technologies. An example is current anecdotal evidence suggesting that the Internet and the e-commerce facilitated by it have increased the magnitude of truck-trips and changed their nature, with as yet un-quantified changes in social costs. That those net effects cannot be addressed within existing theories is a principal motivation for the proposed research to create a new family of urban freight models. The new models envisioned by the project team will directly treat the influence of the Internet and e-commerce, are explicitly dynamic, and account for the gaming behaviors of the multitudinous agents active on an urban freight network. This proposal describes a research plan that may lead to the first comprehensive theory of urban goods movements. The main research objective can be described as the strategic and general tactical modeling of prices, consumption levels, trip generation, capacity expansion, routes, schedules and designs for urban goods networks. This will be accomplished by creating a family of models. The most detailed and general models of this family will be dynamic stochastic hierarchical network games wherein the freight rights-of-way are shared with passenger traffic. The project team expects to show how urban goods movements and decisions appropriate to a within-day time scale can be integrated with network game-theoretic models involving day-to-day, week-to-week and year-to-year time scales. This will: (a) provide a more realistic depiction of the flow of both commodities and commercial vehicles, incorporating considerations of trip chains and empty trips than have not been previously possible; (b) enhance transportation modeling by adding realism to the freight supply-demand equilibration process during its disequilibrium phase; (c) improve the efficiency of carrier-specific traffic management; and (d) set the stage for a future research to decrease network-wide congestion by co-ordination of carrier-specific distributed traffic planning by embedding auctions in the dynamic game theoretic models developed in this study.