Currans Kristina M, Abou-Zeid Gabriella, McCahill Chris, Iroz-Elardo Nicole, Clifton Kelly J, Handy Susan, Pineda Irene
Urban Planning, University of Arizona, Tucson, AZ USA.
Transportation Data, ICF, Tucson, AZ USA.
Transportation (Amst). 2022 Jul 29:1-26. doi: 10.1007/s11116-022-10306-8.
Parking supply is one of the most neglected elements of the built environment in travel behavior research, despite evidence linking parking with vehicle use. As transportation impacts of new development are increasingly measured by vehicle miles traveled (VMT), explicitly connecting parking characteristics with vehicle travel is necessary to better inform transportation and land use policy. In this paper, we begin to address this research gap and explore the relationship between constrained parking and household VMT. Utilizing the 2017 National Household Travel Survey (NHTS) California add-on sample, we estimate residential parking constraint for households in Los Angeles County. Then, we develop a two-level model framework. Level 1 (Cost) models estimate travel costs, represented by vehicle ownership as a function of parking constraints, the built environment, and demographics. Level 2 (Demand) models regress household-level total and homebased-work VMT on predicted vehicle ownership, controlling for temporal and environmental characteristics. To further explore the relationship between parking and VMT by place type, we applied Level 1 and Level 2 models to develop a suite of scenarios for typical households in Los Angeles County. Our findings support the hypothesis that the built environment (including parking) influences VMT through travel costs (vehicle ownership). Results from scenarios analysis reveal constrained on-site residential parking (< 1 parking space per dwelling unit), accounts for an approximate 10-23 percentage-point decrease in VMT within each place type. Finally, implications for practice and future research are presented.
尽管有证据表明停车与车辆使用之间存在联系，但在出行行为研究中，停车供应是建成环境中最被忽视的要素之一。由于新开发项目对交通的影响越来越多地通过车辆行驶里程（VMT）来衡量，因此明确将停车特征与车辆出行联系起来，对于更好地为交通和土地利用政策提供信息至关重要。在本文中，我们开始填补这一研究空白，探讨受限停车与家庭车辆行驶里程之间的关系。利用2017年美国家庭出行调查（NHTS）加利福尼亚补充样本，我们估算了洛杉矶县家庭的住宅停车限制情况。然后，我们开发了一个两级模型框架。一级（成本）模型估算出行成本，以车辆拥有量为代表，它是停车限制、建成环境和人口统计特征的函数。二级（需求）模型将家庭层面的总VMT和基于家的工作VMT对预测的车辆拥有量进行回归分析，同时控制时间和环境特征。为了进一步按地点类型探索停车与VMT之间的关系，我们应用一级和二级模型为洛杉矶县的典型家庭开发了一套情景。我们的研究结果支持这样的假设，即建成环境（包括停车）通过出行成本（车辆拥有量）影响VMT。情景分析结果显示，受限的住宅现场停车（每个居住单元<1个停车位）在每种地点类型中导致VMT下降约10 - 23个百分点。最后，本文阐述了对实践和未来研究的启示。
