打印

作者

Yonah Freemark;Anne Hudson;Jinhua Zhao

来源

JOURNAL OF THE AMERICAN PLANNING ASSOCIATION,Vol.85,Issue2,P.133-151

语言

英文

关键字

autonomous vehicles,municipalities,transportation planning

作者单位

摘要

AbstractProblem, research strategy, and findings: Local government policies could affect how autonomous vehicle (AV) technology is deployed. In this study we examine how municipalities are planning for AVs, identify local characteristics that are associated with preparation, and describe what effects bureaucrats expect from the vehicles. We review existing plans of the 25 largest U.S. cities and survey transportation and planning officials from 120 cities, representative of all municipalities with populations larger than 100,000. First, we find that few local governments have begun planning for AVs. Second, cities with larger populations and higher population growth are more likely to be prepared. Third, although local officials are optimistic about the technology and its potential to increase safety while reducing congestion, costs, and pollution, more than a third of respondents worried about AVs increasing vehicle miles traveled and sprawl while reducing transit ridership and local revenues. Those concerns are associated with greater willingness to implement AV regulations, but there is variation among responses depending on political ideology, per capita government expenditures, and population density.Takeaway for practice: Municipal governments’ future approaches to AV preparation will likely depend on the characteristics of city residents and local resources. Planners can maximize policy advancement if they work with officials in other cities to develop best practices and articulate strategies that overlap with existing priorities, such as reducing pollution and single-occupancy commuting.Keywords: autonomous vehicles, municipalities, transportation planningAdditional informationAuthor informationYonah FreemarkYONAH FREEMARK (freemark@mit.edu) is a PhD candidate in the Massachusetts Institute of Technology (MIT) Department of Urban Studies and Planning (DUSP) studying land use and transportation policy.Anne HudsonANNE HUDSON (awhudson@mit.edu) is a master of city planning and master of science in transportation student at MIT DUSP.Jinhua ZhaoJINHUA ZHAO (jinhua@mit.edu) is the Edward H. and Joyce Linde Associate Professor of city and transportation planning at MIT DUSP and director of the MIT Urban Mobility Lab. He conducts research on travel behavior, systems, and policies.AcknowledgmentsWe thank Nina Brooks, Joanna Moody, Daniel Nichanian, and Shenhao Wang for their aid in reviewing drafts of our article, as well as all members of MIT’s JTL Urban Mobility Lab for their feedback during research development. We also thank three anonymous reviewers for their useful suggestions throughout the revision process.Research SupportThe Singapore–MIT Alliance for Research and Technology (SMART) and University Transportation Center New England (Grant # MITR25-49) partially funded this research.Supplemental MaterialSupplemental data for this article can be found on the publisher’s website.Notes1 For the purposes of this study, we do not distinguish between policies and regulations because we focus on the impetus for municipal action on AV technology.2 “Highly” automated vehicles are levels 3 to 5 on the Society of Automotive Engineers’ classification; this includes system control of steering, acceleration, and monitoring of the driving environment. Levels 4 to 5 also involve no fallback to human drivers in the case of dynamic driving tasks, and level 5 extends automation to all driving modes (Li, Sui, Xiao, & Chahine, 2018 Li, S., Sui, P., Xiao, J., & Chahine, R. (2018). Policy formulation for highly automated vehicles: Emerging importance, research frontiers and insights. Transportation Research Part A. Advance online publication. doi:10.1016/j.tra.2018.05.010[Crossref] , [Google Scholar]).3 See, for example, the 3 Revolutions: Shared, Automated, Electric project of the University of California at Davis, Institute of Transportation Studies (2017 Institute of Transportation Studies at UC Davis. (2017). 3 Revolutions: Sharing, electrification, and automation. Davis, CA: Author. Retrieved from https://3rev.ucdavis.edu/policybriefs/ [Google Scholar]) (http://3rev.ucdavis.edu).4 In some states, such as Pennsylvania, taxi service is regulated at the state level by a special commission as opposed to by cities themselves (Schaller, 2007 Schaller, B. (2007). Entry controls in taxi regulation: Implications of US and Canadian experience for taxi regulation and deregulation. Transport Policy, 14(6), 490–506. doi:10.1016/j.tranpol.2007.04.010[Crossref], [Web of Science ®] , [Google Scholar]).5 We also conducted a search of ordinances or mayoral executive orders related to AVs beyond requirements for testing. Though Portland (OR) developed a draft policy in 2017, in no city did we identify legal documents directly associating city planning issues with AVs.6 In the survey documented here, we did not define what we meant by “AV,” leaving this question open ended. Based on text responses, most interpreted it as meaning a passenger car–sized vehicle operating on city streets, as it has been described commonly in the U.S. press; we did not receive any indication that there was confusion on this matter. Nevertheless, to clarify, there are other forms of automated transportation, such as trains and buses.7 We also tested number of departmental employees per capita, adjusted for local population levels. This produced similar results.8 We find little significant correlation between more liberal political ideology and increasing support for municipal regulations (see Technical Appendix Table A-3; in fact, in model 7, we find the opposite, though these effects disappear with additional controls in models 9 and 10), despite the correspondence in Table 6 between the two. It is true that when we run a single-variable probit regression, we find a significant (p < .05) and strong relationship; however, this relationship disappears once we control for local population size.9 To do this, we transformed Likert responses into a −2 to +2 scale.