Excellence in Research/Collaborative Research: Modeling Transportation Choices Under the Presence of Real-time Information Using Simulated-based Virtual Experiments

卓越的研究/合作研究：使用基于模拟的虚拟实验在实时信息存在下对交通选择进行建模

• 批准号: 2200590
• 负责人: Venktesh Pandey
• 金额: $ 28万
• 依托单位: North Carolina Agricultural & Technical State University
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2022
• 资助国家: 美国
• 起止时间: 2022-10-01 至 2025-09-30
• 项目状态: 未结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=2200590&HistoricalAwards=false
• 关键词: Excellence; Research; Collaborative; Modeling; Transportation

This Excellence in Research (EiR) project will investigate the impacts of real-time information on the sequential choice-making behavior of travelers using transportation infrastructure and services in smart cities. Smart cities of the future are envisioned as a place where digitalization of day-to-day services, interconnectedness of sensing technologies, and intelligent algorithms for city management lead to innovative socioeconomic and socio-technical growth. Travelers using transportation services in these “cities of the future” make various choices in a dynamically evolving system such as the choice of destination, departure time, mode of travel, travel route, and/or parking location. This project studies the interrelation of choices and the impact of real-time information on choices by adopting a data-driven modeling approach. The tools and models developed as part of this project will enable data-driven choice modeling and management of future transportation systems and will contribute toward NSF’s mission of promoting the progress of science and advancing the nation’s prosperity and welfare. Beyond its intellectual merit, this project will (a) enable underrepresented students at Historically Black Colleges and Universities to participate in research on future transportation systems, (b) develop educational tools for accurate choice modeling of travelers in response to real-time information, and (c) develop incentives that guide the design and development of smart cities of the future. Eventually, the project will enable agencies to perform better short-term and long-term transportation planning, which will have a substantial societal impact.As its primary goal, the project will explain travelers’ choices under the presence of real-time information using an advanced econometric framework that identifies the factors affecting these choices, including time-varying characteristics of emerging technologies, level-of-service, and socio-demographic variables. This project will contribute to the choice modeling theory and experiments by (1) explicitly incorporating real-time information in sequential dynamic choices of travelers, (2) modeling the impact of the error-prone or aggregated information on travel choices through D-efficient experimental design, and (3) developing a policy-sensitive tool that will be able to evaluate a wide range of smart-city-related transportation policies. This project will adopt innovative techniques such as creating a simulation-based virtual environment for experiments on choices made by travelers in a dynamically evolving system, and developing and calibrating robust mathematical models useful for accurate forecasting of travel behavior. The focus applications in this project will also provide guidance on optimization and operations of civil infrastructure systems and will enable methodological evaluation of transportation policies in Smart Cities that may impact travel behavior over the long term.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

这个卓越研究（EiR）项目将调查实时信息对智能城市中使用交通基础设施和服务的旅行者顺序选择行为的影响。未来的智慧城市被设想为一个日常服务的数字化、传感技术的互联性和城市管理的智能算法导致创新的社会经济和社会技术增长的地方。在这些“未来城市”中使用交通服务的旅行者在一个动态发展的系统中做出各种选择，如选择目的地、出发时间、旅行方式、旅行路线和/或停车位置。本项目通过采用数据驱动的建模方法，研究选择之间的相互关系以及实时信息对选择的影响。作为该项目的一部分，开发的工具和模型将使未来交通系统的数据驱动选择建模和管理成为可能，并将为NSF促进科学进步和促进国家繁荣和福利的使命做出贡献。除了它的智力价值之外，这个项目将(a)使传统黑人学院和大学中代表性不足的学生能够参与对未来交通系统的研究，(b)开发教育工具，根据实时信息对旅行者进行准确的选择建模，以及(c)制定激励措施，指导未来智慧城市的设计和发展。最终，该项目将使各机构能够更好地进行短期和长期交通规划，这将产生重大的社会影响。该项目的主要目标是，利用先进的计量经济学框架，在实时信息的存在下解释旅行者的选择，该框架确定影响这些选择的因素，包括新兴技术的时变特征、服务水平和社会人口变量。该项目将通过(1)明确地将实时信息纳入旅行者的连续动态选择中，(2)通过D-efficient实验设计对易出错或汇总信息对旅行选择的影响进行建模，以及(3)开发一种政策敏感工具，能够评估各种与智慧城市相关的交通政策，从而为选择建模理论和实验做出贡献。该项目将采用创新技术，例如创建一个基于仿真的虚拟环境，用于在动态发展的系统中对旅行者所做的选择进行实验，以及开发和校准有助于准确预测旅行行为的强大数学模型。该项目的重点应用还将为民用基础设施系统的优化和运营提供指导，并将对智慧城市中可能影响长期出行行为的交通政策进行方法学评估。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

项目成果

Physics Informed Temporal Multimodal Multivariate Learning for Short-Term Traffic State Prediction

用于短期交通状态预测的物理信息时态多模态多元学习

• 发表时间: 2023
• 期刊: INFORMS Transportation and Logistics Society Second Triennial Conference
• 作者: Deshpande, Niharika;Darko, Justice;Park, Hyoshin;Yoon, Gyugeun;Pandey, Venktesh
• 通讯作者: Pandey, Venktesh

Subscription Models for Differential Access to Real-time Information

用于差异化访问实时信息的订阅模型

• 发表时间: 2023
• 期刊: INFORMS Transportation and Logistics Society Second Triennial Conference
• 作者: Pandey, Venktesh;Neupane, Anusha;Park, Hyoshin
• 通讯作者: Park, Hyoshin