Simulating long-term emissions from private automated vehicles under climate policies

https://doi.org/10.1016/j.trd.2023.103665Get rights and content

Highlights

  • Simulates uptake of private light-duty automated vehicles (AVs) and GHG impacts.

  • Consumer uptake of AVs ranges from 15% to 36% of new market share by 2035.

  • Full availability of AVs leads to a small increase in GHG emissions.

  • GHGs increase mainly due to increased vehicle travel (VKT) and reduced zero-emission vehicle sales.

  • Strong climate policy (regulation or tax) can still reduce AV emissions.

Abstract

The future of privately owned, fully automated vehicles (AVs) is highly uncertain, especially the impacts to greenhouse gas (GHG) emissions. We simulate the impacts of climate policy on AV uptake in Canada’s light-duty vehicle sector and the corresponding GHG emissions, including changes to efficiency, vehicle kilometre travelled (VKT), and zero-emissions vehicle (ZEV) sales. We use a technology adoption model which includes consumer preferences and endogenous learning, while also representing automaker decisions. In scenarios where AVs become fully available for sale in 2025, consumer uptake ranges from 15% to 36% new market share by 2035--some of which displace ZEV sales. With or without strong climate policy, the availability of AVs leads to a small increase in GHG emissions, largely due to increased VKT per vehicle and among new user groups. However, enactment of a strong carbon price or ZEV sales mandate can still substantially reduce emissions in the AV scenarios.

Introduction

Widespread uptake of fully automated vehicles (AVs) may significantly transform the transport system in the coming years (Bansal and Kockelman, 2017). In this study we simulate the potential impacts of AV availability (starting in 2025) on light-duty vehicle greenhouse gas (GHG) emissions out to 2035. We use the term AVs (short for fully automated vehicles) to refer to vehicles with an automation level of 4 (or higher) as per the definition used by the US National Highway and Traffic Safety Administration (2022). Others might call these autonomous or self-driving vehicles—but we use the term “fully automated”. We assume that any drivetrain (powered by gasoline or electricity) can be fully automated with the use of LIDARs and cameras and other required software.

Apart from the wide-ranging impacts on road safety, automobility, and travel behaviour, large scale AV usage could also have significant energy and GHG impacts (Wadud et al, 2016). Of course, there is enormous uncertainty regarding AV deployment, including availability and consumer adoption. Of particular interest in this paper, there are a variety of ways that AVs can impact GHG emissions, including changes in driving efficiency (or inversely, energy intensity), vehicle sizing, traffic congestion, and vehicle travel demand for existing and new user groups (vehicle km travelled or VKT). The collective understanding of these future trends and impacts is still in its infancy (Harb et al, 2021).

Simultaneously, governments across the world are implementing climate policies to reduce GHG emissions from the road transport sector. For example, our case study of Canada has implemented a mix of climate policies to achieve net zero emissions by 2050, including carbon pricing and regulations such as a low-carbon fuel standard and vehicle emissions standard (Government of Canada, 2021a). In line with this longer-term objective, many jurisdictions have announced goals for light-duty vehicle sales to be 100% zero-emissions vehicles (ZEVs) by 2035, such as in Canada, the UK, the Netherlands, Norway, and the US state of California. A subset of these regions (in the US and Canada) have translated these goals into a binding ZEV sales mandate (Axsen et al., 2022). This push towards deep GHG reduction goals and climate policies leaves many unanswered questions about how AVs might interact with climate policies and ZEVs, and about their resulting GHG impacts.

Within this context, we presently explore the following research questions, focusing on the case of Canada’s light-duty vehicle sector:

  • 1.

    If AVs become available for sale, what sales rates are expected with and without climate policy (carbon pricing and regulation)?

  • 2.

    How does AV availability impact ZEV sales (starting in 2025)?

  • 3.

    What are the GHG emissions impacts of AVs with and without climate policy, when considering potential changes to efficiency and vehicle travel (VKT)?

In this study, we use AUM, a technology adoption model that simulates the uptake of light duty vehicles in Canada—which we adapt to represent AV technology. By ‘simulating’, we mean projecting future consumer adoption of new and emerging technologies under a certain set of behaviour, technology, and policy assumptions. In this case we study the uptake of AVs, ZEVs, and vehicles that combine both features. Technology adoption models are often associated with the larger category of energy-economy models. Within the literature on technology adoption and energy-economy models, some focus only on conventional internal combustion engine vehicles (e.g. Small, 2012), and some others focus on both internal combustion engine vehicles and ZEVs (e.g. Fox et al., 2017, Wolinetz and Axsen, 2017). ZEVs are vehicles that have the potential for zero tailpipe emissions, namely plug-in hybrid (PHEVs), battery-electric (BEVs), along with fuel-cell electric vehicles (FCEVs). Our study extends this literature space by including AVs, in addition to including internal combustion engine vehicles and ZEVs. Though, since the uptake of light-duty fuel cell vehicles globally remains very low, we exclude this drivetrain technology in our current study.

As discussed in greater detail in the literature review section, this study is novel on several accounts. First, in contrast to the few previous attempts of estimating long-term adoption of AVs (e.g., Bansal and Kockelman, 2017, Talebian and Mishra, 2018), we use a technology adoption model which includes behaviourally realistic representations of consumers, including dynamic consumer preferences, and endogenous representation of automakers including technology learning. Second, this model simulates demand- and supply-side dynamics to allow us to endogenously break down the GHG impacts of AVs into several constituent categories, such as efficiency (or inversely energy intensity), VKT response, and interactions with ZEV adoption. In contrast, previous literature estimating the GHG impacts of AVs tends to rely on exogenous assumptions and calculations (e.g., Wadud et al., 2016, Milakis et al., 2017), or else uses optimization models that do not represent realistic consumer preferences (e.g., Jones and Leibowicz, 2019, Brown and Dodder, 2019, Sheppard et al., 2021). Third, our application includes a novel analysis of the potential interactions between ZEV supportive policies and AV adoption, which is missing from existing literature.

This current study focuses on private AVs. Some researchers argue that future scenarios with widespread usage of shared AVs (SAVs or “robo-taxis” in some form) are feasible and could have positive impacts on society, including dramatic reductions in GHG emissions (Sperling et al., 2018, Vilaça et al., 2022). However, emerging empirical evidence suggests that the majority of consumer segments prefer private AVs over SAVs (Haboucha et al., 2017, Nazari et al., 2018, Saleh and Hatzopoulou, 2020, Saeed et al., 2020), and researchers agree that there is great uncertainty over whether (and under what conditions) consumers will choose private AVs or SAVs (Wadud and Chintakayala, 2021). While a future of SAVs that displace private vehicle ownership is technically possible, we presently focus on the study of AV uptake in the private light-duty vehicle sector.

In short, our present research objectives are to simulate the adoption of privately owned AVs in the Canadian light duty vehicle sector, using a model that represents consumer behaviour, supplier decision-making, and climate policy. Using the time horizon to 2035, we simulate AV new sales each year and changes in light-duty vehicle stock, and estimate the GHG impacts of AV adoption. To simulate the effects of policy on GHG emissions from AV deployment, we consider scenarios with baseline policies, and with added “strong” carbon pricing or ZEV regulation. We also represent uncertainty in numerous parameters, and report the sensitivity of results to these parameters.

The paper is structured as follows. Section 2 reviews the literature on AVs, Section 3 describes the model used for the study, Section 4 presents the results, and Section 5 concludes with a discussion on the results.

Section snippets

Literature review

Adapting the classification from Zhang and Zhang (2022), AV-related literature can be categorized according to at least four sub-themes (though with significant overlap), namely:

  • i.

    survey studies assessing the consumers’ willingness to pay for AVs, or factors affecting consumers’ acceptance of AVs;

  • ii.

    studies on the impact of shared AVs on urban road traffic, trip generation, mode and route choice and road networks;

  • iii.

    studies on (long-term) adoption and diffusion of AV technology; and

  • iv.

    studies on energy

The AUtomaker-consumer model (AUM)

We use the AUtomaker-consumer Model (AUM) to simulate the impacts of AVs on Canada’s light-duty vehicle sector. AUM is a technology adoption model which simulates long-term (from 2020 to 2035) projections of new light-duty vehicle sales (including ZEVs), total light-duty vehicle stock, and the corresponding GHG emissions. AUM is unique in that it includes behaviorally-realistic consumers (also accounting for consumer preferences, rather than just financial cost considerations in technology

AV market share and its interaction with ZEV uptake

Fig. 3 depicts median new AV market share under the three policy scenarios (Baseline, Baseline + Tax, Baseline + ZEV), where the uncertainty in market share estimates is depicted by the optimistic and pessimistic values in Table 9. Under the Baseline scenario with median values, new AV market share (starting in 2025) is expected to be 12.7% in 2025 and rising to 24.1% by 2035.

Relative to the Baseline, the market share of AVs is marginally lower under the climate policy scenarios (Baseline + Tax

Discussion and conclusions

In this study, we attempt to answer the following research questions: i) If AVs become available for sale, what sales rates are expected with and without climate policy (carbon pricing and regulation)? ii) How does AV availability impact ZEV sales? iii) What are the GHG emissions impacts of AVs with and without climate policy?

For our research we use AUM, a light-duty vehicle technology adoption model which includes dynamic consumer preferences, endogenous technology learning, consumer

Author contribution statement

The authors confirm contribution to the paper as follows: study conception and design: Chandan Bhardwaj; Jonn Axsen; Curran Crawford; data collection (literature review), model development and analysis: Chandan Bhardwaj; draft manuscript preparation: Chandan Bhardwaj, Jonn Axsen, Curran Crawford; All authors reviewed the results and approved the final version of the manuscript.

Declaration of Competing Interest

Funding was provided by the Community Trust Endowment Fund (CTEF) of Simon Fraser University, and the Social Sciences and Humanities Research Council (SSHRC) of Canada. The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

References (96)

  • S. Djavadian et al.

    Multi-objective eco-routing for dynamic control of connected & automated vehicles

    Transp. Res. Part D: Transp. Environ.

    (2020)
  • R. Elvik

    The demand for automated vehicles: A synthesis of willingness-to-pay surveys

    Econ. Transp.

    (2020)
  • J. Fox et al.

    Picking winners: modelling the costs of technology-specific climate policy in the US passenger vehicle sector

    Ecol. Econ.

    (2017)
  • K.M. Gurumurthy et al.

    Modeling Americans’ autonomous vehicle preferences: A focus on dynamic ride-sharing, privacy & long-distance mode choices

    Technol. Forecast. Soc. Chang.

    (2020)
  • C.J. Haboucha et al.

    User preferences regarding autonomous vehicles

    Transport. Res. Part C: Emerg. Technol.

    (2017)
  • M. Harb et al.

    What do we (Not) know about our future with automated vehicles?

    Transport. Res. Part C: Emerg. Technol.

    (2021)
  • C.D. Harper et al.

    Estimating potential increases in travel with autonomous vehicles for the non-driving, elderly and people with travel-restrictive medical conditions

    Transport. Res. Part C: Emerg. Technol.

    (2016)
  • R. Jagannathan et al.

    Why do firms use high discount rates?

    J. Financ. Econ.

    (2016)
  • J. Ji et al.

    Trip energy consumption estimation for electric buses

    Commun. Transport. Res.

    (2022)
  • E.C. Jones et al.

    Contributions of shared autonomous vehicles to climate change mitigation

    Transp. Res. Part D: Transp. Environ.

    (2019)
  • V. Kolarova et al.

    Impact of trust and travel experiences on the value of travel time savings for autonomous driving

    Transport. Res. Part C: Emerg. Technol.

    (2021)
  • V. Kolarova et al.

    Assessing the effect of autonomous driving on value of travel time savings: A comparison between current and future preferences

    Transp. Res. A Policy Pract.

    (2019)
  • C. Kormos et al.

    Latent demand for zero-emissions vehicles in Canada (Part 2): Insights from a stated choice experiment

    Transp. Res. Part D: Transp. Environ.

    (2019)
  • Z. Le Hong et al.

    Air quality and greenhouse gas implications of autonomous vehicles in Vancouver, Canada

    Transp. Res. Part D: Transp. Environ.

    (2021)
  • T. Leicht et al.

    Consumer innovativeness and intentioned autonomous car adoption

    J. High Technol. Managem. Res.

    (2018)
  • Y. Liu et al.

    Modelling the energy consumption of electric vehicles under uncertain and small data conditions

    Transp. Res. A Policy Pract.

    (2021)
  • F. Liu et al.

    Can autonomous vehicle reduce greenhouse gas emissions? A country-level evaluation

    Energy Policy

    (2019)
  • A. Manfreda et al.

    Autonomous vehicles in the smart city era: An empirical study of adoption factors important for millennials

    Int. J. Inf. Manag.

    (2021)
  • F. Mensing et al.

    Trajectory optimization for eco-driving taking into account traffic constraints

    Transport. Res. Part D: Transport Environ.

    (2013)
  • D. Milakis et al.

    Policy and society related implications of automated driving: A review of literature and directions for future research

    J. Intell. Transp. Syst.

    (2017)
  • F. Nazari et al.

    Shared versus private mobility: Modeling public interest in autonomous vehicles accounting for latent attitudes

    Transport. Res. Part C: Emerg. Technol.

    (2018)
  • S.M. Pampel et al.

    How I reduce fuel consumption: An experimental study on mental models of eco-driving

    Transport. Res. Part C: Emerg. Technol.

    (2015)
  • A. Pani et al.

    Evaluating public acceptance of autonomous delivery robots during COVID-19 pandemic

    Transp. Res. Part D: Transp. Environ.

    (2020)
  • P. Penmetsa et al.

    Perceptions and expectations of autonomous vehicles–A snapshot of vulnerable road user opinion

    Technol. Forecast. Soc. Chang.

    (2019)
  • S. Pettigrew et al.

    Clusters of potential autonomous vehicles users according to propensity to use individual versus shared vehicles

    Transp. Policy

    (2019)
  • X. Qu et al.

    Automation and connectivity of electric vehicles: Energy boon or bane?

    Cell Reports Phys. Sci.

    (2022)
  • A. Rahimi et al.

    Adoption and willingness to pay for autonomous vehicles: attitudes and latent classes

    Transp. Res. Part D: Transp. Environ.

    (2020)
  • T.U. Saeed et al.

    An empirical discourse on forecasting the use of autonomous vehicles using consumers’ preferences

    Technol. Forecast. Soc. Chang.

    (2020)
  • M. Saleh et al.

    Greenhouse gas emissions attributed to empty kilometers in automated vehicles

    Transp. Res. Part D: Transp. Environ.

    (2020)
  • K.A. Small

    Energy policies for passenger motor vehicles

    Transp. Res. A Policy Pract.

    (2012)
  • C.A. Spurlock et al.

    Describing the users: Understanding adoption of and interest in shared, electrified, and automated transportation in the San Francisco Bay Area

    Transp. Res. Part D: Transp. Environ.

    (2019)
  • C. Stogios et al.

    Simulating impacts of automated driving behavior and traffic conditions on vehicle emissions

    Transp. Res. Part D: Transp. Environ.

    (2019)
  • A. Talebian et al.

    Predicting the adoption of connected autonomous vehicles: A new approach based on the theory of diffusion of innovations

    Transport. Res. Part C: Emerg. Technol.

    (2018)
  • I. Tsouros et al.

    Who will buy alternative fueled or automated vehicles: A modular, behavioral modeling approach

    Transp. Res. A Policy Pract.

    (2020)
  • M. Vilaça et al.

    Life cycle assessment of shared and private use of automated and electric vehicles on interurban mobility

    Appl. Energy

    (2022)
  • Z. Wadud et al.

    To own or not to own–That is the question: The value of owning a (fully automated) vehicle

    Transport. Res. Part C: Emerg. Technol.

    (2021)
  • Z. Wadud et al.

    Help or hindrance? The travel, energy and carbon impacts of highly automated vehicles

    Transp. Res. A Policy Pract.

    (2016)
  • Z. Wadud et al.

    Fully automated vehicles: A cost-based analysis of the share of ownership and mobility services, and its socio-economic determinants

    Transp. Res. A Policy Pract.

    (2021)
  • Cited by (0)

    View full text