Elsevier

Atmospheric Environment

Volume 226, 1 April 2020, 117399
Atmospheric Environment

The impact of automation and connectivity on traffic flow and CO2 emissions. A detailed microsimulation study

https://doi.org/10.1016/j.atmosenv.2020.117399Get rights and content

Highlights

  • Traditional car-following models provide overestimated instantaneous acceleration values.

  • AVs do not facilitate Traffic Flow, and generate the highest emissions values per kilometer.

  • CAVs increase the capacity of the network and therefore during peak hours, they generate more emissions in absolute values.

  • In total, the differences in emissions per kilometre driven between CVs, AVs, CAVs do not exceed 6%.

  • Results show that the use of vehicle dynamics-based modelling in microsimulation (i.e. MFC free-flow model) is in cases, essential.

Abstract

The interest on the impact of vehicle automation and connectivity in the future road transport networks is very high, both from a research and a policy perspective. Results in the literature show that many of the anticipated advantages of connected and automated vehicles or automated vehicles without connectivity (CAVs and AVs respectively) on congestion and energy consumption are questionable. Some studies provide quantitative answers to the above questions through microsimulation but they systematically ignore the realistic simulation of vehicle dynamics, driver behaviour or instantaneous emissions estimates, mostly due to the overall increased complexity of the transport systems and the need for low computational demand on large-scale simulations. However, recent studies question the capability of common car-following models to produce realistic vehicle dynamics or driving behaviour, which directly impacts emissions estimations as well. This work presents a microsimulation study that contributes on the topic, using a scenario-based approach to give insights regarding the impact of CAVs and AVs on the evolution of emissions over a highway network. The motivation here is to answer whether the different driving behaviours produce significant differences in emissions during rush hours, and how significant is the impact of detailed vehicle dynamics simulation and instantaneous emissions in the outcome. The status of the network is assessed in terms of flow and speed. Furthermore, emissions are estimated using both the average-speed EMEP/EEA fuel consumption factors and a generic version of the European Commission's CO2MPAS model that provides instantaneous fuel consumption estimates. The simulation results of this work show that AVs can deteriorate the status of the network, and that connectivity is the key for improved traffic flow. Emissions-wise, the AVs have the highest fuel consumption per km travelled among other types, while CAVs only marginally lower the overall consumption of human-driven vehicles. For the same traffic demand, the total emissions for different vehicle types remain at comparable levels.

Introduction

The impact of automation-related technologies on road transport networks is a topic studied for many years now (Kesting et al., 2010; Louwerse and Hoogendoorn, 2004). Connected and Automated vehicles (CAVs) are expected to bring significant advancements in the existing road transport systems in terms of reducing traffic congestion, energy consumption, CO2 and pollutants emissions (Alonso Raposo et al., 2017; Litman, 2015).

The picture looks promising, but as research on the field progresses, many researchers express their doubts on the anticipated benefits, either energy or traffic-wise. In Fiori et al. (2019) preliminary results were presented showing that moving from internal combustion engine vehicles (ICEVs) to plug-in electric vehicles (PEVs), the relationship between congestion and energy consumption can change, with higher energy consumption connected to the free-flow test-cases. Mattas et al. (2018) presented a simulation case study on the impact of connected autonomous vehicles, focusing on the possible benefits of connectivity. The results show that depending on the traffic demand, the autonomous vehicles can have adverse effects on traffic flow, while connected autonomous vehicles can be beneficial for the network, depending on their penetration rate.

At the same time, the tools used in various impact assessment studies have known limitations that may affect the final predictions. Results in the work of Ciuffo et al. (2018) raise concerns about the capability of the existing car-following models to reproduce observed vehicles’ acceleration dynamics and thus estimate vehicles emissions and energy/fuel consumption. Most studies focus on congested conditions where the free-flow regime is expected to play a minor role. However, recent studies (Laval et al., 2014; Marczak et al., 2015) highlight that also in congested conditions, the acceleration regime, which incorporates vehicle dynamics and driving behaviour, affects the capacity drop, the hysteresis and possibly other traffic-related phenomena.

Panis et al.(Int Panis et al., 2006) highlighted the need for a detailed analysis of not only average speeds but also other aspects of vehicle operation such as acceleration and deceleration. On the same page, Lejri et al. (2018) proposed a model that accounts for traffic speed dynamics in order to provide more accurate emissions estimations. Finally, in the recent simulation study of Stogios et al. (2019), the importance of considering the different driving behaviors is highlighted.

The prospect of CAVs in reducing the environmental impact of vehicles is of great importance. In this context, it is important to investigate how and to what extent CAVs technologies will affect vehicle energy use and reduce traffic emissions. On the other hand, if the technology does not deliver the expected results, it is essential to identify the correct traffic management strategies to help reach the desired emissions-reduction goals.

The present paper studies the impact of automation and connectivity on future road transport networks, based on certain assumptions regarding the vehicle technology, the traffic supply, and the demand. A microsimulation study on a highway network conducted using AIMSUN traffic simulation software. The results are based on various scenarios with state of the art models for the simulation of different vehicle technologies involved, that is, conventional human-driven vehicles (CVs), AVs and CAVs. The same scenarios were replicated using the same models for the congested part but accounting explicitly the vehicle dynamics on free flow. Different technologies generate different driving behaviours that impact the levels of congestion in the network. The motivation of this work is to answer whether the variation in the driving behaviours produces significant differences in emissions during rush hours. Furthermore, it is important to understand the role of realistic vehicle dynamics simulation and how this impacts traffic flow and emissions.

The network used is the ring road of Antwerp presented in (Mattas et al., 2018). Initially, we build four scenarios using state of the art models for the simulation of the different vehicle types. Each of the four scenarios refers to 3-h simulation with a) CVs, b) AVs, c) CAVs and d) CAVs with 20% increased traffic demand. Furthermore, in order to study whether the explicit simulation of vehicle dynamics can change these results, the above-mentioned car-following models were modified in order to explicitly consider realistic vehicle dynamics. In order to achieve the latter, we introduce the simulation of vehicle dynamics on free-flow, using the Microsimulation Free-flow aCceleration model (MFC) (Makridis et al., 2019a). A publicly available implementation in Python of the last version of the MFC model can be found online (https://pypi.org/project/co2mpas-driver/). The above-mentioned four scenarios are replicated using now the modified car-following models, leading to 8 scenarios in total for this work. Results demonstrate the impact of connectivity and automation on traffic flow and emissions.

The reference models used for the assessment of CO2 emissions estimates are the fuel consumption and CO2 emissions factors proposed by the EMEP/EEA guidebook, and a generic version of CO2MPAS (Fontaras et al., 2018) similar to the one described in (Tsiakmakis et al., 2017). The EMEP/EEA guidebook methodology (European Environment Agency, 2016), is more widely known by its software implementation, COPERT. It foresees an average-speed model and is frequently used in most European countries in order to estimate emissions of all major air pollutants produced by different vehicle categories. CO2MPAS is a vehicle-specific simulation model recently introduced by the EU in its vehicle CO2 certification system (Fontaras et al., 2018). Finally, it should be highlighted that the results of this study are bounded by a set of assumptions, which are summarised a) in the accuracy of the car-following models used for the simulation of different vehicle types, b) the accuracy of the emissions models, c) the accuracy of the vehicle dynamics simulation model and d) the absence of electric vehicle dynamics simulation.

Section snippets

Simulations

This section describes the driver models used in order to simulate the different technologies, the models used to provide emissions estimations and finally, the physical network and the different scenarios.

Results

Simulated data regarding the state of the network were retrieved for 10-min intervals. The results focus on three major dimensions, the impact of the vehicle/driver technology, i.e. CVs, AVs or CAVs, the impact of the simulation of vehicle dynamics, i.e. MFC and finally, the differences in the emissions estimations from EMEP/EEA and the instantaneous generic CO2MPAS.

Fig. 2 illustrates the evolution of the average harmonic speed per 10-min period over the 3-h simulation for different vehicle

Discussion and conclusions

This work proposes a microsimulation framework that investigates the impact of vehicle automation and connectivity in terms of traffic flow and emissions on a realistic highway transport network, the ring road of Antwerp, Belgium. The proposed framework takes into consideration by modelling the different vehicle technologies that generate different vehicle behaviours and traffic patterns. Moreover, it investigates the potential impact of simulating realistic vehicles dynamics through modelling.

CRediT authorship contribution statement

Michail Makridis: Conceptualization, Methodology, Writing - review & editing, Data curation, Formal analysis. Konstantinos Mattas: Conceptualization, Methodology, Writing - review & editing, Data curation, Formal analysis. Caterina Mogno: Conceptualization, Methodology, Writing - review & editing, Data curation, Formal analysis. Biagio Ciuffo: Conceptualization, Methodology, Writing - review & editing. Georgios Fontaras: Conceptualization, Methodology, Writing - review & editing.

Declaration of competing interest

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 (26)

  • B. Ciuffo et al.

    Capability of current car-following models to reproduce vehicle free-flow acceleration dynamics

    IEEE Trans. Intell. Transport. Syst.

    (2018)
  • European Commission

    CO2MPAS: vehicle simulator predicting NEDC CO2 emissions from WLTP — CO2MPAS 2.0.0+9.g5b3144d documentation

  • European Environment Agency

    EMEP/EEA Air Pollutant Emission Inventory Guidebook

    (2016)
  • Cited by (0)

    View full text