SCC-IRG Track 1: Inclusive Public Transit Toolkit to Assess Quality of Service Across Socioeconomic Status in Baltimore City

大多数拥有大量公共交通乘客的美国城市都有一个严酷的统计数据：乘坐公共交通的通勤者的收入比使用汽车的通勤者低得多。以前的研究还表明，使用公共交通的高收入居民通常依赖于单一的登机旅行，而低收入者则需要忍受复杂而漫长的旅行，需要多种模式或换乘。传统上，运输机构使用服务质量（QoS）调查来衡量乘客对性能的看法。然而，这些调查存在重要的局限性，往往掩盖了具有复杂流动经历的低收入居民所面临的挑战。为了解决这些差距，已经开发了几种允许居民收集GPS标记的QoS数据的智能手机应用程序。虽然前景看好，但这些应用程序不仅无法收集关键信息来描述复杂的旅行，而且缺乏隐私，透明度和决策支持系统。该项目将创造新的方法，回答开放的经验问题，并为设计，开发，部署和评估尊重隐私的工具包提供基于研究的指导方针，以确定和描述低收入居民经常经历的复杂旅行的典型多因素挑战;并通过社区对话和决策支持系统推动自下而上，众包知情的可操作解决方案。这项跨学科的研究工作将推进隐私，调查设计，数据分析，公交公平，数据驱动的公民参与和公交模拟的最新技术;并将创造一个独特的机会来了解这些研究领域之间的相互依赖关系，从而产生最终推动QoS公交改进所需的新知识。通过让所有相关利益相关者参与该项目，并将居民和交通公平放在中心位置，该项目提出了一种更加公平和以人为本的智慧城市方法，其中技术不提供给居民，而是与他们一起设计，以满足明确的需求。为了实现这一目标，本项目将需要回答沿着四个研究主题的研究问题：（1）理解参与：分析可能阻止低收入居民参与移动体验数据收集工作的隐私障碍，如何降低这些障碍以维持参与，以及可能在居民参与和质量数据之间提供良好平衡的QoS调查策略;（2）移动体验数据分析：创建新颖的、可解释的机器学习和统计方法，从大规模、高维度、门到门的移动体验数据中识别和描述交通挑战和公平性，并以所有利益相关者都能理解的方式做到这一点;（3）公民参与和解决方案构思的透明度：确定创建民主空间所需的条件，流程，工具和数据，通过所有相关利益攸关方（居民，倡导团体和决策者）之间的透明，数据驱动的邻里对话，可以确定公共交通挑战的解决方案;（4）基于公交服务质量的仿真决策支持系统：小说创作，可解释的模拟，以确定对城市规模的交通的影响，结合当地的解决方案，以解决特定的社区确定-该奖项反映了NSF的法定使命，并已被认为是值得通过评估使用基金会的智力价值和更广泛的支持影响审查标准。

Most American cities with substantial public transit ridership share a stark statistic: commuters on public transportation have disproportionately lower incomes than commuters who use automobiles. Previous research has also shown that higher income residents who use public transit typically rely on single-boarding trips, while lower-income individuals endure complex, lengthy trips, requiring several modes or transfers. Traditionally, transit agencies use quality of service (QoS) surveys to gauge passenger perceptions of performance. However, these surveys suffer important limitations that more often mask challenges faced by low-income residents with complex mobility experiences. In an attempt to address these gaps, several smartphone applications that allow residents to collect GPS-tagged, QoS data have been developed. While promising, these apps not only fail to collect critical information to characterize complex trips, but also lack privacy, transparency and decision support systems. This project will create novel methods, answer open empirical questions and provide research-based guidelines for the design, development, deployment and evaluation of a privacy-respectful toolkit to identify and characterize the multi-factorial challenges typical of complex trips often times endured by low-income residents; and to drive bottom-up, crowdsourced-informed actionable solutions via community conversations and a decision support system. This interdisciplinary research effort will advance the state of the art in privacy, survey design, data analytics, transit equity, data-driven civic engagement, and transit simulations; and will create a unique opportunity to understand the interdependencies between these research areas generating new knowledge necessary to ultimately drive QoS transit improvements. By involving all relevant stakeholders in the project, and putting residents and transit equity at the center, this project propounds a more equitable and human-centered approach to smart cities, one in which technologies are not presented to residents, but rather designed with them to address articulated needs. To achieve this goal, this project will need to answer research questions organized along four research thrusts: (1) Understanding Participation: analysis of the privacy barriers that might prevent low-income residents from participating in mobility experience data collection efforts, how to lower them to sustain participation, and the QoS survey strategies that might provide a good balance between resident participation and quality data; (2) Mobility Experience Data Analysis: creation of novel, interpretable machine learning and statistical methods to identify and characterize transit challenges and equity from large-scale, high-dimensional, door-to-door mobility experiential data, and to do so in a way that is interpretable to all stakeholders; (3) Transparency for Civic Engagement and Solution Ideation: identification of the conditions, processes, tools and data needed to create democratic spaces where solutions to public transit challenges can be identified via transparent, data-driven, neighborhood conversations among all stakeholders involved: residents, advocacy groups and decision makers; and (4) Simulation-based Decision Support Systems based on transit QoS: creation of novel, interpretable simulations to identify the impact on city-scale transportation by incorporating local solutions to address specific community-identified-needs.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.