CAREER: Impact of Freeway Geometric Design on Congestion Characteristics

学院早期职业发展(Career)计划奖项的目标是建立高速公路设计元素(如坡度和曲率)与交通拥堵特征(如走走停停的频率、滞后和通行能力下降)之间的关系。研究方法基于实证数据分析，这些数据既有，也是作为本项目的一部分收集的。车辆轨迹数据将通过数值模拟和统计分析相结合的方式，将拥堵特征与个别驾驶员的行为相关联。这将有助于揭示司机行为和高速公路设计之间的联系，并提出并验证能够捕捉到这项研究结果的跟车理论。如果成功，这项研究的结果将导致改进高速公路设计指南，目前该指南主要侧重于提供足够的通行能力。这项研究将为以下问题提供答案：我们应该如何设计高速公路，以便在拥堵时：(I)加减速操作次数最少？或(Ii)瓶颈流量最大化？或(Iii)换道次数最少？这些问题的答案可能有助于将温室气体、通勤延误和事故降至最低，并将促进未来更可持续的基础设施系统。此外，这项研究将为从业者和研究人员提供可以说是第一个能够捕捉到经验观察到的复杂拥堵现象的交通流模型。这将提高现有基于模拟的高速公路基础设施项目评估的准确性，并将允许测试更高效的拥堵控制策略，这些策略侧重于将排放水平、延误和事故降至最低。教育和推广目标包括一个促进交通流理论、模拟和建模的在线门户，学生、研究人员和行业专业人员将在其中受益。该门户包括每周的网络研讨会、数据和程序工具、一个论坛和一个维基。

The objective of this Faculty Early Career Development (CAREER) program award is to establish the relationship between freeway design elements, such as grade and curvature, and traffic congestion characteristics, such as the frequency of stop-and-go movements, hysteresis and capacity drop. The research approach is based on empirical data analysis, both existing and collected as part of this project. Vehicle trajectory data will be used to correlate congestion characteristics and individual driver behavior using a combination of numerical simulations and statistical analyses. This will allow for the unveiling of the link between driver behavior and freeway design, and to propose and validate a car-following theory able to capture the findings in this research. If successful, the results of this research will lead to improved freeway design guidelines, which currently mostly focus on providing adequate capacity. This research will provide answers to questions such as: How should we design freeways so that, when congested, (i) the number of acceleration/deceleration maneuvers is minimized?, or (ii) the bottleneck discharge rate is maximized?, or (iii) the number of lane-changes is minimized? The answer to these questions will potentially help minimize greenhouse gases, commuter delays and accidents, and will promote a more sustainable infrastructure system in the future. Additionally, this research will provide practitioners and researchers with arguably the first traffic flow model able to capture complex congestion phenomena as observed empirically. This will improve the accuracy of existing freeway infrastructure project simulation-based assessments, and will allow testing more efficient congestion control strategies that focus on minimizing emission levels, delays and accidents. The educational and outreach objectives include an on-line portal for promoting traffic flow theory, simulation and modeling, where a community of students, researchers and industry professionals will benefit. This portal includes weekly webinars, data and program tools, a discussion forum and a wiki.