Methods and Tools to Strengthen Transit System Resilience

根据世界各地大多数乘客满意度调查的结果，过境乘客最不喜欢的就是长时间的延误和过境车辆的过度拥挤。许多因素都可能造成过境延误，从运营环境的小扰动到紧急情况或系统故障造成的重大事故。在多伦多，仅地铁系统每年就发生12000多起事故，其中约1000起被认为是“重大”事故。过境延误往往加剧过度拥挤，反之亦然。用户和运营商的成本可能是巨大的。最近的一项研究估计，纽约地铁事故造成的用户延误的经济成本每年高达3.89亿美元。该研究计划的目的是直接解决大型多式联运系统中的延误和过度拥挤问题。它采用了一种广泛的、多方面的方法，将为需求管理、供应管理和流量管理提供解决方案，所有这些都旨在一贯和协调地减轻交通延误和过度拥挤，特别是在地铁系统中。为了管理需求，我们的研究将开发创新的激励计划，为那些从过度拥挤的地点和时间转换的过境乘客提供奖励积分。基于激励和忠诚度的计划已经有效地应用于各种其他服务和行业，虽然它们正在慢慢引入公共交通，但它们需要支持经验工具来最大限度地发挥其有效性。这些工具将在本研究中开发。为了管理供应，特别是在地铁服务发生重大中断的情况下，我们将开发一个模型和工具系统，帮助交通机构部署有效的穿梭巴士服务，以填补地铁服务的差距。更具体地说，我们将开发分析工具，以实时预测任何给定地铁事故的预期持续时间，并确定最佳公交车数量以及这些公交车应该从哪些路线出发，以便将这些公交车调度到中断的地铁路段，从而最大限度地减少总延误。本研究的第三个主要轨道是关于管理进入地铁线路所有车站的乘客流量，以避免由于过度拥挤而造成的容量损失。这将通过创造性地安排与地铁线路连接的公共汽车和有轨电车支线来实现，沿着采用实时控制策略，以避免任何给定车站的过度饱和。拟议的解决方案将利用关于过境业务和需求的丰富多样的数据来制定，这些数据现在已经非常丰富。建立在健全的理论基础上，拟议的模型将使用严格和新的建模方法建立。拟议的研究可能会对过境复原力政策和技术产生变革性影响。

Transit riders dislike nothing in their journeys more than long delays and overcrowding of transit vehicles, according to the results of most rider satisfaction surveys around the world. Many factors can cause transit delays, ranging from minor perturbations in the operating environment all the way to major incidents due to emergencies or system failure. In Toronto, the subway system alone is afflicted by more than 12000 incidents/year, approximately 1000 of which are considered “major”. Transit delays often exacerbate overcrowding, and the opposite is also true. The cost to users and the operator can be enormous. One recent study estimated the economic cost of user delays due to subway incidents in New York to be upward of 389 million USD annually. This research program is aimed directly at tackling the problems of delays and overcrowding in large multi-modal transit systems. It adopts a broad, multi-faceted approach which will produce solutions for demand management, supply management and flow management, all aimed consistently and coherently at alleviating transit delays and overcrowding, particularly in subway systems. To manage demand, our research will develop innovative incentive-based programs which offer reward points to transit riders who switch away from overcrowded spots and times. Incentive-based and loyalty programs have been applied effectively to a variety of other services and industries, and while they are slowly being introduced to public transit they require supporting empirical tools to maximize their effectiveness. These tools will be developed in this research. To manage supply, specifically in the event of a major disruption to subway service, we will develop a system of models and tools to help transit agencies deploy effective shuttle bus services to fill the gap in subway service. In more specific terms, we will develop analytics to predict in real-time the expected duration of any given subway incident, and we will determine the optimal number of buses and which routes these buses should be pulled from for dispatching to the disrupted subway section such that total delays are minimized. The third major track of this research is concerned with managing the flows of passengers into all stations of a subway line to avoid capacity loss due to overcrowding. This will be achieved through creative scheduling of the bus and streetcar feeder lines connecting with the subway line, along with a real-time control strategy to avoid oversaturation at any given station. The proposed solutions will be developed using the rich and diverse data on transit operations and demand which are now available abundantly. Founded in sound theory, the proposed models will be built using rigorous and new modelling approaches. The proposed research is likely to have a transformative effect on transit resilience policy and technologies.