Policy recommendations for connected cooperative and automated mobility

This deliverable presents the summary of key findings based on the impact assessment of Cooperative, Connected, and Automated Mobility (CCAM) technologies and services, performed within the Levitate project. Based on these findings this Deliverable identifies key factors with implications for future policy making and recommends areas for deeper consideration to policymakers. To build the background, a brief summary of the impact assessment framework within Levitate and discussion is presented on the vision of the two city project partners ‘Transport for Greater Manchester’ and ‘City of Vienna’.

The details of the results can be found in the following Levitate Deliverables

• 5.2-5.4 (Roussou et al, 2021a; Roussou et al., 2021b; Roussou et al., 2021c),
• 6.2-6.4 (Haouari et al., 2021; Sha et al., 2021; Chaudhry et al., 2021),
• 7.2-7.4 (Hu et al, 2021a; Hu et al, 2021b; Hu et al, 2021c),
• Weijermars et al, 2021
• Deliverables 5.5 (Goldenbeld et al., 2021a), 6.5 (Gebhard et al.,2022), and 7.5 (Goldenbeld et al., 2021b),
• Case Studies documents (Hu et al., 2022, Johannes et al., 2022, Richter, G., 2022, Singh et al., 2022, Haouari et al., 2022),
• Transferability Working Paper (Sha et al., 2022).

Various impacts (studied within Levitate) of CCAM are discussed both under baseline conditions (i.e., with increasing penetration of CAVs without any policy intervention) and then with the implementation of various policy measures. Findings from cost and benefit analyses (D3.4 Hartveit and Veisten, 2022) have also been presented.

The deliverable presents the broader implications of various CCAM related policy measures, the key influencing factors for ensuring the effective and sustainable implementation, and hence, enables the selection of suitable policy options while minimising any adverse impacts.

Key highlights based on the consolidated findings on broader impact dimensions are as follows:

General issues

• CCAM services with similar names and broad approach may have very different impacts depending on the manner in which they are implemented.
• Future CCAM services and technologies may have a mixture of positive and negative societal impacts. Policy measures should be based on a full impact assessment in order to identify improved opportunities to achieve city policy goals or set measures to mitigate negative impacts. Depending upon network characteristics and fleet compositions, the early phases of CAV deployment with a mixed fleet of automated vehicles and vehicles with human drivers in the transport system can result in marginal decrease and in some cases increased conflicts and collisions. Local and national policies will be essential to monitor and mitigate these detrimental impacts during the transition phase.
• As advanced automated vehicles form the largest part of the vehicle fleet, it is anticipated that crash rates will reduce substantially below the current levels. When these vehicles meet or exceed the performance of humans it is expected that traffic impacts may improve beyond existing levels.
• Early generations of automated vehicles, which operate below the level of human driven vehicles with increased headways, highly cautious sensitivity to the detection of other road users– so increased stops - and therefore slower travel and increased delays, are expected to reduce the capacity of cities for traffic. City policies will be required to mitigate these impacts.
• The magnitude of the impacts of CCAM services and technologies is broadly in line with the fleet penetration. Small scale deployments are unlikely to result in a large impact at network level as these impacts remain dominated by the background traffic.
• Several policy measures that have been examined can bring positive environmental impacts; however, powertrain electrification has an overwhelmingly larger impact on emissions compared to the studied policy interventions
• Commonly any improvement in passenger car mobility through the increased automation will have the effect to reduce the use of public transport and active travel. Similarly, improvements in public transport will reduce personal car use and active travel. Automated ride sharing as well as last mile shuttle services are likely to negatively impact active travel with respect to the baseline scenario due to providing pick-ups and drop-offs closest to the origins and destinations of passengers, where last mile shuttles can potentially have much stronger impact on active travel than automated ride sharing.
• Close monitoring of the manner in which CAVs moved, their interactions within the transport network and a calibration of the societal impacts is essential to improve future impact forecasts and to prepare more effective interventions so that city goals can be achieved.
• The Levitate project has shown the benefits of conducting detailed impact forecasts based on a broad spectrum of modelling methods. The methods can be applied to other CCAM interventions and can also be adapted to evaluate real-world trials of CCAM services and technologies.

Economic cost-benefit analysis

All single interventions have been tested in cost-benefit analysis, applied to a hypothetical case area. Although lacking input about the costs of implementing the policy interventions, the following summarises the overall results in net present value (NPV).

• There is a considerable variation in NPV between the interventions and between the various implementation methods.
• Automated urban shuttle services, automated freight delivery and the implementation of GLOSA show routinely positive NPV, given relatively limited costs of implementation.
• Automated ride sharing (ARS) and the replacement of on-street parking show variable NPV results, in the case of ARS the proportion of the total demand and the willingness to share are critical factors.
• The introduction of road pricing within a CAV traffic environment will result in negative NPV; the gains in external environment and health impacts do not outweigh the increased costs to private car users, under the given assumptions.
• Even without policy measures, automation in freight transport will likely gain popularity once the technology is mature and the operating costs become cheaper than the costs nowadays.

Specific interventions

• Road use pricing can be a promising option for improving use of active modes and public transport with increasing prevalence of CAVs. The benefits from Road Use Pricing policy may be slower but will potentially lead to sustainable benefits. It is expected to lead to a number of additional benefits over the baseline impacts: better energy efficiency (dynamic toll more than static toll or empty km pricing), higher vehicle occupancy rate, and lower parking space demand.
• The implementation of Dedicated Lanes for CAVs shows small benefits for traffic until CAVs comprise the majority of vehicles in the fleet. The use of the innermost lane provides the greatest traffic and safety benefits.
• The optimum parking behaviour of CAVs can be managed by adjusting the price of parking. The scenario where a CAV drops passengers off then parks locally minimises impacts on travel time and congestion. Other scenarios where a CAV may return to base or park remotely will increase impacts because of the additional distance travelled.
• CAV parking that is remote from the drop-off location enables on-street parking to be replaced by public spaces or cycle lanes with associated benefits to travel delay and speed.
• Green Light Optimised Speed Advisory (GLOSA) systems in general showed small improvements in traffic impacts when used with fixed time controllers. Increasing the number of GLOSA controlled intersections on arterial roads resulted in small additional improvements in traffic impacts. The impacts need to be carefully assessed when human-driven vehicles comprise the largest proportion of traffic.
• The impact of Automated Rideshare Services depends heavily on the proportion of total demand fulfilled by the service and also the passengers’ willingness to share with others. When fulfilling low levels of demand there are low, adverse impacts on traffic indicators and there are many empty journeys but, when there is a high willingness to share and a large part of the total demand are covered, traffic impacts become positive.
• Under all of the deployment scenarios examined the impacts of Automated Urban Shuttle Services were relatively low as the vehicles routinely formed only a small part of the total fleet. Most societal impacts were positive. However, care should be taken to prevent the anticipated unwanted impacts of these services, for example on the use of active travel modes. Anticipatory research and anticipatory and flexible planning approaches are recommended to prevent these negative developments.
• Freight vehicles also tend to be a small proportion of the total fleet nevertheless Automated Urban Freight Delivery services provide many positive benefits. Automated freight vehicles that enable night-time deliveries to be made produce additional benefits to travel time and congestion. Automation alone will most likely lead to an increase in freight mileage (because of smaller and cheaper freight vehicles), so corresponding policy measures in favour of freight consolidation should be considered to mitigate this trend. Fortunately, automation is expected to facilitate the consolidation process.
• A focused assessment of the impact on bridges of truck platooning has identified the need to improve the structural resistance of bridges over 55m span in bending and over 60m span in shear. Alternatively, increased forward headways must be imposed.

Other key remarks

• To govern new forms of smart mobility and automated urban transport, public authorities will need to cooperate with many new partners and assume new roles in the process of governance. Although many ideas and plans for new forms of mobility may come from private companies, public authorities should promote preferred directions of innovation by setting up strategic agendas and by establishing suitable standards, regulations and guidelines.
• However, care should be taken to prevent the anticipated unwanted impacts of these services, for example on equal accessibility of travel and on the use of active travel modes. Anticipatory research and anticipatory and flexible planning approaches are recommended to prevent these negative developments.
• Given the potential that increasing automation may attract part of public transport users and/or pedestrians/cyclists to switch to a private automated vehicle it is recommended that city planners and managers enhance the public transport network, by providing point-to-point Automated Urban Shuttle Services as well as on-demand AUSS, in order to promote the reduction of the use of private cars.
• Clear communication to transport users and other road users is necessary to clearly explain new transport operations, to explain what users and other road users can expect and to prevent idealised expectations. The effectiveness of specific interventions may be very sensitive to changes in mobility behaviour.
• In decisions about new forms of automated transport, waiting time, travel time, travel costs, comfort, safety and security should play a dominant role in setting policy goals, as these are likely to determine long-term and wider acceptance once the novelty value wears off.
• In future projects the long-term planning of successive implementation phases is recommended, for example going from operator to remote operator operations, and from simple to complex traffic environments.
• Although new forms of automated urban transport may be operated and controlled by private companies, it is recommended that these are developed to complement the public transport system in useful ways, for example by providing their services in regions not served by the public transport, usually outside the city center, or by providing automated shuttles connecting different existing public transport stations.
• Guidelines - including ethical guidelines - and lists of impacts for future automated urban mobility and transport have been formulated, within LEVITATE and generally by the transport research community, and should be partly or fully adopted in strategic plans to facilitate successful implementation of new transport services.
• Multimodality and synchro modality are important factors to aim towards a sustainable logistic supply chain.
• All the above points require homogenous and shared data among operators, which is perhaps the most difficult challenge due to the competition between service providers as well as freight operators.

The many different scenarios of CCAM, the many different potential policy options and its interdependencies show a very complex pattern of effects. However, the effects of CCAM on cities and society largely depend on the regulatory framework in which CCAM is deployed. It is up to policy makers to define a regulatory framework supporting the goals of the respective Smart City Strategies, SUMPs (Sustainable Urban Mobility Plans), Climate Strategies etc. while avoiding adverse effects.

LEVITATE has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No 824361.