Delft University of Technology
Automated vehicles and how they may affect urban form
A review of recent scenario studies
Stead, Dominic; Vaddadi, Bhavana
DOI
10.1016/j.cities.2019.03.020
Publication date
2019
Document Version
Final published version
Published in
Cities: the international journal of urban policy and planning
Citation (APA)
Stead, D., & Vaddadi, B. (2019). Automated vehicles and how they may affect urban form: A review of
recent scenario studies. Cities: the international journal of urban policy and planning, 92, 125-133.
https://doi.org/10.1016/j.cities.2019.03.020
Important note
To cite this publication, please use the final published version (if applicable).
Please check the document version above.
Copyright
Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons. Takedown policy
Please contact us and provide details if you believe this document breaches copyrights. We will remove access to the work immediately and investigate your claim.
This work is downloaded from Delft University of Technology.
Green Open Access added to TU Delft Institutional Repository
‘You share, we take care!’ – Taverne project
https://www.openaccess.nl/en/you-share-we-take-care
Otherwise as indicated in the copyright section: the publisher
is the copyright holder of this work and the author uses the
Dutch legislation to make this work public.
Contents lists available atScienceDirect
Cities
journal homepage:www.elsevier.com/locate/cities
Automated vehicles and how they may affect urban form: A review of recent
scenario studies
Dominic Stead
a,⁎, Bhavana Vaddadi
baFaculty of Architecture and the Built Environment, Delft University of Technology, Julianalaan 134, Delft 2628BL, the Netherlands bIntegrated Transport Research Laboratory, KTH Royal Institute of Technology, Drottning Kristinas Väg 40, 114 28 Stockholm, Sweden
A R T I C L E I N F O Keywords: Automated vehicles Urban form Urban design Land use Scenarios A B S T R A C T
Motorisation in cities has fundamentally transformed urban patterns of development, ranging from residential parking and density standards of single buildings on one hand to urban infrastructure construction and the expansion of entire cities on the other. The introduction of automated vehicles (AVs) has enormous potential to transform urbanisation patterns and urban design even further. However, the direction in which this technology will change the city is contested and a diverse set of views can be found. This paper provides a review of scenarios on these issues to date. Although some scenario studies provide useful insights about urban growth and change, very few consider detailed impacts of AVs on urban form, such as the density and mix of functions, the layout of urban development and the accessibility of locations, including the distance to transit.
1. Introduction
Within one or two decades, cities may witness the widespread in-troduction of fully automated vehicles (AVs),1 vehicles capable of sensing the urban environment and navigating without human input. These vehicles have the potential to fundamentally transform urbani-sation patterns, urban spaces, street life and the quality of cities to a similar extent as the introduction of the first motorised vehicles in the city. Optimists point to the potential for AVs (especially when shared) to substantially reduce the total number of vehicles in the city, creating the opportunity for more car-free areas (e.g. parks, squares), whereas pessimists speak of the risk of more suburbanisation and urban sprawl as a result of longer travel times and distances that are considered ac-ceptable due to the increased opportunities to work (or relax) while travelling by AV (see for examplePapa & Ferreira, 2018).
Assessing the long-term impacts of AVs on urban development is difficult and complex since many impacts are dependent on decisions related to the future role of AVs in society and in the city (Litman, 2019; Townsend, 2014). These include decisions such as where and when AVs will be allowed to drive in the city, whether AVs can be operated alongside human driven vehicles (or segregated), how car users (of both conventional and automated vehicles) will respond to the introduction of AVs, whether AVs will be primarily used as private or shared ve-hicles, and the degree of safety and security they manage to achieve. To
understand a range of possible impacts of AVs, several studies involving scenario-building exercises have been carried out in recent years. While many of them focus on the technical aspects of their introduction, re-latively few studies consider the potential impact of AVs on spatial development of cities, which is the main focus of this review. Amongst the scenario studies exploring how AVs may impact cities in the long term, several aspects of impacts on urban form have been considered, including parking, road capacity and residential location.
Various researchers have started to consider the potential positive and negative effects of this technology on the city. Speculation about the potential future impacts of AVs on urban development has captured the attention of urban planners and designers around the world (Stone, Ashmore, Scheurer, Legacy, & Curtis, 2018). However, there is little consensus about the future impacts of AVs on urban development, sometimes not even in terms of the overall direction of change for certain types of impacts (Table 1). Conjecture still abounds when it comes to the potential impacts of AVs on urban development patterns and the quality and use of urban space (Cavoli, Phillips, Cohen, & Jones, 2017). According toPapa and Ferreira (2018), vehicle automa-tion may open the door to a large variety of technological, design-re-lated, legal and cultural changes that could either improve or worsen accessibility in urban environments. Whatever the direction of these changes in accessibility, the introduction of a new transport technology like AVs is likely to have significant impacts on urbanisation patterns
https://doi.org/10.1016/j.cities.2019.03.020
Received 28 September 2018; Received in revised form 20 January 2019; Accepted 27 March 2019 ⁎Corresponding author.
E-mail addresses:d.stead@tudelft.nl(D. Stead),bhavana@kth.se(B. Vaddadi). 1Otherwise called autonomous, driverless, self-driving or robotic cars.
Cities 92 (2019) 125–133
0264-2751/ © 2019 Elsevier Ltd. All rights reserved.
and urban form. AVs will also have implications for the governance of urban development and urban transport (Stone et al., 2018). In parallel, urban planning policies and spatial development decisions will influ-ence the design, introduction and take-up of AVs in the city.
It is important to recognise here that the introduction of a new technology like AVs can have multiple and often parallel impacts on the transport system (see for exampleCohen-Blankshtain & Rotem-Mindali, 2016). First, new technologies have the potential to change travel de-mand directly by reducing, inducing or modifying travel patterns. Second, new technologies have the potential to change the capacity of transport infrastructure and thereby affect the efficiency (and/or ca-pacity) of the transport system and transport users' behaviour. Third, new technologies can directly affect urban development patterns which in turn can influence urban mobility patterns.
AVs represent a potentially highly disruptive technology that could lead to shifts in mobility patterns and urban development of a similar order of magnitude as the first introduction of the car, metro or tram. Moreover, AVs have the potential to transform the urban form of the city which in turn can have a strong influence on travel patterns, transport demand and infrastructure use. At the same time, travel patterns, transport demand and infrastructure use can all influence urban form (Kasraian, Maat, Stead, & van Wee, 2016). Urban form is concerned with the physical characteristics of built-up areas, including the shape, size, density and configuration of development (RTPI, 2015). It can be considered at a range of different scales, varying from regional to local (see for example Stead & Marshall, 2001), and frequently measured in terms of five inter-related dimensions: density (of activ-ities), diversity (of land uses), design (of streets and paths), destination accessibility (to jobs or services) and distance to transit (seeEwing and Cervero (2010)).2
The aim of this paper is to review and synthesize the content of the different types of scenarios that have been put forward in recent studies concerning the introduction of AVs and their implications for urban form. While some of these studies consider the role of urban form in influencing the introduction and take-up of AVs, most consider the reverse relationship – the way in which the introduction and use of AVs may affect the future development of urban form.
The paper is divided into five main parts. The first part considers the role and importance of scenario studies for long-term decision-making in transport and urban planning. The second part outlines the research approach and methods employed to identify and review the various scenarios of AVs and their impacts on cities. In the third part, recent scenario studies are reviewed and a synthesis of their findings is pre-sented, including an overview of the extent to which the scenario stu-dies have considered different aspects of urban form. Fourth, the paper identifies gaps in attention to different aspects of urban form (i.e. the scales or types of urban form which the scenario studies do not con-sider). In the fifth and final part, the conclusions of the research are identified together with potential new directions for future research.
2. Scenario studies
Scenarios are hypothetical sets of developments for focusing atten-tion on a certain issue or to take an informed decision. Scenarios can be of two main types: (i) quantitative scenarios often based on models which contain implicit assumptions about the future; and (ii) qualita-tive scenarios which are primarily narraqualita-tive and describe the future in the form of words or visual symbols. In addition, three basic approaches of scenario construction can be distinguished: (i) forecasting ap-proaches; (ii) exploratory apap-proaches; and (iii) backcasting approaches. In many scenario studies, combinations of these different approaches can be found (Banister & Hickman, 2013). The nature of these three approaches is briefly outlined below.
First, forecasting approaches are often used to investigate both probable futures and possible futures. They are useful for situations where substantial external change is not expected and where there is some expectation that current trends will continue into the future. This type of approach when applied to land use and transport planning is frequently quantitative, focusing on ways to reduce travel and emis-sions compared to trend forecasts, with land use variables such as density and location of growth (Bartholomew, 2007).
Second, exploratory approaches are the most widely used form of scenarios and typically take two dimensions of change within which four scenarios are constructed, usually within a two by two matrix (Banister & Hickman, 2013). Here, the main aim is to identify key challenges and to trigger new thinking. Exploratory approaches tend to be trend-breaking. They are usually interactive and participatory and involve the identification of key independent and dependent variables, the analysis of actor roles and strategies, and the construction of plausible scenarios (Godet, 2001).
Third, backcasting approaches are often used to look at preferred futures over the longer term and are designed to be trend-breaking in their thinking. They are often combined with exploratory approaches, and examine the means by which those futures can be reached. A central part of the backcasting process is the identification of pathways from a desirable future back to the present. This includes timelines for critical decisions in terms of implementation of change and other key issues. This approach often puts emphasis on visioning the future and the use of expert and other groups so that discussion, feedback and participation all take place.
The process of scenario-building generally involves the develop-ment of two or more scenarios that are used to compare and examine alternative futures. Scenario-building involves multiple stages (see Stead & Banister, 2003for a description of these different stages), in-cluding the development of a reference case (sometimes referred to as the Business as Usual or BAU scenario), against which other scenarios can be compared.
Scenario studies provide a useful tool for supporting long-term thinking and decision-making on transport and urban development policy. Such studies offer a way of testing the robustness of long-term decisions under a wide range of policy-making conditions (see also Pearman, 1988;Stead & Banister, 2003). They are particularly useful in Table 1
Optimistic and pessimistic views about the impacts of AVs on land use and parking. (Adapted fromPapa & Ferreira, 2018.)
Optimistic Pessimistic
Land use The quality of the built environment will be improved (re-centralization or regeneration of inner areas, re-densification, land use changes to new green public areas, residential locations). AVs are seen as a way of promoting better quality of life in cities.
The built environment will be reshaped to accommodate the needs of AVs and their users in preference to the needs of other social groups. AVs increase suburbanisation or sprawl due to the comfort of trips.
Parking Parking policies will facilitate the conversion of redundant parking places into new recreational, green, and building areas, or into transport infrastructures for active modes of transport. Parking is limited within cities. Pick-up and drop-off spaces are only provided at a limited number of locations in the city.
Parking policies will remain as they are (i.e. AVs will use on-road parking spaces). A growth in AV ownership and use leads to an increase in demand for parking spaces in the city. Large numbers of collection/drop-off points are created in the city which add to the amount of space allocated for vehicles.
2Ewing and Cervero (2010)refer to these dimensions as the five Ds.
D. Stead and B. Vaddadi Cities 92 (2019) 125–133
situations of high uncertainty such as long timescales or rapid techno-logical changes. Considering and comparing alternative scenarios, re-flecting major uncertainties, can provide useful information for policy-makers (and academics) about how to anticipate and respond to future developments and can provide insights into ways of avoiding adverse outcomes that might occur in the future. In order to understand the nature of future transport systems, an important part of scenario-building is the identification of key factors which may determine qua-litatively different paths for the socio-economic environment. May (1982)identifies a number of potential benefits of policy scenarios to decision-making, including:
•
providing useful frameworks for decision-making – scenarios allow decision-making issues to be explored using a range of alternative scenarios, reflecting different assumptions about the future;•
identifying dangers and opportunities – considering a range of al-ternative futures increases the likelihood of identifying possible problems and opportunities in policy-making; suggesting a variety of possible approaches – the use of scenarios may generate a range of approaches to tackle issues or problems whereas the use of forecasts, often based on single theories or simple extrapolations, often leads to the pursuit of singular solutions;•
helping to assess alternative policies and actions – scenarios may for example be used to identify the usefulness of different policies under alternative future conditions; and•
increasing creativity and choice in decision-making – identifying possible future developments and avoiding the acceptance of cur-rent trends as inevitable opens up new possibilities for policy de-velopment.While scenario studies offer a number of benefits, it is recognised that there are also limitations associated with their use. First, like any other form of considering long-term futures, scenario studies are always a simplification of reality since they cannot include all possible factors or developments that will occur. Second, scenario studies can never provide a completely accurate view or image of the future since there are often many unpredictable factors involved and an almost infinite number of alternatives and variables regarding what may and may not happen in the future.3Third, a range of methods and techniques can be employed for constructing scenarios, which can result in a diverse set of interpretations about future development.
3. Research methods
This paper reviews the scenarios contained in recent studies drafted during the last 5 years (2013–2017), most of which employ exploratory approaches (see below). All of the studies reviewed in this paper con-sider the issue of urban form and structure to some degree (some go into more detail than others). Scenario studies that do not consider the impact on urban form and structure were not included in this review.4 The scenarios contained in these papers and reports present a range of different futures, the majority of which is speculative in nature (as opposed to predictive). The scenario studies reviewed in this paper come from a mixture of two main sources: (i) academic papers (jour-nals, conference proceedings and books); and (ii) reports produced by government agencies or consultants. The studies were identified using a keyword search in Scopus and Google Scholar (using combinations of the keywords: self-driving, automated, autonomous, vehicles, land use,
space, spatial, scenarios) supplemented by forward and backward snowball techniques.5
Each of the scenarios reviewed in this paper have been classified into one of four main types. These types were inductively developed on the basis of the main content and assumptions of the scenarios. This classification distinguishes between the following four main types:
1. Business as Usual (BAU). Reference scenarios which assume the continuation of one or more current trends (in mobility, urban de-velopment and/or demographics), without the introduction of AVs. 2. Technology + Non-Shared (T). Scenarios which assume the troduction of AVs which are either solely or predominantly in-dividually owned and used.
3. Technology + Shared (T+). Scenarios which assume the in-troduction of AVs which are solely or predominantly shared. 4. Technology + Shared+ Infrastructure/Policy (T++). Scenarios
which assume the introduction of AVs which are solely or pre-dominantly shared. In addition, supportive policies and/or infra-structures are introduced to actively promote the uptake and use of AVs.
In total, 13 literature sources were identified and selected for review in this paper, comprising three academic reports (Fulton, Mason, & Meroux, 2017; Pernestål Brenden, Kristoffersson, & Mattsson, 2017; Townsend, 2014), three government agency reports (Chapin et al., 2016;International Transport Forum, 2015;Tillema et al., 2015), two conference papers (Gruel & Stanford, 2016;Thakur, Kinghorn, & Grace, 2016), two consultants' reports (Bouton, Knupfer, Mihov, & Swartz, 2015; Corwin, Vitale, Eamonn, & Cathles, 2015), one journal article (Milakis, Snelder, van Arem, van Wee, & Correia, 2017), one book chapter (Heinrichs, 2016) and one report from a transport umbrella organisation (Röhrleef, Deutsch, & Ackermann, 2015). Almost all the sources contain several scenarios (Table 2), each of which are reviewed below.6As can be seen inTable 2, the majority of studies uses scenarios for exploratory purposes while most of the other studies use scenarios in order to develop forecasts.
4. Content of scenarios – assumptions about future travel and land use patterns
4.1. Business as usual (BAU)
Of the 13 papers reviewed, 6 papers contained a Business as Usual scenario. Of the ones which specify a time frame, they range from 2028 to 2050.7While several papers are global in scope (e.g.Corwin et al., 2015;Fulton et al., 2017), others focus on specific countries such as Australia (Thakur et al., 2016), Germany (Heinrichs, 2016; Röhrleef et al., 2015), Netherlands (Milakis et al., 2017;Tillema et al., 2015) and the United States (Townsend, 2014).
The Business as Usual scenarios largely assume that current trends, attitudes and priorities remain largely unchanged in the future. These trends refer amongst other things to changes in technology, economics, demographics and politics. The underlying assumptions here are that technological innovations are not taken up to any great extent, parti-cularly due to the high cost of the necessary infrastructure. Significant technological development does take place and is mainly restricted to efficiency gains in specific areas (Heinrichs, 2016). The scenarios
3In fact, the main purpose of many scenario studies is not to predict the most plausible or likely future but to illustrate the range in different possible futures. 4Many other scenario studies can be found which examine various impacts of AVs (e.g. VMT, modal share, energy consumption) but do not consider how AVs may affect urban form (e.g.Childress, Nichols, Charlton, & Coe, 2015;Meyer, Becker, Bösch, & Axhausen, 2017;Trommer et al., 2016).
5The forward snowball method involves searching for newer articles which cite any of the studies already on the list of papers to be reviewed. The backward snowball method involves searching for older articles cited by studies on the list of papers to be reviewed.
6A few scenarios which do not closely fit into the classification presented above are not included in the review.
7Time frames in some studies are not explicitly stated.
D. Stead and B. Vaddadi Cities 92 (2019) 125–133
generally assume that car ownership and travel gradually rise (Fulton et al., 2017).
While most of the assumptions are made on the basis of current polices and trends in the cities, others consider the importance of fac-tors such as ownership of vehicles, travel time and new mobility models such ride sharing for evaluation of the impact of automated technology. Of the few scenarios which do mention urban form or land use, the BAU scenario in Heinrichs (2016)assumes that cities still continue to be dominated by cars and are characterised by low-density, fragmented settlement structures.
4.2. Technology + non-shared (T)
The technology and non-shared scenarios assume the introduction of AVs which are either solely or predominantly individually owned and used. Of the literature reviewed, ten of them contain scenarios very similar to this concept (Table 1). Various scenarios assume a gradual roll-out of automated vehicles and relatively widespread adoption of AVs from around 2020, and a rapid growth in AVs around 2025. These scenarios generally assume a continuation of existing trends for vehicle sharing, public transport use, and urban planning (e.g.Bouton et al., 2015; Fulton et al., 2017;Röhrleef et al., 2015). Meanwhile, vehicle ownership does not change significantly as individuals are attached to the ownership of their own cars (Corwin et al., 2015;Gruel & Stanford, 2016;Thakur et al., 2016).
In terms of impacts on land use, these scenarios assume that the demand for transport infrastructure and parking space may require large amounts of additional space in the city. In particular, this would lead to significant increases in the demand for parking in core areas of the city. Due to the low rates of sharing, traffic pressure in the city will increase. This will then result in the formation of urban sub-centres and/or the suburbanisation of some recreational and commercial functions (Tillema et al., 2015).
4.3. Technology + shared (T+)
The technology and shared scenarios assume a future state where automated vehicles are fully developed and the current mobility trends have changed and evolved along with the technology. Various mobility models such as ride sharing platforms, mobility on demand systems and car sharing platforms are expected to operate in cities. Almost all the literature reviewed contains one or more scenarios of this type (Table 2).
In these scenarios, shared vehicle services expand rapidly in terms of geographical coverage and customer segments. Car-sharing is the norm. New forms of shared vehicles and ride-sharing platforms are often included in these scenarios. Many of these scenarios assume that ridesharing is generally seen as a more economical and convenient form of mobility, particularly for short point-to-point movements. As shared mobility serves a greater proportion of local transportation needs, Table 2
Summary of scenarios reviewed.
Source, date Type of approach (forecasting, backcasting or
exploratory) Geographical focus Number of scenarios contained in each source:
BAU T T+ T++ Othera Total reviewed (excluding
‘other’)
Bouton et al., 2015 Exploratory Global – 1 2 – – 3
Chapin et al., 2016 Backcasting USA – 1 – – – 1
Corwin et al., 2015 Exploratory Global 1 1 1 – – 3
Fulton et al., 2017 Forecasting Global 1 1 – 1 – 3
Gruel & Stanford, 2016 Forecasting Global – 1 1 – 1 2
Heinrichs, 2016 Exploratory Germany 1 – 2 – – 3
International Transport Forum,
2015 Forecasting Portugal (Lisbon) – – 1 – 1 1
Milakis et al., 2017 Exploratory Netherlands – 1 – 1 2 2
Pernestål Brenden et al., 2017 Exploratory Sweden – 2 2 – – 4
Röhrleef et al., 2015 Exploratory Germany 1 1 – – 2 2
Thakur et al., 2016 Forecasting Australia 1 1 1 – 1 3
Tillema et al., 2015 Exploratory Netherlands – 1 2 – 1 3
Townsend, 2014 Exploratory USA 1 1 1 – 1 3
Total 6 12 13 2 9 33
a The scenarios in the ‘other’ category are not reviewed in this paper.
Table 3
Issues of urban form contained in the scenarios reviewed.
Source, date How issues of urban form are considered in the study (as an
assumption or prediction) Issues of urban form contained in the study
Bouton et al., 2015 Assumption Density, mixed use, transit-oriented development
Chapin et al., 2016 Prediction Parking area, drop-off space, parking location, roadspace
Corwin et al., 2015 Assumption Parking area
Fulton et al., 2017 Assumption Density (compact city), mixed use
Gruel & Stanford, 2016 Prediction Location of new development (dispersion/sprawl)
Heinrichs, 2016 Assumption Density, location of new development (suburbanisation), parking
area, roadspace
International Transport Forum, 2015 Assumption Parking area, roadspace
Milakis et al., 2017 Assumption Density (compact city)
Pernestål Brenden et al., 2017 Assumption Parking area, roadspace
Röhrleef et al., 2015 Assumption Location of new development (dispersion/sprawl), parking area
Thakur et al., 2016 Prediction Location of new development (inner and outer suburbs)
Tillema et al., 2015 Assumption Parking location, transport transfer points
Townsend, 2014 Assumption Location of new development (sprawl/edge city), transit-oriented
development
D. Stead and B. Vaddadi Cities 92 (2019) 125–133
households begin to reduce the number of cars they own while others abandon ownership altogether.
Many of the scenarios assume a multimodal public transport system alongside highly networked (automated) mass taxi systems as the backbone of urban mobility with the formation of multimodal hubs. However, it is also conceivable that AVs (particularly when shared) may reduce the demand for conventional public transport. Due to the shared ownership of vehicles, a considerable reduction in the land consumption for urban parking spaces may be possible, providing op-portunities for additional development in cities or more open space (see below). At the same time, a growth in low-density suburban develop-ment is envisaged in some scenarios.
4.4. Technology + shared + infrastructure/policy (T++)
Technology, Shared and Infrastructure/Policy scenarios (which as-sume the introduction of AVs which are solely or predominantly shared in conjunction with supportive policies and/or infrastructures to ac-tively promote the uptake and use of AVs) are contained in only two of the literature sources reviewed (i.e.Fulton et al., 2017;Milakis et al., 2017). In these scenarios, shared automated mobility is combined with substantial policy support for electrification, automation, shared-use mobility and urban planning to promote walking, cycling and public transport use.
According toFulton et al. (2017), strong policy support for compact Fig. 1. How AVs may affect road design and urban form.
(Images courtesy ofsehinc.com. Source:Short Elliott Hendrickson Inc., 2018.)
D. Stead and B. Vaddadi Cities 92 (2019) 125–133
cities is needed to avoid significantly higher vehicle kilometres, even in a situation of fairly high levels of vehicle sharing in smaller vehicles. This scenario assumes that government coordination of mobility-as-a-service, linking many transportation options into a seamless network of trip planning and payment via a single interface and increased use of local development impact fees (e.g. charges that account for the nega-tive externalities of car use in order to help fund investment in sus-tainable transport solutions). Other planning policies contributing to the implementation of shared AVs in cities include coordination of re-gional land-use and transportation decisions, increased, ongoing in-vestments in walking, cycling, and public transport infrastructure and systems, implementation of bike and e-bike sharing programs in urban
areas with sufficient density. Meanwhile, one of the scenarios put for-ward byMilakis et al. (2017)supposes that the growth in travel and urban expansion can be curbed when AV technology is combined with road pricing and stricter land use policies (e.g. compact city regula-tions).
5. Content of scenarios – potential impacts on urban form
The potential impacts of AVs on urban form are discussed below in terms of three interrelated issues, each of which encompasses one or more of the 5 Ds proposed byEwing and Cervero (2010): (i) freeing up parking space/reallocating roadspace; (ii) improving the attractiveness Fig. 2. How roadspace could be reallocated after introducing AVs.
(Images courtesy of HOK. Source:Jencek & Unterreiner, 2018.)
D. Stead and B. Vaddadi Cities 92 (2019) 125–133
of the urban environment; and (iii) making more efficient use of transport infrastructure.8In the majority of cases, issues of urban form are mainly discussed in the assumptions used to construct scenarios rather than being outputs or results from the scenarios (Table 3). Nevertheless, there are few studies in which the latter is the case (e.g. Chapin et al., 2016;Gruel & Stanford, 2016;Thakur et al., 2016).
5.1. Freeing up parking space/reallocating roadspace
The scenarios reviewed in this paper often consider that AVs will have a large impact on parking spaces, depending on the number and type of vehicles required. This is related to the type of scenario, parti-cularly in terms of whether vehicles continue to be individually owned and used (as they are in the BAU and T scenarios) or whether they will be shared (as they are in the T+ and T++ scenarios). As such, AVs can have a significant impact on urban spaces and their quality. If in-dividual car ownership decreases and vehicles are shared, more op-portunities may arise for reallocating existing parking spaces. Some spaces could be converted into other uses, while others could be used more efficiently to accommodate vehicles in smaller spaces (e.g. stacked silos) since cars without drivers can be parked more closely together. New parking spaces can therefore be smaller. Moreover, AVs offer the potential to decouple car parking from other land uses since AVs can potentially take themselves away to another part of the city where parking is cheap (or free). As such, the location of parking does not have to be in close proximity to major transport destinations. The relocation of parking space from city centres to the outskirts is possible, thereby freeing up space in the city.
In situations where automated shared and on-demand vehicles are introduced, some of these may be in near continuous operation and have less need to park and may return to depots in less expensive lo-cations where more land is available (e.g. on the urban periphery or industrial areas). This means that the amount of off-street parking could be significantly reduced in city centres. However, a substantial amount of space will still be needed for picking up and dropping off passengers, especially at transport interchanges and major destinations in the city (offices, public buildings, etc.). Clearly, these developments may sub-stantially affect parking revenue for local governments which can re-present a critical and substantial funding source for some local autho-rities (Gühnemann, 2009).
5.2. Improving the attractiveness of the urban environment
Lowering the amount of on-street parking offers more opportunities for reallocating urban space for use by transport modes that are cleaner and/or visually less intrusive (e.g. cycling, walking and mass transit). Freed-up parking space also has the potential to be used for public spaces (e.g. parks, gardens, café seating) to promote more attractive spaces for people. As such, AVs can potentially help to promote more efficient use of land in cities and facilitate new forms of streetscaping (Figs. 1 and 2), thereby creating new opportunities for improving urban attractiveness.
Beyond the opportunities for creating new public spaces, AVs po-tentially offer new opportunities for adding to the building stock where land is freed up either by reducing parking space (see above) or by using transport infrastructure space more efficiently (e.g. narrower
carriageways or shorter headways – see below). In both cases, creating greater diversity of land uses can help to improve the attractiveness of urban areas as well as facilitating shorter travel distances, thereby re-ducing the need for motorised transport.
While several scenarios consider the opportunities that AVs offer for increasing the attractiveness of city centres as places to live, work and play, some scenarios also put forward the idea that AVs may also in-crease the desirability of suburban areas for residential purposes. With AVs, travel times could increase as journeys are no longer considered to be unproductive times of the day. Without having to drive a vehicle, passengers can potentially use journeys for work or recreation purposes and the car becomes a mobile office, library, gym or games room. Thus, people's choices of where to live may become more heavily influenced by factors such as the quality of residential environments than the de-sire to live in close proximity to work (Heinrichs, 2016). If suburban residential environments are considered to be more attractive (in terms of price and/or spatial quality), AVs could contribute to increased suburbanization trends and longer travel distances. In order to address these impacts and avoid large-scale suburbanisation, urban planning tools will be critical in order to increase the quality and attractiveness of residential environments in urban areas and to limit the space for development in suburban and exurban locations. The latter will require strong regulations (i.e. land-use controls) as well as economic instru-ments to reduce the demand for suburban and exurban expansion.
5.3. Making more efficient use of transport infrastructure
AVs provide opportunities to reduce the dimensions of carriageways (and parking spaces – see above) since they are potentially able to operate safely with shorter distances between vehicles, meaning that they can drive alongside each other or behind each other with lower clearances. Moreover, full introduction of AVs also offers the potential for removing some of the existing transport-related hardware from ci-ties including traffic signals, direction and access signs and safety bar-riers. This can have benefits both for the costs of transport infra-structure as well as for the quality of urban streetscapes.
In addition, AVs provide new opportunities for more flexible use of road infrastructure, so that it can be used for different purposes at different times of the day (or week). Examples include pedestrian-only zones or public transport lanes at specific times of the day. Moreover, AVs can also reduce the need for features such as central reservations between carriageways on main roads, resulting in the potential for using the infrastructure more flexibly such as regulating the direction of flow of different lanes according to prevailing traffic conditions. Clearly, using transport infrastructure more efficiently can potentially generate additional traffic, particularly if travel times are reduced and/ or reliability of transport times are increased. Consequently, attempts should not be made to reduce overall travel times for AVs if there is a desire to prevent suburbanisation.
5.4. Gaps in the literature
The review of the scenarios helps to identify various gaps con-cerning the potential impacts of AVs on urban form which is currently an under-examined issue in scenario studies of AVs. The effects on urban form can differ greatly depending on the type of AVs introduced, the extent to which they are shared system and the introduction of supporting policies and infrastructure. The impacts of different types of AVs (from individual pods to driverless trains) on future urban form are scarcely addressed in the scenarios.
Most scenarios assume that AVs systems will have some significant impacts on the land-use patterns of cities. Some speculate more dis-persed low-density land-use patterns surrounding metropolitan regions where AVs result in the further growth of suburbs and even push de-velopment and population into exurban areas (see alsoSoteropoulos, Berger, & Ciari, 2019). On the other hand, others speculate that AV
8The impact labelled ‘freeing up parking space/reallocating roadspace’ per-tains to a combination of Ewing and Cervero's dimensions of urban form, pri-marily their first and third dimensions (i.e. density and design); ‘improving the attractiveness of the urban environment’ pertains to a combination of their second, third and fourth dimensions (i.e. diversity, design and destination ac-cessibility); and ‘making more efficient use of transport infrastructure’ pertains to a combination of their first, third, fourth and fifth dimensions (i.e. density, design, destination accessibility and distance to transit).
D. Stead and B. Vaddadi Cities 92 (2019) 125–133
technology reduces the need for parking space, thereby freeing up space for other uses (see for example Zhang and Guhathakurta, in press). Some estimates suggest that AVs could halve the demand for parking and reduce the demand substantially more if most AVs are shared (Dupuis et al., 2015;Milakis et al., 2017). However, it is still unclear how much urban parking space can be freed for other land uses since significant amounts of space will still be needed for passenger drop-off and collection. Moreover, there are many uncertainties about the spa-tial and temporal distribution of parking demand after AVs are in-troduced and used. Amongst other things, new spatial and temporal patterns in parking demand will be dependent on the degree of access in cities afforded to AVs, and the locations where they are permitted to collect and drop off passengers. Furthermore, there has been little at-tention to the inevitable peaks in transport demand and how these will affect the need for roadspace, collection and drop-off points and parking space.
Clearly, the price, location and availability of parking spaces across different parts of the city will affect the extent to which AVs circulate when not used (i.e. running empty) or reposition themselves in antici-pation of passenger demand later in the day (Papa & Ferreira, 2018). The reallocation of roadspace and parking areas will affect the intensity of land uses. In general, many of the studies tend to devote more at-tention to exploring the small-scale impacts of AVs on urban form, such as parking spaces and carriageway dimensions, than the larger scale impacts such as suburbanisation and the reallocation of parking spaces to other parts of the city which lead to long term implications.
6. Conclusions and recommendations
Many scenarios suggest that fully automated vehicles might be in-troduced into the city within one or two decades. This will have sig-nificant implications for transport and urban planning policies. According to the different scenario studies reviewed, views about the potential consequences of AVs for urban development patterns are di-verse. Some are much more optimistic about the impacts for city centres than others. Others suggest a picture of greater suburbanisation of cities (e.g.Gruel & Stanford, 2016;Heinrichs, 2016).
One possible long-term impact of the adoption of AVs is the de-velopment denser urban cores, more buildings and fewer parking spaces. At the same time, AVs could lead to a greater dispersion of low-density development in metropolitan fringe areas given the ability of owners to engage in other activities while vehicles drive themselves. These impacts may vary depending on how car-dependent a city is (e.g. Bouton et al., 2015). City governments need to have a clear apprecia-tion of where urban land may become available as a result of any re-ductions in parking or infrastructure space after the adoption of AVs. A clear vision for the re-allocation or re-use of this space is needed.
Connections in decision-making processes between long-term mo-bility (location choice) and daily momo-bility (destination and transport mode selection) should be brought into debates surrounding the in-troduction of AVs. The extent to which suburbanisation may materi-alize as AVs are introduced, and how the impacts can be assessed, should be considered as a matter of priority.
The reduction of traffic management measures, combined with improvements in safety, lower emissions and less noise pollution have the potential to improve public realm, making it more attractive and more utilized. The space currently occupied by car parking or made unattractive by transport infrastructure could be put to new and in-novative uses. To a limited degree, this has already begun in cities by introducing the concept of ‘pavements to parks’, which has started to encourage the conversion of roadside parking spaces (sometimes on a temporary basis) into new uses such as gardens and cafes (see also Pernestål Brenden et al., 2017;Townsend, 2014).
The role and position of mass public transport (e.g. trains, trams and metros) is crucial in shaping future urban form in cities as well as the roll-out of AVs. In some scenarios reviewed, AVs act in competition
with mass public transport, such as shared AVs which operate with comparable tariffs and similar speeds as public transport. In other scenarios, mass public transport provides an essential transport back-bone for the city while AVs primarily provide services that link to and from the mass public transport network. In order to achieve the latter, policies will be required to ensure that AVs work in concert with public transport and other highly efficient modes, rather than compete with them (see alsoFulton et al., 2017). Policy options to achieve this might involve restrictions on operations within certain corridors and in-centives to serve stations. Ultimately, the future role of AVs in influ-encing urban form and structure is not so much dependent on the technology and level of automation of vehicles but rather on the reg-ulation of this technology and the governance of cities and regions.
Acknowledgements
The authors wish to thank the three anonymous reviewers for their constructive comments and suggestions on an earlier version of this manuscript. The authors are grateful to Short Elliott Hendrickson Inc. (sehinc.com) and HOK for their kind permission to reproduce the images contained inFigs. 1 and 2respectively.
References
Banister, D., & Hickman, R. (2013). Transport futures: Thinking the unthinkable. Transport Policy, 29, 283–293.
Bartholomew, K. (2007). Land use-transportation scenario planning: Promise and reality. Transportation, 34(4), 397–412.
Bouton, S., Knupfer, S. M., Mihov, I., & Swartz, S. (2015). Urban mobility at a tipping point. Detroit: McKinsey Center for Business and Environment report.http://www. mckinsey.com/business-functions/sustainability-and-resource-productivity/our-insights/urban-mobility-at-a-tipping-point.
Cavoli, C., Phillips, B., Cohen, T., & Jones, P. (2017). Social and behavioural questions
associated with automated vehicles a literature review. London: UK Department for
Transport. http://www.gov.uk/government/publications/social-and-behavioural-questions-associated-with-automated-vehicles.
Chapin, T., Stevens, L., Crute, J., Crandall, J., Rokyta, A., & Washington, A. (2016). Envisioning Florida's future: Transportation and land use in an automated vehicle world. Florida Department of Transportation, final report of FDOT contract BDV30 934-10. Tallahassee: FDOT.
Childress, S., Nichols, B., Charlton, B., & Coe, S. (2015). Using an activity-based model to explore the potential impacts of automated vehicles. Transportation Research Record, 2493, 99–106.
Cohen-Blankshtain, G., & Rotem-Mindali, O. (2016). Key research themes on ICT and sustainable urban mobility. International Journal of Sustainable Transportation, 10(1), 9–17.
Corwin, S., Vitale, J., Eamonn, K., & Cathles, E. (2015). The future of mobility: How
transportation technology and social trends are creating a new business ecosystem. Deloitte
University Presshttp://www2.deloitte.com/content/dam/Deloitte/br/Documents/ manufacturing/Future_of_mobility.pdf.
DuPuis, N., Martin, C., & Rainwater, B. (2015). City of the Future: Technology & Mobility. Washington: National League of Cities. https://www.nlc.org/resource/city-of-the-future-technology-mobility.
Ewing, R., & Cervero, R. (2010). Travel and the built environment: A meta-analysis. Journal of the American Planning Association, 76(3), 265–294.
Fulton, L., Mason, J., & Meroux, D. (2017). Three revolutions in urban transportation: How to
achieve the full potential of vehicle electrification, automation and shared mobility in urban transportation systems around the world by 2050. UC Davis Institute of Transportation studies report to the Institute for Transportation and Development policy. Davis (CA):
University of California. http://www.itdp.org/wp-content/uploads/2017/04/ITDP-3R-Report-FINAL.pdf.
Godet, M. (2001). Creating futures: Scenario planning as a strategic management tool. Paris: Economica.
Gruel, W., & Stanford, J. M. (2016). Assessing the long-term effects of autonomous ve-hicles: A speculative approach. Transportation Research Procedia, 13(2016), 18–29.
Gühnemann, A. (2009). Current practice in funding of urban transport: The case of Germany. Institute for Transport Studies working paper. Leeds: University of Leeds.
Heinrichs, D. (2016). Autonomous driving and urban land use. In M. Maurer, J. C. Gerdes, B. Lenz, & H. Winner (Eds.). Autonomous driving: Technical, legal and social aspects (pp. 213–231). Berlin: Springer.
International Transport Forum (2015). Urban mobility system upgrade: How shared self-driving cars could change city traffic. Corporate partnership board reportParis: ITFhttp://www.itf-oecd.org/sites/default/files/docs/15cpb_self-drivingcars.pdf. Jencek, B., & Unterreiner, J. (2018). People-driven design: Planning for the urban future of
autonomous vehicles. Urban Land Magazine https://urbanland.uli.org/planning-design/people-driven-design-planning-urban-future-autonomous-vehicles.
Kasraian, D., Maat, K., Stead, D., & van Wee, B. (2016). Long-term impacts of transport infrastructure networks on land-use change: An international review of empirical
D. Stead and B. Vaddadi Cities 92 (2019) 125–133
studies. Transport Reviews, 36(6), 772–792.
Litman, T. A. (2019). Autonomous Vehicle Implementation Predictions. Implications for
Transport Planning. Victoria Transport Policy Institute: Victoria.http://www.vtpi. org/documents/evaluation.php.
May, G. (1982). The argument for more future-oriented planning. Futures, 14(4), 313–318.
Meyer, J., Becker, H., Bösch, P. M., & Axhausen, K. W. (2017). Autonomous vehicles: The next jump in accessibilities? Transportation Economics, 62(June 2017), 80–91.
Milakis, D., Snelder, M., van Arem, B., van Wee, B., & Correia, G. (2017). Development and transport implications of automated vehicles in the Netherlands: Scenarios for 2030 and 2050. European Journal of Transport and Infrastructure Research, 17(1), 63–85.
Papa, E., & Ferreira, A. (2018). Sustainable accessibility and the implementation of au-tomated vehicles: Identifying critical decisions. Urban Science, 2(1), 1–14.
Pearman, A. D. (1988). Scenario construction for transport planning. Transportation Planning and Technology, 12(1), 73–85.
Pernestål Brenden, A., Kristoffersson, I., & Mattsson, L. G. (2017). Future scenarios for
self-driving vehicles in Sweden. Integrated Transport Lab report, TRITA-MMK 2017:07.
Stockholm: KTH.http://urn.kb.se/resolve?urn=urn:nbn:se:vti:diva-11890.
Röhrleef, M., Deutsch, V., & Ackermann, T. (2015). Position paper: Scenarios for autono-mous vehicles – Opportunities and risks for transport companies. Position paper November 2015. Cologne: Verband Deutscher Verkehrsunternehmen e.V. (VDV).
RTPI (2015). Urban form and sustainability. RTPI Research Briefing No. 9. March 2015. London: RTPI.http://www.rtpi.org.uk/media/1360966/urban%20form%20and %20sustainability%20briefing.pdf.
Short Elliott Hendrickson Inc (2018). Driverless vehicles set to change the way we design our
roadways? St. Paul (MN): Short Elliott Hendrickson Inc.http://www.sehinc.com/ news/future-what-do-driverless-cars-mean-road-design.
Soteropoulos, A., Berger, M., & Ciari, F. (2019). Impacts of automated vehicles on travel behaviour and land use: An international review of modelling studies. Transport
Reviews, 39(1), 29–49.
Stead, D., & Banister, D. (2003). Transport policy scenario-building. Transportation Planning and Technology, 26(6), 513–536.
Stead, D., & Marshall, S. (2001). The relationships between urban form and travel pat-terns. An international review and evaluation. European Journal of Transport and Infrastructure Research, 1(2), 113–141.
Stone, J., Ashmore, D., Scheurer, J., Legacy, C., & Curtis, C. (2018). Planning for dis-ruptive transport technologies: How prepared are Australian transport agencies? In G. Marsden, & L. Reardon (Eds.). Governance of the smart mobility transition (pp. 123– 137). Emerald Publishing Limited.
Thakur, P., Kinghorn, R., & Grace, R. (2016). Urban form and function in the autonomous era. Proceedings of the Australasian Transport Research Forum, 16–18 November 2016, Melbourne, Australia.
Tillema, T., Berveling, J., Gelauff, G., van der Waard, J., Harms, L., & Derriks, H. (2015).
Driver at the wheel? Self-driving vehicles and transport system of the future. The Hague:
KiM – Netherlands Institute for Transport Policy Analysis.http://english.kimnet.nl/ binaries/kimnet-english/documents/documents-research-publications/2015/10/14/ driver-at-the-wheel/driver-at-the-wheel.pdf.
Townsend, A. (2014). Re-programming mobility: The digital transformation of trans-portation in the United States. Rudin Center for Transtrans-portation Policy & Management, Robert F. Wagner Graduate School of Public Service, New York Universityhttp://www.reprogrammingmobility.org.
Trommer, S., Kolarova, V., Fraedrich, E., Kröger, L., Kickhöfer, B., Kuhnimhof, T., ... Phleps, P. (2016). Autonomous driving. The impact of vehicle automation on mobility
behavior. Munich: Institute for Mobility Research Report, IFMO.http://www.ifmo. de/files/publications_content/2016/ifmo_2016_Autonomous_Driving_2035_en.pdf. Zhang, W., & Guhathakurta, S. (2019). Residential location choice in the era of shared
autonomous vehicles. Journal of Planning Education and Research.https://doi.org/10. 1177/0739456X18776062(in press).
D. Stead and B. Vaddadi Cities 92 (2019) 125–133
