The development of autonomous vehicles [AV] is a disruptive technology for the whole transport sector. Most developments in this field have concentrated on the private car. The benefits of AVs are multiple, most commonly they are associated with higher safety, lower congestion, fewer crashes, higher fuel efficiency and declining human resource costs.
The associated benefits of AV technologies in goods transport can be categorised into three groups: (1) traffic related gains (lower travel time, shrinking costs, less traffic), (2) economic (financial benefits for transport companies, e.g. lower costs, restructuring of market), (3) safety and environment. These or similar effects have been widely analysed with regard to passenger cars (Milakis, van Arem & van Wee 2017). But in the case of goods transport there are other factors which need to be taken into account.
The term autonomous vehicle has a wide variety of interpretations. Most commonly it refers to private cars driving without driver interaction. But there are other forms of automated driver interaction such as adaptive technologies which overrule a driver’s decision and/or are giving recommendations. Also, there are technologies for following a lead driver in convoy. Autonomous vehicles can be completely unmanned or supervising personnel can be present for a safety and/or maintenance role. One needs to analyse if and when full automation would be the most efficient development or if there could be situations in the case of goods transport where a lower degree of automation would also be sufficient.
The literature and research is mainly focuses on personal AVs and much less on freight transport. But the latter can also have a vital role in the future, and the economic and social effects can be even greater. New ways of delivery can speed up transport and reduce its costs. Also, with personal AVs a change in business mode from full ownership to shared ownership is highly likely (Masoud & Jayakrishnan 2015). In freight traffic there is no need for such changes and in the business-to-business market new technologies are easier to adapt and there are already established financing capabilities for higher value investments.
It has been shown that growth in global trade and economic output is connected to declining transport costs (Baier & Bergstrand 2001). The spread of containerised traffic and the shrinking costs of intercontinental traffic played a major role in the globalisation of the economy and contributed to the emergence of globally integrated supply chains. The understanding of automated transport systems could be vital for the understanding of the growth potential and how it can affect the future of the global economy.
In this paper the automation of road, rail, and urban transport will be analysed by a review of available research results and data which can guide the development of future scenarios of transport automation. The possible economic and social aspects of AVs will also be evaluated.
The emergence of AVs sooner or later is awaited by science and industry, but there are still at least two major challenges ahead before rollout on a mass scale. The first one is an ethical issue, that is - how to programme the vehicles. In the case of a collision it has to be decided whom and what to protect: the drivers, the vehicles or to protect the highest number of humans? If freight vehicles could commute without drivers, and in first priority they had to protect humans, this could lead to the fact that operators would not actually see too much of a decline in accident-related costs.
Secondly, the current legal framework has also to be upgraded to be able to deal with the AVs, the liabilities of producer and customer are not regulated sufficiently adequately, which poses a great business risk for transport companies (Buning & de Bruin 2017). The legal framework for the transport industry involves other issues about reliability and international laws and regulations have to be upgraded to include issues arising due to the AVs. It can include INCOTERMS rules on how to load and unload goods in a fully automatic system, or International Union of Railways [UIC] regulations regarding railways. The implementation of these industry specific complex regulations will involve a long process. But before these changes happen it is worth determining in which transport mode can AVs be economically, socially and environmentally viable.
The study includes a literature review for road, rail and urban AV freight transport. Its methodology includes data collection on social, safety and economic aspects of automation and includes estimations of the impacts of AV technology. Data has been used on a global scale for transport volumes and scenario analysis, and the economic impact was analysed at the level of a single representative firm. Finally the environmental impacts – specifically CO2 emissions – are calculated for freight transport and the impact scenarios for AV technology are analysed.
Automation of road transport
Most of the car and truck industry is promising self-driving cars and trucks on the road in the years to come. As S. Hassler (2017) discovered, Ford, Google, Mercedes-Benz, Tesla, and Uber have declared that by 2021 this could be a reality.
Vehicles currently on the roads are already very intelligent and are equipped with Advanced Driver Assistance Systems (ADAS). This equipment can not only help to prevent accidents but as J. Zolock et al. (2016) have pointed out, this is a crucial way to collect data to further improve ADAS systems. These already feature most of the technologies that will be used in AV but more data is needed to be able to develop systems that are more reliable than a human not just on average, but in all situations. And currently this is the main challenge: in ideal situations AVs perform better but in the case of bad weather or in the case of very bad road conditions and/or insufficient infrastructure quality, they are not yet reliable enough. AVs should be beneficial: D. J. Fagnant and K. Kockelman (2015) found that in the case of US the savings can be as high as 2 000 to 4 000 USD per year per vehicle. Yet, they see that within the US the different regulations in states and high production costs are hindering the spread of this technology.
There are already numerous intelligent technologies on trucks which are autonomously overwriting driver control. This is not a new phenomenon as such technologies as ABS have been around for decades. But the complexity and the cardinality of these technologies are already making such vehicles almost autonomous, for instance on highway cruise mode. The following features are common in current trucks:
- –adaptive cruise control and collision warning with emergency brake in practice makes it possible for trucks to travel in convoys without active driving;
- –stretch brake to prevent jack-knifing;
- –Lane Keeping Support prevents trucks leaving their lanes, especially on highways;
- –Driver Alert Support monitors driver behaviour and advises when to take a break;
- –roll control assists in adjusting air suspension to adjust cargo height to road quality;
- –stability control assists in maintaining the stability of truck and trailer.
These technologies are already close to autonomous driving, but their integration is still a lengthy process.
Most ADAS systems target safety. This is a very stressful goal as heavy goods vehicles are overrepresented in serious accidents and fatalities when compared to their number and mileage. In the United States, for example, they make up 3% of all vehicles, they drive 7% of all miles, but they are involved in 11% of all road fatalities (Bezwada 2010). Truck safety is not only dependent on the vehicle itself: L. Mooren et al. (2014) reviewed 42 articles on truck safety and found that organisational safety policies and safety management systems are the most important factors in raising safety levels. Necessary maintenance and its safeguarding cannot be taken over by AVs.
It is not easy to predict how much impact AVs will have on safety and the reduction of accidents. S. Newnam & N. Goode (2015) analysed 27 highway crash reports and found that the truck driver’s fault was not the main problem. The conclusions of the study are also interesting in the light of the emergence of AVs.
In the European Union there is a decade of experience of research on specialised autonomous driving: trucks driving in convoys. The technology – also called platooning – involves a lead truck driven by a human driver and 2 to 4 other trucks following this vehicle automatically so that their drivers can spend their rest time. This way the utilisation of trucks can be higher, as drivers can drive more hours daily. This is anticipated to occur since trucks can travel 8-10 metres from each other (Turri, Besselink & Johansson 2016).
This has various advantages but does not require such huge investments as full automation. The Safe Road Trains for the Environment project started in 2009 (Safe Road Trains… n.d.). In the project a convoy of 5 vehicles has been tested: a lead truck with a human driver, a following truck, and 3 following cars. The outcome was that a truck could save up to 2.8 tons of CO2 equivalent per year and a car up to 0.1 tons. Heavy goods vehicles (>3.5 tons, HGV) can save up 2.7% to 5.3% fuel which reduces costs (Lammert et al. 2014). Safety barriers on motorways have been identified as a weak point in this technology: these could not bear the load of the impact of vehicles travelling close to each other.
From the results of platooning research, in 2017 the European automotive industry gained enough experience to come up with a detailed roadmap for the introduction of semi-automated convoys by 2025 (European Automobile Manufacturers Association 2017). In 2017 the Department for Transport and Highways England gave the green light to testing platooning on UK motorways from 2018 on. It can be assumed that this semi-automatic technology will be rolled out in Europe in the period up to 2023 and that full automation will require a much longer timeframe.
HGV drivers and passengers are relatively safe from accidents: according to the European Road Safety Observatory (European Commission 2016) database in 2014 there were only 499 fatal casualties in the EU, down from 1,126 in 2005. The equivalent values for trucks weighing less than 3.5 tons are 774 and 1,194. But 3,850 people died in accidents involving HGVs and the fatality rate was 7.7 per million inhabitants per year. Trucks are involved in many more accidents; it was estimated that 100 thousand people yearly are injured in accidents which involve HGVs.
The European Truck Accident Causation (n. d.) study found that in 85.2% of cases human factors are the main cause of accidents, in 5.3% the vehicle, 5.1% the condition of the infrastructure, and in 4.4% the weather conditions. If we assume that 90% of the human factors could be overcome by AVs, 77 thousand accidents and 3 thousand fatalities could be overcome annually in the EU.
The next step towards further automation would require more advanced, more reliable and faster vehicle-to-vehicle communication. In practical terms this means the existence of a 5G mobile network with full coverage. As the finalisation of the standardisation of this technology is not expected before 2020 and the development and roll-out will continue at least until the middle of the 2020s (Rost et al. 2016) autonomous driving on a mass scale cannot be expected before that.
T. J. Gordon & M. Lidberg (2015) pointed out that there are various feasible technologies that are available for the development of the efficiency and safety of road traffic and it could be more beneficial to concentrate on these rather than full autonomous driving. In the case of trucks, platooning on highways could be such a solution or even an AV mode for highways and conventional driving on land and city roads. After leaving highway exits automation seems to be challenging: going through highly populated areas, treating unexpected pedestrians, animals or even improper drivers, finding industrial parks and companies which are pretty commonly not well marked on digital maps, dealing with traffic control, police, illegal migration, safeguarding security of the cargo, fixing minor technical defects, and loading and unloading cargo are challenges which automation cannot fix, all will need human interaction.
Economic effects of AVs in freight transport
Reducing staff costs and raising efficiency are required to compensate for investment in the new technology. But there will also be social effects of AVs if employment in the transport sector will reduce dramatically. To understand the magnitude of this, we have to understand that there are 1.5 million people working in the road transport industry in the US, 3.0 million in the EU and 3.9 million in China as can be seen in Table 1. Of these – which also include railways, storage and warehousing – 40% are employed in road transport, where the bulk of the jobs are truck drivers.
Employment statistics for transport in the major economies
Sources: own study based on data by: Eurostat, World Bank, National Bureau of Statistics of China, Bureau of Labour Statistics (US), Ministry of Statics and Programme Implementation (India)
|Employed in transport and storage (million)||Employment in road transport (million)||Total population (million)||Share of population working in transport||Share of population working in road transport|
If we consider the 4 main economies which have around half of the total global population we can see how important the transport sector is socially: 1.1% of all people are working in it, and 0.4% in road transport. With the different employment rates it can be stated that 1 of every 100 workers is most probably a truck driver. Therefore, AV in road transport potentially threatens 15.3 million jobs.
For road transport currently the main constraint on improving efficiency is that drivers can work a maximum of 9 hours daily – but regulations are different in most of the countries. The 9 hour limit is especially restrictive on long routes, it is less important in urban and sub-urban collection and delivery. If trucks could travel 24/7 that would mean that their average travel range per day could be 2-3 times higher.1 This would make road traffic considerably cheaper, which is good news for trucking companies.
An important question concerns the costs that trucking companies would be able to pay for the AV technology. Waberer’s was taken as an example as a wide range of financial data is publicly available about the corporation and it is one of the biggest trucking companies in Europe. The efficiency of an autonomous truck could grow from 9 to 23-24 hours a day, so theoretically 39% of trucks would be sufficient for the same traffic (Tab. 2).
Financial data of Waberer’s International Inc. (2016) and AV scenarios, values in euros
Remarks: EBIT stands for earnings before interest and taxes which represents the profit of the corporations
|0: Current model||1: Partial automation||2: Full automation|
|Cost of trucks||76,680,000||43,132,500||30,005,217|
|Additional AV costs||0||21,566,250||30,005,217|
|Office and other||914||1,154||1,393|
|Million km / year||450||450||450|
|Number of trucks||3,550||1,997||1,389|
|Driver’s salary / year / truck||20,483||18,207||0|
|Yearly leasing cost / truck||21,600||21,600||21,600|
|AV costs / truck / year||0||10,800||21,600|
|Revenue / year / truck||161,226||286,624||412,022|
|Km / truck / year||126,761||225,352||323,944|
|Km / truck / day||347||617||888|
|Operation / day (hours)||9||16||23|
In the 1st scenario partial automation was considered, like platooning. Half of the truck drivers could be dispensed with, but an additional 10% of office workers are needed. The cost of the AV technology is estimated at 50% of the leasing costs of the truck. The daily working hours were estimated at 16 hours as business hours will restrict the operation – e.g. one cannot deliver in the middle of the night. In the 2nd scenario full automation was modelled where there are no drivers, but 20% more office workers are needed, and the cost of the AVs is the same as current truck prices. If automation will be at such a high level, delivery can be automated too and a daily 23 hours of running time could be theoretically achieved although in reality this will be influenced by several factors, e.g. the dispersion of the need for transport services. This scenario can be seen as a theoretical maximum. We can conclude that there is a very viable business opportunity to introduce self-driving trucks. This simplified cost and benefit calculation shows that AVs in road transport have a high economic potential if all legal and technical challenges can be overcome.
Automation of rail transport
Autonomous driving is already a reality in railway traffic: in Australia Rio Tinto completed the first fully autonomous rail journey without an engine driver on board in October 2017 (Rio Tinto completes… 2017). The company has been using autonomous locomotives since the beginning of 2017, but there were drivers present for security reasons. In Europe there are new projects for testing the technology: in Austria a 25 km test section has been allocated to do that on a track which is used for both passenger and freight traffic (Test track for… 2017). In Germany, the national rail operator Deutsche Bahn [DB] is testing a fully autonomous hump locomotive for shunting operations (DB presses ahead… n. d.).
Railway automation is at an advanced stage since automatic train control and protection systems are used worldwide. High-speed trains can be considered almost autonomous, even if there is a driver onboard. H. Dong et al. (2010) describe in detail how the Chinese highspeed rail train control, protection and operation systems work. As China operates 80% of all high-speed trains in the world, they have not only advanced technologies but a very large amount of data. Even so, the main challenge they see is the better modelling of the tracks and the trains, how to create models that incorporate the growing number of variables and data, and how to deal with the ever-growing complexity of such models.
European railway systems are also close to automation: the European Rail Traffic Management System [ERMTS] was formed in 1998. It was necessary as there were more than 20 types of train protection and control systems in Europe. In the last two decades common technology could be rolled out in most of the EU member states, but we are still a long way away from an interoperable system. In 2014 the system was deployed on 5 800 km of European tracks and the goal is 56 000 km in 2030. That is only one quarter of the total 210 615 km of the total EU rail network, but it handles at least 80% of all the traffic. And this is only the infrastructure side, rail vehicles also have to be equipped with on-board units. That said, for the implementation of the ERMTS technology on the mostly heavily used European track the time frame will be more than 30 years. Within the ERMTS scheme there are different levels, and these figures also include the less developed level-1 too. The European Court of Auditors (2017) calculates that the implementation in Europe up to 2030 will cost at least 80 billion euros on the core network, and implementation on the comprehensive network up to 2050 a further 190 billion euros.
ERMTS is a comprehensive system to assist the train driver and train control, but is not intended for completely autonomous driving. This will though completely change the role of train drivers from an active position to a much more supervisory role. A. Naghiyev et al. (2017) analysed the different strategies that can be implemented for the role of train driver in the semi-autonomous ERMTS landscape. They showed that the design of future train driving systems is a complex process where human factors also need to be taken into consideration.
The average lifespan of railway locomotives and rolling stock is 30-50 years, of signalling and interlocking systems is 15–50 years, while structures such as platforms must be usable for up to 100 years (Tokody, Mezei & Schuster 2017). This means that changing technology in the rail sector is extremely slow.
In the case of AV in the rail industry, it can be assumed that it will only make sense to use full automation in the special circumstances of a new network being built. As of today this maybe would be the case in the US if it starts to create a high-speed rail network as passenger traffic there is very underdeveloped. In the case of freight transport automation will more likely be implemented in places where there is a lack of workforce and the environment is too harsh, and in the US there is also potential for automation as there the majority of tracks are freight-only and they are privately owned.
Full automation does not pose a similar social challenge in the case of railways as it does in the case of road transport. As it will be a very long process, train drivers will be around for some decades yet. Financially it is just a fraction of total costs to employ train drivers even if they only serve as supervisors in case of faults. Due to the high value of trains and the possible damage and the number of casualties a train could create in case of an accident there is far less encouragement for railway operators to fully eliminate human control.
Even if it did come to full automation, the number of engine drivers is much lower than truck drivers. According to UIC statistics in 2016 there were 80 297 locomotives and 75 813 railcars and multiple units (Railway Statistics – Synopsis), but there is no information on the number of engine drivers. For full utilisation 24/7 4-5 full time equivalent employee is needed, so the theoretical maximum of engine drivers worldwide can be estimated as between 625 and 780 thousand.
Automation has already had a much greater impact on railway than on road traffic. Its control has been centralised and there is no need for station staff anymore. In China 58.1% of the stations are computer controlled, in the EU that value is close to 100% whereas in India there is still a lot of manual labour. Table 3 shows the difference in employment compared to performance on a railway. The US rail sector, which mainly serves freight traffic, is the most efficient.
Employment and performance of the rail sector
Source: calculation based on UIC 2016 Synopsis and National Bureau of Statistics of China
|Country/region||Number of employees (thousand)||Train kilometres (millions)||Employment per train kilometres (thousand)|
The rail freight business has not really been renowned for innovation and a technology driven approach. This is rather a low-tech, low-profit industry which specialises in bulk products. There have been numerous attempts to position the industry for higher value-added logistics services but with less success.
For rail freight the autonomous shunting of wagons could be the first step in automation. Railway stations are closed territories, the movement of people can be controlled. In container terminals the shunting of railway wagons is already working in praxis (Hansen 2004).
Even if technologies are provided, railways are not focusing too much on automation. The reason for this can be understood if we look at the financial data for railways. DB Cargo from Germany is the biggest European rail freight corporation with an estimated market share of 25% (DB Cargo Company Presentation 2017). Based on financial data from 2016 we can calculate that engine driver labour costs make up around 6% of total costs. If it could be halved by AV technology – for instance with automatic locomotives used on major lines and in major shunting yards – with moderate costs of 30 000 euros per year per locomotive this could help the profit of the corporations, but only to a minor extent, as can be seen in Table 4 in the partial automation scenario.
DB Cargo AG Financial data as of 2016 [EUR]
Sources: calculated based on DB AG annual report 2017
|0: Current model||1: Partial automation||2: Full automation|
|Calculated depreciation of locomotives||394,380,000||394,380,000||394,380,000|
|Additional AV costs||0||84,510,000||211,275,000|
|Office and other||21,728||22,125||22,522|
|Trains / year||1,591,400||1,591,400||1,591,400|
|Number of locomotives||2,817||2,817||2,817|
|Engine driver’s salary - % of revenue||6.0%||3.0%||0.0%|
|Engine Driver’s salary / year / locomotive||96,462||48,231||0|
|AV costs / locomotive / year||0||30,000||75,000|
|AV costs / train / year||0||21,600||22,500|
But to be able to automate trains there is also an investment need on the infrastructure side which can be estimated in billions of euros as suggested by the European Court of Auditors (2017). In this model calculation only 30 000 euros per locomotive per year was calculated. If full automation could be achieved with costs of 75 000 euros per year per locomotive the engine drivers would not be needed, but more office workers would be required – it was assumed that the additional jobs would be 10% of the engine drivers dispensed with. This scenario does not generate a considerably higher EBIT (earnings before interest and taxes) for the companies than partial automation. Compared with the possible financial gains of road vehicle automation, AVs are much less lucrative in the case of railways.
Automation of urban logistics
According to the World Bank (Urban population (% of total)) today 54% of the world’s population is living in cities, and one in five people live in cities with more than 1 million inhabitants – and these figures are constantly rising. The UN is projecting that by 2030 the world will have 41 mega-cities with more than 10 million inhabitants. These highly populated urban areas are the centre of the global economy: a McKinsey report found that today 600 urban centres generate about 60% of global GDP (Dobbs et al. 2011). Although services in megacities are typically more important than industry, freight traffic is getting more and more challenging due to congestion and because urban mobility is creating air pollution. In the European Union urban mobility accounts for 40% of all CO2 emissions from road transport and up to 70% of other pollutants from transport (Directorate General for Mobility and Transport n. d.).
Therefore, the most interesting and promising field for AV is urban transport. Here the loads are smaller and distances shorter which makes this type of transport also the easiest to shift to electric mobility. The change in consumer behaviour is a great push towards faster and more efficient city logistics. The rise of e-commerce and the shift from bulk products to personalised goods is also a challenge for goods transport in cities and this is leading to the emergence of the importance of city logistics (Taniguchi 2014).
Numerous analyses have been prepared on what could be the effect of shared AV usage for cities. They all modelled individual transport and came to the conclusions that automated, usually electric, taxi-bot systems would be much more efficient than private car usage. K. Spieser et al. (2016) showed in a case study for Singapore that the same personal mobility needs as of today could be fulfilled with just one third of the vehicles with such a system. L. M. Martinez, H. A. C. Gonçalo & J. Viegas (2014) did a model for the city of Lisbon and even though it is considerably smaller with a population of 2.8 million in the metropolitan area, they were able to show that that the proposed shared-taxi system could significantly reduce fares and travel time for passengers. What has not been taken into consideration until now is the possibility of integrating passenger and freight traffic in cities (in the Singapore case study there was only one remark on city logistics). The integration of these services could lead to further cost cutting and further reduced usage of space. As shown in Figure 1, passenger mobility typically has peak hours in the mornings and late afternoons but vehicles could not be fully utilised during the day and especially at nights. Using AVs for goods delivery – like making taxi-bots able to haul trailers – could make the investment in the new technology more feasible.
The appearance of taxi-bots for transport will have further effects on city logistics. Most parking lots and garages will be superfluous and new usage models will be necessary for them. As the International Transport Forum (2015) observes, the freed-up space could be very useful for urban logistics purposes. This way smaller distribution centres could be established which would form a dense network. In a similar manner to the internet network this would mean a faster and much more independent network of goods delivery and could serve the growing need for faster delivery.
L. G. Clarembaux et al. (2016) presented a project in which they are planning to use autonomous electric vehicles in city logistics. Here they see very well the main challenge in the docking of the load and the vehicle. It is an important question to answer, what would be more efficient: using future urban trucks like today, stopping for loading and unloading, or would it be wiser to pack cargo in containers which could be attached and detached at the designated address so that the containers could be collected back again. This could lead to a similar efficiency gain in urban logistics as that provided by containerised traffic for the world economy. But it would require smaller containers in standard sizes. There was already an EU funded experiment with MicroCarriers in the city of Hanover and the results showed that such micro container carriers could be efficient in urban logistics (Brüning & Schönewolf 2011). The use of containers can be advantageous as in this way the utilisation of vehicles could be higher since there is no time lost in loading and unloading. However, this requires much more sophisticated planning of the trajectories of the vehicles.
Automated urban freight transport has a challenge that is less frequently discussed: autonomous box collection and delivery requires space. Even if technology is improving rapidly, autonomous vehicles can navigate less efficiently than human driven trucks. To make the collection smooth, special drop-off places should be created. In front of a supermarket which has a parking lot this does not seem a big challenge, or in the case of most US cities as they have been designed for motor traffic. In the case of European cities which have narrow streets, or in very densely built up Asian cities, autonomous vehicles can have more disadvantages than advantages. Parking for a longer period, and lack of human communication with other drivers can lead to traffic jams. And current autonomous driving technologies are not yet up to standard in coping with such a complex environment as an alley, where road widths change constantly, children and pets are running through, and loading and unloading sometimes cannot even be organised according to traffic rules and involves improvisation by drivers.
One of the anticipated positive effects of AVs is that on the environment: their fuel efficiency is thought to be higher than human driven vehicles. In the case of light-duty transportation modes there is already evidence. Without considering efficiency as a target the fuel efficiency of AVs is 3% better than in the case of human driven vehicles and if the software is focused on fuel consumption the gain can be up to 10% (Mersky & Samaras 2016). These results could suggest that AV in goods transport leads to lower emission rates and higher efficiency.
But growing efficiency in road transport due to AVs has also meant that road transport can gain a yet higher market share in the transport business. Road transport is already the leading transport mode globally (Fig. 2). As road transport has gained an ever-higher share of total freight transport in recent decades, this trend can be strengthened with the help of AVs. Electric trucks – especially for longer distances – are seen as inefficient in the decades to come, therefore diesel trucks will be used in the near future. AVs pose a high risk of CO2 emission growth if automation of trucking improves at a faster pace than other transport modes. It is a very different threat in the major economic zones.
The CO2 emissions of the major economic centres of the world can be calculated if we take per ton-km emission values from The European Chemical Industry Council [CEFIC] (2011). Models can be made of the kind of impact the rising share of road transport would have on the modal split for emission values (Tab. 5). In Scenario 1 it was assumed that more efficient road traffic attracts 5% of rail traffic and in Scenario 2 10% and in Scenario 3 25% of the related CO2 emissions for the same total traffic volumes. It can be seen that even with the growing efficiency of road transport, global emissions can grow significantly, between 1 and 9%.
Calculated CO2 emissions of freight transport [million tons]
Source: calculated based on EU Transport in figures – Statistical pocketbook 2016 and CEFIC 2011
|Scenario 1 change||1%||2%||1%||6%||0%||1%|
|Scenario 2 change||2%||3%||1%||11%||1%||3%|
|Scenario 3 change||5%||11%||4%||41%||2%||9%|
Autonomous vehicles will have a remarkable impact in moulding the future of the transport industry in the coming decades. The individual traffic modes will be affected very differently. The road vehicles are mostly likely to achieve full automation but before that can happen more and more adaptive digital technologies will be used which will serve not only to produce greater safety and efficiency, but also will provide a means to collect data for further developments. The trucking industry has substantial efficiency growth potential, but this will have a considerable social cost as millions of truck drivers can lose their jobs, which can affect 0.4 percent of all employees world-wide.
The other field for innovative autonomous solutions can be urban logistics where the connection between the transport of individuals and freight transport can also offer efficiency gains. Regulation and safety in this type of development can be a bigger challenge, but the market push effects of the rise of e-commerce will help developments in this field. AVs are being developed for trucking by incumbent corporations and lots of the innovations can be seen on the roads by the middle of the 2020s. In urban logistics new companies and business models can emerge as the current operators are less aware of the challenges of automation.
One other means of transport, especially rail, will also see some degree of development in the field of automation but the introduction of either of these new technologies would involve very great investment and/or a very low return on investment. Therefore, it is not expected that these traffic modes will face significant effects from automation.
Autonomous driving will most probably make road transport even more competitive in comparison with other modes of transport. This will have considerable environmental impacts, as road transport had proportionately the highest rates of CO2 emissions. Every 5 percent modal shift gain of road transport will add an extra 1 percent to global CO2 emissions in the transport sector. The potential effects on modal split from AVs have to be taken into account in future transport and environmental policies and new regulations have to be implemented to be able to reduce CO2 emissions.
Baier S. L. & Bergstrand J. H. (2001) The growth of world trade: tariffstransport costsand income similarity Journal of International Economics 53 (1) 1–27.
Bezwada N. (2010) Characteristics and Contributory causes Associated with Fatal Large Truck Crashes Department of Civil Engineering. Available from: http://krex.k-state.edu/dspace/bitstream/handle/2097/6820/NishithaBezwada2010.pdf?sequence=1&isAllowed=y [accessed: 30.10.2017].
Brüning M. & Schönewolf W. (2011) Freight Transport System for Urban Shipment and Delivery IEEE Forum on Integrated and Sustainable Transportation Systems Vienna 136–140.
Buning M. C. & de Bruin R. (2017) Autonomous intelligent cars: proof that the EPSRC Principles are future-proof Connection Science 29 (3) 189–199.
Clarembaux L. G. Pérez J. Gonzalez D. & Nashashibi F. (2016) Perception and control strategies for autonomous docking for electric freight vehicle Transportation Research Procedia 14 1516–1522.
DB presses ahead with train automation n. d. Available from: https://www.deutschebahn.com/en/Digitalization/automation.html [accessed: 07.09.2017].
Directorate General for Mobility and Transport (n. d.) Clean transportUrban transport European Commission. Available from: https://ec.europa.eu/transport/themes/urban/urban_mobility_en [accessed: 02.12.2017].
Dobbs R. Smit S. Remes J. Manyika J. Roxburgh Ch. & Restrepo A. (2011) Urban world: Mapping the economic power of cities Report of the McKinsey Global Institute. Available from: https://www.mckinsey.com/global-themes/urbanization/urban-world-mapping-the-economic-power-of-cities [accessed: 02.12.2017].
Dong H. Ning B. Cai B. & Hou Z. (2010) Automatic Train Control System Development and Simulation for High-Speed Railways IEEE Circuits and Systems Magazine 10 (2) 6–18.
European Automobile Manufacturers Association 9.05.2017 Truck industry gears up for wide-spread introduction of semi-automated convoys by 2023. Available from: http://www.acea.be/press-releases/article/truck-industry-gears-up-for-wide-spread-introduction-of-semi-automated-conv [accessed: 28.10.2017].
European Commission (2016) Annual Accident Report European Commission Directorate General for Transport. Available from: https://ec.europa.eu/transport/road_safety/sites/roadsafety/files/pdf/statistics/dacota/asr2016.pdf [accessed: 23.09.2017].
European Court of Auditors (2017) A single European rail traffic management system: will the political choice ever become reality? Special Report no. 13. Available from: http://www.eca.europa.eu/Lists/ECADocuments/SR17_13/SR_ERTMS_RAIL_EN.pdf [accessed: 22.11.2017].
European Truck Accident Causation (n. d.) European Truck Accident Causation ETAC Scientific Study European Commission Directorate General For Energy and Transport volume 1 – final report. Available from: https://ec.europa.eu/transport/road_safety/sites/roadsafety/files/pdf/projects_sources/etac_final_report.pdf [accessed: 23.09.2017].
Fagnant D. J. & Kockelman K. (2015) Preparing a nation for autonomous vehicles: opportunitiesbarriers and policy recommendations Transportation Research Part A: Policy and Practice 77 (6) 167–181.
Gordon T. J. & Lidberg M. (2015) Automated driving and autonomous functions on road vehicles Vehicle System Dynamics 53 (7) 958–994.
Hansen I. (2004) Automated shunting of rail container wagons in ports and terminal areas Transportation Planning and Technology 27 (5) 385–401.
International Transport Forum (2015) Urban Mobility System Upgrade How shared self-driving cars could change city traffic – Corporate Partnership Board Report. Available from: https://www.itf-oecd.org/sites/default/files/docs/15cpb_self-drivingcars.pdf [accessed: 05.11.2017].
Lammert M. Duran A. Diez J. Burton K. & Nicholson K. (2014) Effect of Platooning on Fuel Consumption of Class 8 Vehicles Over a Range of SpeedsFollowing Distancesand Mass SAE International Journal of Commercial Vehicles 7 (2) 626–639.
- Export Citation
Lammert, M., Duran, A., Diez, J., Burton, K. & Nicholson, K. (2014))| false Effect of Platooning on Fuel Consumption of Class 8 Vehicles Over a Range of Speeds, Following Distances, and Mass, SAE International Journal of Commercial Vehicles, 7 (2), 626–639. 10.4271/2014-01-2438
Martinez L. M. Gonçalo H. A. C. & Viegas J. (2014) An agent-based simulation model to assess the impacts of introducing a shared-taxi system: An application to Lisbon (Portugal) Journal of Advanced Transportation 49 (3) 475–495.
Masoud N. & Jayakrishnan R. (2015) Formulations for Optimal Shared Ownership and Use of Autonomous or Driverless Vehicles Proceedings of the 94th Annual Meeting of the 20 Transportation Research Board in Washington DC.
Mersky A. C. & Samaras C. (2016) Fuel economy testing of autonomous vehicles Transportation Research Part C: Emerging Technologies 65 (4) 31–48.
Milakis D. van Arem B. & van Wee B. (2017) Policy and society related implications of automated driving: A review of literature and directions for future research Journal of Intelligent Transportation Systems 21 (4) 324–348.
Mooren L. Grzebieta R. Williamson A. Olivier J. & Friswella R. (2014) Safety management for heavy vehicle transport: A review of the literature Safety Science 62 (2) 79–89.
Naghiyev A. Sharples S. Ryan B. Coplestone A. & Carey M. (2017) Expert knowledge elicitation to generate human factors guidance for future European Rail Traffic Management System (ERTMS) train driving models Proceedings of the Institution of Mechanical Engineers Part F: Journal of Rail and Rapid Transit 231 (10) 1141–1149.
- Export Citation
Naghiyev, A., Sharples, S. Ryan, B., Coplestone, A., & Carey M. (2017))| false Expert knowledge elicitation to generate human factors guidance for future European Rail Traffic Management System (ERTMS) train driving models, Proceedings of the Institution of Mechanical Engineers, Part F: Journal of Rail and Rapid Transit, 231 (10), 1141–1149. 10.1177/0954409717695902
Newnam S. Goode N. (2015) Do not blame the driver: A systems analysis of the causes of road freight crashes Accident Analysis & Prevention 76 (C) 141–151.
Rio Tinto completes first fully autonomous rail journey in Western Australia 2.10.2017 Rio Tinto Media Release. Available from: http://www.riotinto.com/media/media-releases-237_23264.aspx [accessed: 01.12.2017].
Rost P. Banchs A. Berberana I. Breitbach M. Doll M. Droste H. Mannweiler C. Puente M. A. Samdanis K. & Sayadi B. (2016) Mobile Network Architecture Evolution toward 5G IEEE Communications Magazine 54 (5) 84–91.
Safe Road Trains for the Environment; developing strategies and technologies to allow vehicle platoons to operate on normal public highways with significant environmentalsafety and comfort benefits (n.d.) Cordis. Available from: http://cordis.europa.eu/project/rcn/92577_en.html [accessed: 18.11.2017].
Spieser K. Treleaven K. B. Zhang R. Frazzoli E. Morton D. & Pavone M. (2014) Toward a Systematic Approach to the Design and Evaluation of Automated Mobility-on-Demand Systems A Case Study in Singapore [in:] G. Meyer & S. Beiker eds. Road Vehicle Automation (Lecture Notes in Mobility) Springer Chem Switzerland.
Taniguchi E. (2014) Concepts of city logistics for sustainable and liveable cities Procedia-Social Behavioral Sciences 151 310–317.
Test track for autonomous trains in Austria 27.09.2017 ReilFreight. Available from: https://www.railfreight.com/business/2017/09/27/test-track-for-autonomous-trains-in-austria/ [accessed: 07.10.2017].
The European Chemical Industry Council [CEFIC] (2011) Guidelines for Measuring and Managing CO2 Emissions from Freight Transport Operations. Available from: http://www.cefic.org/Industry-support/Responsible-Care-tools-SMEs/5-Environment/Guidelines-for-managing-CO2-emissions-from-transport-operations/ [accessed: 01.04.2018].
Tokody D. Mezei I. J. & Schuster G. (2017) An Overview of Autonomous Intelligent Vehicle Systems [in:] K. Jármai & B. Bolló eds. Vehicle and Automotive Engineering. Lecture Notes in Mechanical Engineering Springer Cham (Switzerland) 287–307.
Turri V. Besselink B. & Johansson K. H. (2016) Cooperative Look-Ahead Control for Fuel-Efficient and Safe Heavy-Duty Vehicle Platooning IEEE Transactions on Control Systems Technology 25 (1) 12–28.
Zolock J. Senatore C. Yee R. & Larson R. (2016) The Use of Stationary Object Radar Sensor Data from Advanced Driver Assistance Systems (ADAS) in Accident Reconstruction SAE Technical Paper 2016-01-1465. https://doi.org/10.4271/2016-01-1465.
Budapest metro timetable for passenger trains: https://bkk.hu/en/timetables/ [accessed: 01.04.2018].
Company register of Hungary: http://e-beszamolo.im.gov.hu/oldal/kezdolap [accessed: 14.10.2017] [in Hungarian].
DB AG Annual Report 2017: https://www.db.com/ir/en/download/DB_Annual_Report_2017.pdf [accessed: 15.10.2017].
DB Cargo Company Presentation 2017: http://www.xrail.eu/documents/13422/0/170330_DB+Cargo+company+presentation_english.pdf/15893927-bcc4-40ee-b7cc-52c854dfc970 [accessed: 14.06.2018].
EU Transport in Figures – Statisticalpocketbook 2016: https://ec.europa.eu/transport/facts-fundings/statistics/pocketbook-2016_en [accessed: 12.10.2017].
National Bureau of Statistics of China: National Bureau of Statistics of China [accessed: 12.12.2017].
Railway Statistics – Synopsis: https://uic.org/IMG/pdf/synopsis_2016.pdf [accessed: 01.12.2017].
The Waberer’s Group website: http://www.waberers.com/en/investors [accessed: 14.10.2017].
UIC 2016 Synopsis: https://uic.org/IMG/pdf/synopsis_2016.pdf [accessed: 15.11.2017].
Urban population (% of total): https://data.worldbank.org/indicator/SP.URB.TOTL.IN.ZS [accessed: 15.10.2017].