Excerpt
Table of Contents
Abstract
List of Abbreviations
List of Figures
List of Tables
1 Introduction
2 Literature Review
2.1 The History and Development of the German Automobile Industry
2.2 Autonomous Driving
3 Research Aims and Contribution
3.1 Research Aim
3.2 Significance of the Study
3.3 Research Question and Objectives
4 Methodology
4.1 Research Design
4.2 Data Collection
4.3 Sampling Strategy
4.4 Data Analysis
5 Primary Research Analysis
5.1 Qualitative Analysis
5.2 Quantitative Analysis
6 Discussion
6.1 Technology
6.2 Legal and Regulatory Environment
6.3 Customer Readiness
6.4 Key Recommendation for German OEMs
7 Guidance for Future Research
7.1 Theoretical and Methodological Limitations
7.2 Suggestions for Future Research
7.3 Personal Reflection of the Paper
References
Appendices
Appendix A: VW vs. main Competitors: Units Sold in 2004 and 2014
Appendix B: SAE International Levels of Autonomous Driving
Appendix C: GM Leads, Tesla & Apple Trail Deeply In Navigant Research Self-Driving Report
Appendix D: US States with Enacted Autonomous Vehicle Legislation
Appendix E: Rogers Product Adoption Curve
Appendix F: Survey on Customer Readiness
Appendix G: Age Distribution of Survey Respondents
Appendix H: T-test on Willingness to use AVs by Gender
Appendix I: ANOVA on Willingness to give up One's own Car and Importance of Driving Experience for Female Participants.
Appendix J: T-test on Influence of Positive and Negative News of AVs on males
Bibliografische Information der Deutschen Nationalbibliothek:
Die Deutsche Nationalbibliothek verzeichnet diese Publikation in der Deutschen Nationalbibliografie; detaillierte bibliografische Daten sind im Internet über http://dnb.d-nb.de abrufbar.
Impressum:
Copyright © ScienceFactory 2018
Ein Imprint der Open Publishing GmbH, München
Druck und Bindung: Books on Demand GmbH, Norderstedt, Germany
Covergestaltung: Open Publishing GmbH
Abstract
The automobile industry is one of Germany’s strongest and most important industries. However, the diesel emission scandal is a crack in the perceived quality, reliability and the overall premium image of Germany. The automotive industry is facing its greatest and most critical change since the invention of the automobile. Four megatrends (connected, autonomous, shared and electric) are dominating the 21st Century and the entire automotive industry around the world. Various companies are working intensively on the development of the next major milestone in human history that incorporate all megatrends, autonomous vehicles. The technology is perceived to be disruptive and thus many challenges and obstacles remain before the new technology becomes superior to human drivers.
This research paper aims to explore and investigate the status quo of the development of autonomous vehicles at German OEMs. It further aims to identify future impediments until market entrance and recommend managerial actions. The paper focuses on three key pillars: Technology, legal and regulatory environment, and customer readiness which are indispensable for a successful implementation of autonomous vehicles. Various interviews are conducted with experts from OEMs and automobile associations and potential German end-users are surveyed. The analysis revealed various insights which are discussed and recommendations for German OEMs are given.
List of Abbreviations
illustration not visible in this excerpt
List of Figures
Figure 1. Production of Passenger Vehicles by Country from 1961-2015
Figure 2. Car Brand Map based on Cost and Product Excellence
Figure 3. Levels of Autonomous Driving. Adapted from SAE J3016
Figure 4. Types of Regulation for Automated Driving
Figure 5. Research Onion
Figure 6. Willingness to use AVs.
Figure 7. Willingness to give up one's own Car.
Figure 8. Willingness to give up one's own Car by Gender.
Figure 9. Evaluation of Daily Mobility Factors.
Figure 10. Preferred Shape of AV and Concerns of Riding in an AV without Self-Steering Possibilities.
Figure 11. Safety Considerations of AVs.
Figure 12. Trustworthiness of AV Providers.
Figure 13. Fear of Cyberattacks on AVs.
Figure 14. Influence of Positive and Negative News on Males.
Figure 15. Influence of Positive and Negative News on Females.
Figure 16. Knowledge of AVs by Gender.
Figure 17. Key Recommendation for German OEMs.
List of Tables
Table 1: TCA Steps.
Table 2: TCA Identified Themes - Technology.
Table 3: TCA Identified Themes - Legal and Regulatory Environment.
Table 4: Chi-Square-Test of Gender and Willingness to give up one's own Car.
Table 5: Anova on Willingness to give up one's own Car and Importance of Driving Experience for Male Participants.
Table 6: T-test on Influence of Positive and Negative News of AVs on Gender.
Table 7: T-test on the Influence of Positive and Negative News of AVs on Females.
Table 8: Difference in Knowledge of AVs by Gender.
1 Introduction
With a turnover of around 404 billion Euro which account for approximately 20% of the total German industry revenue, the automobile industry is one of Germany’s greatest and most important industries (GTAI, 2017). The great success of the German automobile industry evolved over decades and ensured Germany a place at the top of the world’s carmakers. However, with the diesel emission scandal, the industry and OEMs took loses, especially VW (Volkswagen). The VW’s stock price plunged by more than 50% within a few months (La Monica, 2015). The scandal is a crack in the perceived quality, reliability and the overall premium image of cars made in Germany (McGuinness, 2015). The timing of the scandal is rather unfortunate since the automotive industry is facing the greatest change of all time. Four megatrends are dominating the 21st Century and the entire automotive industry around the world. The tremendous and disruptive[1] trends and its emerging technologies[2] are affecting and changing the industry completely (Heineke, Möller, Padhi, & Tschiesner, 2017; Morgan Standley, 2016). The first megatrend is shared. Similar to other industries, the sharing economy influences the automotive industry as car and ride sharing players entered the market (Grosse-Ophoff, Hausler, Heineke, & Möller, 2017). The second megatrend is electric. Global warming, the tightened Co2 reduction targets and the pioneering of Tesla lead to a drastic increase in electric vehicles. The adoption of electric vehicles is ongoing as carmakers around the world are electrifying their vehicles (Times, 2017). Connected is considered to be the third megatrend of the automotive industry. As the digitalization process of the society proceeds, it vastly affects automobiles with respect to internal and external connectivity for optimizing vehicle operation and maintenance as well as for passenger’s utilization of travelling time (McKinsey, 2014). Though autonomous is considered to be the fourth megatrends, it is also perceived as the ultimate combination of all megatrends. As a matter of fact, the emerging technology is seen by Germany’s transport minister Alexander Dobrindt as the ‘greatest mobility revolution since the invention of the car’ (BMVI, 2016). The reason is that AVs (autonomous vehicles) are utilized in shared manners, are electrified and connected to the environment. The technology is perceived as the solution to the market movements and answer to the megatrends in the automotive industry. The reason is that AVs have the potential to reduce the Co2 emission and eliminate a significant portion of traffic jams while increasing mobility and space utilization in urban areas (Kuper, 2016).
While these megatrends appose threats to some, Dr. Dieter Zetsche (CEO of Daimler AG) conceives the transformation differently as he believes that ‘changes are not a threat, but an opportunity’ (VDA, 2017). Nevertheless, the changing environment levels the playing field to some extent which increases the number of competitors as newcomers[3] are entering the automotive industry. This impacts the global competition and increases the pressure on German OEMs (original equipment manufacturers) to innovate. But not only vehicles are affected, rather the entire ecosystem of mobility has to be adopted and change in order to enable a successful implementation of AVs (Wladawsky-Berger, 2016). Consequently, the infrastructure must be adjusted simultaneously to the development of AVs. This requires a co-working and cooperation between political and economic forces. Moreover, cooperating is vital in order to stem the significant burden of the transformation as well as to ensure and maintain the top level position of the German automobile industry on a macroeconomic perspective. Otherwise, competitors from the USA and China might take over the German market and harm one of Germany’s most critical industries (McKinsey, 2016).
This research paper is guided by the following structure: The paper starts with a literature review including the three research pillars of AD (autonomous driving): Technology, legal and regulatory environment, and customer readiness. The chapter is followed by a description of the research aim and contribution which contains the defined research question and objectives. The methodological research design, data collection, and data analysis approach is elaborated within the methodology section. Afterward, the conducted primary research is analysed based on the characteristics of the collected data (qualitative and quantitative). In the following chapter, the key summarized findings are discussed regarding their theoretical as well as managerial implications and aggregated recommendations for German OEMs are given. The paper ends with an elaboration of the theoretical and methodological limitations of the research project and gives an aggregated prospect for future research.
2 Literature Review
2.1 The History and Development of the German Automobile Industry
The history of the automobile industry can be divided into the following five stages or phases: Firstly, the initial automobile development, secondly the development of the modern automobile, thirdly the pre-World War II-phase, fourthly the post-World War II-phase and fifthly the phase of the 21st century.
The early roots of the automobile can be found in the very beginning of the history, around 100 C.E when the Greek philosopher Heron of Alexander used steam to resolve a ball on an axle (Collier, 2006). Various engineers from different countries developed interim stages of a fully functioning and usable automobile (Weeks, 2011). The ultimate breakthrough in the development was made in Germany by Carl Benz as well as Gottlieb Daimler in 1885 (Weeks, 2010). The two engineers developed the first modern automobiles which used an Otto cycle engine (Sinclair, 2004). Carl Benz’s engine was attached to a three-wheeled vehicle while Gottlieb Daimler used a two-wheeled bicycle (McNeese, 2000). During the 1890s many other engineers started building modern automobiles such as the German OEM Opel in 1899 (Pohl & Rudolph, 1990). The industry started to grow very slowly as only a few duplicates were made (Flink, 1990). At that time, French car manufacturers led the way in terms of production (McNeese, 2000). Due to the costly production process and the materials, the price of the products was high and industry growth was limited (Mullin, 2010). The next major capstone in the history of the automobile was achieved by the American Henry Ford who founded the Ford Motor Company in 1903 and introduced the assembly line to the automobile production in 1913 (Banham, 2002).
During the beginning of the 20th century, the early growth phase of the automobile industry, many new and today’s leading car manufacturers entered the industry. The American OEM GM (General Motors) was founded in 1908 (Fourie, 2016) which acquired Opel during the end of the 1920s (Kudo, Kipping, & Schröter, 2004). Shortly after, the German OEMs Audi (1909) and BMW (Bayerische Motoren Werke) followed in 1916 (Joseph, 2013). In 1931 Porsche was founded (Joseph, 2013) and in 1937 VW (Parment, 2014) as well as Toyota entered the industry (Fujimoto, 1999). Before World War II, engineers made significant improvements with respect to the technical abilities and applications of the vehicles. Shortly before and during the beginning of World War II, the number of vehicles produced increased, mainly due to military reasons (Lepage, 2007). The real production boom occurred after World War II during the major reconstruction phase in Germany between 1954 and 1965. During this period, local production and export increased significantly to 52% in 1965, while imports were relatively low at about 10-12% (IFO Institute for Economic Research & Shuka Institute of Research, 1997). The production in Germany further increased in the following years, though with a weaker slope, while the Japanese kept rocketing to the top, as illustrated in Figure 1 (U.S. Department of Transportation, n.d.).
illustration not visible in this excerpt
Figure 1. Production of Passenger Vehicles by Country from 1961-2015
In accordance with Table 1-23: World Motor Vehicle Production, Selected Countries (Thousands of vehicles), by U.S. Department of Transportation, n.d., Retrieved July 16, 2017, from Bureau of Transportation Statistics: https://www.rita.dot.gov/bts/sites/rita.dot.gov.bts/files/ publications/national_trans.
While the US car industry started struggling, German OEMs benefited from an unexpected growth due to the opening up of East Germany and the Eastern European market (Preissl & Solimene, 2003). Besides, German OEMs kept widening their portfolio through acquisitions as well as the development of new models (Parissien, 2013). From early on and until the end of the 20th century, the ‘Made in Germany’ stamp and the internationally associated quality with the mark, lead to a position at the top of the premium passenger vehicle market (Joseph, 2013). At the beginning of the 21st century, all German brands were positioned at the top of each segment as illustrated by Figure 2 (Hirsh, Hedlund, & Schweizer, 2003). While the brand VW operated at the top of the mass-market segment and Audi positioned itself between the premium and luxury segment, BMW, and Mercedes as well as Porsche were already perceived as the top-notch of the premium or luxury segment.
illustration not visible in this excerpt
Figure 2. Car Brand Map based on Cost and Product Excellence
Adopted from Reality Is Perception: The Truth about Car Brands, by E. Hirsh, S. Hedlung & M. Schweizer, 2003, Retrieved July 16, 2017, from Strategy + Business: https://www.strategy-business.com/article/03302?gko=fbb50.
Nevertheless, German OEMs were hit hard by the financial crisis in 2008 (Boston, 2008), but the industry was able to recover faster than expected (Kollewe, 2012). By the end of 2014, VW was at the top of the global automotive industry (see Appendix A). By then the company had recovered from the financial crisis and build a significant portfolio of different automobile brands. Besides already holding several commercial vehicle brands and sub-companies, in 1988 and 1999 VW acquired three top passenger luxury brands: Bugatti, Lamborghini, and Bentley (Roberts, 2005). The OEM SEAT was already part of the portfolio since 1986 and in 1991 Skoda joined the group (English, 2011). A few years later in 2012, Porsche was also acquired (Topham, 2012). Mercedes and BMW widened their portfolio as well with respect to passenger and commercial vehicles (Freyssenet, 2009). By 2015, two third of the total production from German OEMs was made outside of Germany, indicating the continuation of outsourcing low-value assembly processes (VDA, 2016).
The diesel scandal in 2015 (Taylor & Potter, 2017) as well as the increasing performance of Tesla and the technological trends changed the environment of the German automobile industry significantly (Heneric, Licht, & Sofka, 2005). Today, the industry is subject to the dynamic megatrends of connected, autonomous, shared and electric (Daimler AG, n.d). As a result of the dynamic trends, VW and Mercedes changed their corporate strategy, moving from an automobile manufacturer towards a mobility provider (McKinsey, 2016). While electric vehicles, connectivity features and shared services are already available or entering the market within the next couple of years, AD is considered to be the next major milestone in the automotive industry that will influence everyone and change the world (Alexander, 2013).
2.2 Autonomous Driving
2.2.1 The history and development of autonomous driving
The first significant step in the history of AD is dated back to the 1930s and 1940s when the idea of an automated highway system was proposed (Meyer & Beiker, 2014). During the second half of the 20th century, more American researchers and developers became active and concentrated on the field of an automated highway driving. In Germany, it was Mercedes Benz which reached a milestone during the 1980s as the first automobile manufacturer to test AD on streets in Bavaria, though without any traffic at that time (Siciliano & Khatib, 2016). At the beginning of the 21st century, the research filed was expanded and AVs started to be the main focus of manufacturers and researchers (Meyer & Beiker, 2014).
Though AD is considered to be a disruptive technology, the development and implementation process are rather continuous. While the NHTSA (2013) defined only five phases or four AD-levels, SAE International (2014) extended the definition. Figure 3 illustrates the six phases or five AD-levels ranging from no automation to full automation. The phases or AD-levels can be divided into two main parts (SAE International, 2014): In the first part, the human driver monitors the driving environment (up to AD-level 2) and in the second part, the automated driving system monitors the driving environment (from AD-level 2 until AD-level 5).
illustration not visible in this excerpt
Figure 3. Levels of Autonomous Driving. Adapted from SAE J3016
Taxonomy and Definitions for Terms Related to Driving Automation Systems for On-Road Motor Vehicles by SAE Internal, 2014: SAE International.
The first two AD-levels are already existing and AD-level 3 will be reached between 2018 and 2020 (Roland Berger, 2014). The first production of AD-level 3 vehicles has already started as Audi revealed its new A8 which demonstrates its intelligence and self-driving applications with Audi’s new traffic jam pilot (Taylor, 2017). The first AVs with AD-level 4 are estimated to be leaving the factories between 2020 and 2025 (Roland Berger, 2014). Whereas, AD-level 5 is seen as a future prospect as it requires AVs to be not limited to any driving mode that a human drier can manage which includes aspects such as no limitation by speed or driving environment (see Appendix B)[4].
Various different OEMs as well as newcomers are developing the system behind AVs. Though Hanley (2018) considers certain companies to be leading in developing AVs (see Appendix C), the overall competition is still ongoing. While Tesla, VW, Apple, and Toyota are developing the technology on their own, others formed strategic alliances and cooperations such as GM and Lyft, Volvo and Autoliv, Daimler and Bosch, BMW and Intel (including Mobileye) and Waymo with FCA (Fiat Chrysler Automobiles) (CB Insights, 2017). The partnership of Waymo and FCA is limited to a certain number of vehicles. Therefore, Waymo aimed but failed so far to recruit Ford as a vehicle supplier (Carty, 2017). Ford on the other hand is now using the resources of Lyft (Isaac, 2017).
The technology of AVs, including specially designed AVs and conventional vehicles with AD-ability[5], offers many benefits with respect to the commercial as well as private transportation of passengers and goods (Ozimek, 2014; Omohundro, 2014). Firstly, the number of accidents could be reduced significantly, as about 90% of all car accidents are caused by a human error. Secondly, time and efficiency could be increased drastically, as traffic jams could be reduced since fewer vehicles are needed due to attractive ride sharing and the increased utility of each shared AV. Furthermore, congestion could be diminished because AVs are able to drive in packs and do not take wrong turns. As a result of the increasing efficiency, consumption as well as pollution and carbon emissions are reduced, especially because AVs are powered by electricity. As a consequence of the gained efficiency, commuting and travelling time in general as well as the associated cost could decrease (Bunghez, 2015). Therefore, the consumers in the taxi and logistic industry could benefit from faster and cheaper deliveries (Mark, 2017; DHL, 2014). Disabled and handicapped people could gain mobility because driving a car would not be necessary for high mobility standards anymore. Thirdly, the technology could have positive implications on the infrastructure. The parking situation could be de-escaladed as no permanent parking spot inside urban areas would be needed. The size of car lanes could be reduced since AVs make no steering mistakes and thus less room for steering adjustments is required. Fourthly, the tedium as well as the stress of commutes could diminish due to the fact that passengers are able to utilize their time differently (Kuper, 2016).
2.2.2 Status quo of autonomous driving in the German automobile industry
The following three paragraphs display a secondary literature review of the three research pillars: Firstly, the technology, secondly the legal and regulatory environment and thirdly the customer readiness.
Technology. The emerging technology offers various promising advantages and benefits, though several R&D (research and development) obstacle remain before implementing the technology becomes possible. The technological development challenges are similar for all OEMs around the world as most components are developed nonexclusively by top-tier suppliers. One exception is the ‘brain’ of the vehicle, the AI (artificial intelligence) which interprets the driving environment and steers the vehicle. The key development areas can be divided into the following three categories: Sensing and interpreting the driving environment, communication and cybersecurity (Anderson, et al., 2016).
Sensing and interpreting the driving environment. Anderson, et al. (2016) define a three-phase sense-plan-act design that AVs underline. Firstly, the driving environment is sensed. Secondly, the vehicle’s action is planned and eventually carried out through an actionable command in the third step. These loops occur all the time and in parallel. Cameras and various sensors are responsible for scanning the environment and gathering input information. The difficulty for the sensors is to accurately perceive the driving environment as it is a very dynamic and complex situation. Especially sensing the variety of different elements is a challenge such as various road obstacles including pedestrians, wildlife, debris, cyclists or traffic events such as roadwork, congestion or crashes as well infrastructure conditions including various and rough surfaces, poorly marked roads or lanes, detours and defect traffic lights. The lidar (light detection and ranging) system is able to determine obstacles by using laser ranger finders. However, it functions reliably only over shorter ranges and on certain well reflecting materials. Whereas the Velodyne system works to a range of up to 120 meters. But, on low reflecting materials such as asphalt, the system only works up to 50 meters. On small distances, infrared systems are capable of detecting lane markings without lighting. Another sensor technology used in AVs is radar (radio detection and ranging) which works well on metallic objects such as vehicles but poorly on non-metallic obstacles such as pedestrians. Therefore, pedestrians and bicycles can be spotted by infrared sensors. Ultrasonic sensors are accurate on short ranges of 1 to 10 meters and thus are very useful for parking assistance and backup warning. In general, each sensor provides a different kind of data and therefore has other benefits and limitations. Due to that reason, AVs usually use a combination of sensor systems in order to offset the limitations. For example, sensor suits are used to perfectly localize a vehicle. It is a combination of GPS (global positioning system) and INS (inertial navigation system). Even under ideal conditions, the GPS system is often inaccurate by several meters. This error increases drastically when obstacles or terrain occlude the sky such as in urban areas. Therefore, the system is coupled with the INS. The greatest challenge is the drift that occurs when the system relies only on the INS system as GPS is unavailable. As a result, many AVs draw on prebuild maps, but the difficulty with prebuild maps is the construction and maintenance of an accurate map. Since AVs underlie tight financial constraints, integrating adjustable, multifunctioning sensors is uneconomically. OEMs have to choose between different sensors and consequently between various capabilities and limitations at different price points in order to fulfil their customer’s needs. Another key technical challenge is the accurate functionality under extreme environmental circumstances. The accuracy of sensors declines dramatically during serve precipitation, heavy snow, extreme temperatures or dense fog. Different terrains pose challenges as well as sensors are rather made for one terrain or another. For example, sensors might have issues either on a flat environment or steep hills. The different materials of the road such as asphalt or dirt require different abilities of the sensors due to varying reflection of materials (Attias, 2017). Apart from environmental challenges, sensor failure is another key difficulty since safety is of utmost importance. Sensors may fail due to physical damage, age or electrical reasons (Ploeg, 2017). Therefore, an internal sensing system and algorithm that can detect components that are not performing adequately is required. More precisely, an ultra-reliable simple low-level system with basic sensor functions is needed that takes over in the event of degradation or failure. The system has to detect and override control rapidly in order to be able to steer the vehicle away from dangerous areas such as high traffic roads or blind curves (Anderson, et al., 2016).
Interpreting the driving environment is the second step. After the sensors delivered the input information and data, the ‘brain’ of the vehicle analyses the information and plans the action as well as eventually carries it out. With perfect perception, computers are extremely accurate and reliable. However, they lag humans especially when it comes to interpreting the driving environment. The interpretation is done by the ‘brain’ or AI of the vehicle. AI comprises of machine learning and deep learning which are two different approaches of analysing and processing input data (Pathak, 2017). Deep learning is applied by many top-notch high tech companies such as Drive.ai, AImotive, and FiveAI. While it works exceptionally well in most situations the major drawback is the impossibility of understanding a decision of the AI. The decision process is a ‘black box’ embodied by neural networks consisting of millions of nodes that analyse raw data and make decisions. Due to the limitation of understanding the decision process, some developers such as Ford rather focus on machine learning which is grounded on algorithms and logic (Toews, 2017). The underlying algorithms in this approach are not as sophisticated as human brains or neural networks yet. Thus, they are more vulnerable to difficult environmental conditions which impose great interpretation challenges. The algorithms allow to comprehend and re-enact the machine’s decision and thus problems can be explicitly addressed. Another key challenge of AVs is the decision in an ethical dilemma situation such as the trolley problem in which the AV has to choose between the life of humans (Overly, 2017). Daimler takes a leading position in this dilemma, as the OEM announced to priorities its passengers and not pedestrians based on the reason that AVs should primarily save those that the AV has direct control over (Morris, 2016).
Secondly, vehicle communication. This includes HMI (Human Machine Interface), the communication of the vehicle with humans, as well as communication of V2V (vehicle-to-vehicle) and V2I (vehicle-to-infrastructure). Communicating is essential for informing humans and for ensuring a smooth traffic flow as it allows AVs to coordinate actions, especially when crossing large intersections or making a turn onto a fast-moving road (Savic, Schiller, & Papatriantafilou, 2017). The communication between the vehicles could relive the sensors in some tasks and thus solve some technical challenges such as recognizing black ice, other hazardous conditions or crashes. To enable the communication, various federally funded research programs aim to develop standards for DSRC (dedicated short-range communication) (United States Department of Transportation, n.d.). The decisive challenges to overcome are the creation, maintenance, and insurability of an ultrareliable public infrastructure. An adequate V2V communication system requires the deployment of high-level communication technology as well as communication standards and platforms. Additionally, the localization of each AV has to improve and be extremely accurate (Anderson, et al., 2016).
Thirdly, cybersecurity. Cybersecurity is a key issue because AVs heavily rely on electronic systems and thus require much higher security standards and virus detection systems to ensure functionality and passenger’s safety. Due to the strong dependency on software and thus software updates, the concern of cyberattacks is high (Driscoll, Roy, Ponchak, & Downey, 2017). Assuring software quality and reliability is an increasing challenge as AVs need to connect and perform tasks on diverse platforms. A simple software update requires access to the internet which may enable computers to attack vehicles with viruses. Also due to the connection to other vehicles, infrastructure, and the internet, AVs are more vulnerable to cyberattacks. Software and hardware of AVs need to be also locked up from users in order to prevent ‘jail breaking’ attempts. Overall, it is unalterable that cybersecurity standards of AVs are put on another level for all system in an AV as one small hacked system such as the tire pressure monitor may provide access to the entire vehicle due to the connection of the systems. A possible solution could lie in the emerging technology of blockchain as it could allow an efficient validation of transmitted information (Bordonali, Ferraresi, & Richter, 2017).
Legal and regulatory environment. Another key success factor of AVs is an adequate and corresponding legal and regulatory environment (Anderson, et al., 2016). Overall, the key concerns or challenges to be addressed are general regulatory aspects of AD and insurance or product liability regulations. A sound and modernized legal framework is necessary to enable the realization of AVs. Since the governmental dependency on the automotive industry is high, the adoption of the regulatory framework could bring many benefits from a macro perspective. Moreover, non-binding enactments and rules can be used to create a sustainable regulatory environment. Figure 4 illustrates various tools separated by the time horizon (ex ante – forward looking and ex post – backward looking) and the actor (public and private) (International Transport Forum, 2015).
illustration not visible in this excerpt
Figure 4. Types of Regulation for Automated Driving
Adapted from Automated and Autonomous Driving: Regulation under Uncertainty by the International Transport Forum, 2015. Retrieved from https://www.itf-oecd.org/sites/default/files/docs/15cpb_autonomousdriving.pdf.
The primary consideration and decision to make is to determine whether AVs should be treated specially or generally with respect to regulations. This means, that either new and standalone rules as well as agencies regulate AVs or current slightly adopted rules are in place and ordinary agencies are responsible. The next key consideration is to decide whether policy or technology should lead. This has significant implications on the uniformity or flexibility of AV regulations. While proactive policies can create legal clarity that supports investment and deployment decisions, unsuitable or unrealistic laws and regulations can freeze the development and reduce the flexibility. Due to the advantages and disadvantages, informal dialogues can be a beneficial solution in order to provide and form legal certainty for investment decisions while remaining flexible and adaptable. A further decision has to be made concerning ethical issues and the product liability and responsibility. The obligations and liabilities of owners and operators as well as OEMs have to be clarified (International Transport Forum, 2015).
In Germany, AVs are mainly treated generally and not specially, thus, the Road Traffic Regime is responsible. The German Road Traffic Regime is based on the German national law which is influenced by European and international law. In general, the technological development is much further advanced than the regulatory framework. Consequently, various non-national laws and regulations impose challenges and contradictions to the utilisation of automated functions such as the VC (Vienna Convention) or UN ECE (United Nations Economic Commission for Europe) regulations. The VC defined in Article 8(1) that a driver must be a person. In 2016, the VC was adopted and partly refers now to the UN ECE regulations. This simplifies and increases the adaptability of the legal and regulatory environment as changing only the corresponding UN ECE regulations in the future is sufficient (Norton Rose Fulbright, 2016). AD-level 2 functions such as the lane keeping and changing assistant, adaptive cruise control or parking assist bear no legal risk anymore and are already implemented in various vehicles (Roland Berger, 2014). Though, limitations remain, for example for the lane keeping and lane changing assistant. The UN ECE Regulation No. 79 allows a short period steering assistance by the vehicle before the driver has to take over again. The lane changing assistant has to be initiated by the driver via the direction indicator in order to comply with the UN ECE Regulation No.48 (Norton Rose Fulbright, 2016). Laws for AD-level 3 are installed since 2017. As a result, specific driverless functions such as an autonomous parking system on private areas are usable and will be implemented in future vehicles (Norton Rose Fulbright, 2017). The German legislation enabled road-testing for AD-level 4. This allows the driver to take the hands of the wheel and eyes of the road while the vehicle steers and breaks autonomously. In case of an accident, the driver is held responsible if the accident occurs under his watch, while the manufacturer takes responsibility when the self-driving system is in charge and the vehicle crashes due to a system failure. The law is to be reviewed in two years. Then, adjustments will be made to keep up with to the technological development and clarification regarding the data protection as well as data collection will be attempted (Wacket, Escritt, & Davis, 2017). Since testing AVs is indispensable for the future success, two critical German automobile clusters installed further testing infrastructure. Firstly, digitalization processes of the A9 motorway in Bavaria is proceeding. Secondly, test beds in cities and highways are planned including an inner-city section in Ingolstadt and a combination of motorways, highways and inner-city areas in Baden-Wuerttemberg (Norton Rose Fulbright, 2016; Daimler, 2017).
Ethical implications of AVs are not only a development consideration of OEMs but also a governmental decision. The German Federal Minister of Transport and Digital Infrastructure published a list of 20 ethical rules for the guidance of automated and connected driving. The guidelines are clear regarding a possible collision with animals or property, as humans’ lives have priority. However, the trolley problem is considered as unsolvable. Thus, the guidelines require AVs to avoid such situations in the first place. Primarily due to the reason that these complex situations cannot be standardized or programmed to replace the judgement of a responsible driver. (Federal Minister of Transport and Digital Infrastructure, 2017).
Product liability and insurance policies impose another barrier for AD (BBC, 2015). The current liability regime consists of three pillars, the driver, the keeper (owner) and the manufacturer which includes suppliers. The current German Road Traffic Act, sec. 7 states that the owner of a vehicle may be held liable for any damages caused by the vehicle, regardless of any fault since the owner bears all the risk of operation. On the other hand, according to the paragraph 18 of the German Road Traffic Act, the driver is generally responsible unless it is proven that he did not negligently cause any damages (Norton Rose Fulbright, 2016). Whereas the OEM is considered to be liable whenever a product defect occurs regardless of any negligent behaviour. While the German Product Liability Act is not very comprehensive, the liability based on the general tort law is more extensive. This law requires proof of the negligence of the OEM or the suppliers. In general, the owner and the OEM are joint debtors, thus the plaintiff may request compensation from either. Even though both are may be held reliable, one or the other may claim resource from the joint debtor due to the actual reason for the accident. In case of a technical failure, the owner or practically the insurer of the owner may request resources from the OEM or the supplier. Since AVs are likely to reduce the number of human failures, accidents caused by technical issues may become the primary reason for accidents and thus the liability shifts to the OEM. Therefore, attempts are being made to hold the OEM reliable for any damages occurred in order to simplify and clarify the liability issue. While most OEMs omit a clear commitment, Volvo decided to announce that it will take full responsibility for all technical defects, but not for accidents caused due to inappropriate use of the customer (BBC, 2015). In the US, the legal and regulatory environment for AVs is more advanced with respect to certain aspects. Up to 2017, legislations or executive orders related to AVs were passed within 26 states (see Appendix D). The state of Nevada requires a certificate of compliance which must be issued for an AV by the manufacturer or a licenced AV certification facility. The certification comes with additional cost and the compliance with several technical specifications (Anderson, et al., 2016). The major weakness in the USA is the absence of a federal law and thus the difference between the state laws. The variance imposes potential challenges and obstacles. Firstly, users might have issues when an AV is intended to operate in several states. Secondly, manufacturers face difficulties in complying with every AV law within the USA. In order to increase the deployment of AVs, the government aims to impose the first major federal legislation within 2018 which would regulate the safety requirements of AVs. As the legislation becomes effective, OEMs would be required to build AVs that are equally safe or safer than current vehicles. But, OEMs would not be penalized for missing the safety requirements in the first year (Shepardson, 2017).
Customer readiness. In order to create a successful product, the market side and the customer demand is indispensable. By 2035 more than 12 million AVs are expected to be sold per year on a global scale. This results in a market capture of around 25% of the vehicle market (Mosquet, et al., 2015). To achieve this, the population needs to be ready and accept the emerging technology. Due to the initial high price of specially designed AVs, the customers of OEMs are estimated to be on the one hand almost exclusively taxi companies and on the other hand their subsidiary mobility providers, in case OEMs are repositioning themselves in the supply chain such as VW and Daimler. In the B2B market, customer readiness depends on rational and economic arguments. Whereas in the B2C market the success of AVs depends various considerations of end-users. The step from a conventional vehicle without driver assistant function to a specially designed AV or a conventional vehicle with AD-ability is significant. Thus, incremental steps in the development of driver assistant systems and AD-levels are critical for the technology acceptance (Lipson & Kurman, 2016). As every innovation, AVs are subject to the adoption curve (see Appendix E) and in the early phase, only innovators are expected to use specially designed AVs and buy conventional vehicles with AD-ability (Rogers, 2010). Over time, more people adapt to the new technology for various reasons. The most rational reason is the decrease in cost. While in the beginning mostly companies are expected to operate specially designed AVs in a B2C sharing model due to the high costs, this is expected to change over the lifecycle of the technology and C2C (customer to customer) specially designed AV sharing could increase (Janasz, 2016). The pace and the range of adoption is difficult to predict. For example, the transition from horses to motor vehicles lasted several years. According to Chief Economist for the Consumer Technology Association Shawn DuBravac the noticeable transition might take another 20 to 25 years. The duration of transition depends on various factors. According to Lang, et al. (2016) technology acceptance and customer readiness imposed a major obstacle in the past but this is likely to vanish over time as experience and familiarity of AVs increase. Lang, et al. (2016) conducted a survey that indicates significant customer readiness of AVs in India and China (only 5% and 8% respectively where unwilling to take a ride in an AV) whereas in the US, UK and Germany the rates of unwillingness were significantly higher at 30%, 31% and 36% respectively. Nevertheless, different surveys investigating the customer readiness regarding technology acceptance, displayed significant variations in the findings depending on the formulation of the questions (Putre, 2016). This indicates the indecision of the population regarding the technology and the associated benefits and threats. While people acknowledge the benefits on the one hand, on the other hand people are being strongly influenced by the risk and unknown of the technology (Regan, Horberry, & Stevens, 2014; Jahankhani, et al., 2017). Overall, six key concerns that are holding back the acceptance of AVs can be identified. Firstly, uncertain reliability which is associated with the increased responsibility of the technology and a corresponding, potentially higher rate of failure. Secondly, safety concerns including harm caused by computational (AI) failure or programmed damage reducing accidents (ethical dilemma) as well as cyberattacks. Thirdly, giving up control and putting the personal welfare in the responsibility of a machine is a major concern. Similar issues can be found in the aerospace industry, as people are more worried about taking a plane than driving their car, regardless of their statistically increased safety (Peterson, 2016). Fourthly, AVs are confronted by scepticism as the technology is rather unknown, which in general is a typical issue of emerging technologies and new products (Friar & Balachandra, 2016). Fifthly, the lost freedom that people associate with the ability to drive instantly when and where they want (Peterson, 2016). Sixthly, many people associate fun with driving a car and thus they are not willing to give this up. However, the fun of driving a car substantially depends on the driving environment, as driving during traffic jams or in crowded urban areas is rather stressful (Lang, et al., 2016).
3 Research Aims and Contribution
3.1 Research Aim
This research paper aims to explore and investigate the status quo of the development of AVs at German OEMs. It further aims to identify future impediments until market entrance and recommend managerial actions. Due to the paper’s requirements including the limited time scope, the research project is considered to be cross-sectional in nature. This means, that it concentrates on a specific point in time and changes over a long period are not observed as it is in longitudinal research projects (Saunders, Lewis, & Thornhill, 2012). Furthermore, the paper focuses and thus primary research is conducted in the following three key pillars that are indispensable for a successful implementation of AVs: Technology, legal and regulatory environment and customer readiness.
3.2 Significance of the Study
The automobile industry is one of Germany’s strongest and most important industries (Lewis & Zitzlsperger, 2016). It is also one of the largest employers in Germany as around 800.000 people are directly and one out of seven jobs are indirectly linked to this industry (Bargende, Reuss, & Wiedemann, 2017; Smale, 2015). At this time the automobile industry is at its greatest and most critical change since the invention of the automobile (Maxton & Wormald, 2004). Traditional OEMs are changing their strategy and are now aiming to become mobility provider (Isaac & Boudette, 2016). This change requires different key skills and core competitive advantages since the primary essence of an AV is no longer the vehicle’s hardware but the software. Therefore, skills in the field of IT (information technology) are required as artificial intelligence and machine learning are the key cornerstones of AVs (Foster, Ghani, Jarmin, Kreuter, & Lane, 2017). The industrialization of AVs is a top strategic priority of German OEMs for several reasons. Firstly, OEMs are not seeking a cost but technology leadership. Hence, OEMs are mostly operating at the top of the premium passenger vehicle segment and technology advancement is a core competitive advantage. But the danger of losing the superior market position and becoming a first-tier supplier of the automobile industry is at a high-level (Abuelsamid, 2017). Apart from the market position, many OEMs have invested a significant amount of money in the development of the technology while expecting higher returns. Overall, their existence is at stake, especially if OEMs fail to commercialize the technology (The Economist, 2017). Besides, the government is also very interested in a successful implementation of the technology due to the importance of the industry to the German economy as well as the technology’s benefits for the end-users. The advantages of the technology are very appealing to Germany since the government has to react to the tightened CO2 regulations, the related emission debate regarding diesel engines and ICEs (internal combustion engine) in general, and to reduce the increasing issues of urban density and traffic jams (Arthur D. Little, 2009). But the customer readiness with respect to the technology is unclear, especially because many people still believe AVs to be a part of the far future (Rosenzweig & Bartl, 2015).
In summary, industrializing AVs is essential for the survival and the market position of German OEMs. It is also critical for the German government in order to ensure employment, maintain tax incomes, reduce emission, and offset the effects of urban density. However, many obstacles and unanswered questions are yet to be resolved and addressed to enable the implementation of the emerging technology. This research project supports and enhances the development by further investigating in this domain, identifying challenges, and proposing managerial actions.
3.3 Research Question and Objectives
Based on the purpose of this research project, the following overall research question is formulated and investigated:
What is the development status and what are the impending obstacles of AVs in Germany?
In order to uncover the formulated research question accordingly, the following three research objectives are addressed:
Objective 1
Explore the current technological development of AVs at German automobile manufacturers and identify possible future obstacles.
Objective 2
Establish the current legal and regulatory environment regarding AVs in Germany and investigate the potential future development.
Objective 3
Investigate the current customer readiness and technology acceptance of future private end-user in the German market.
4 Methodology
4.1 Research Design
4.1.1 Empirical strategy
A rigorous methodological approach is essential when it comes to researching and investigating a futuristic and complex topic such as AD with the aim of obtaining valid findings (Laws, Harper, & Marcus, 2003). Saunders et al. (2012) developed a research onion which serves as a general framework to guide the methodological reasoning underlying this research (see Figure 5).
illustration not visible in this excerpt
Figure 5. Research Onion
Adapted from Research Methods for Business Students, by M. Saunders, P. Lewis & A. Thornhill, 2012, Essex: Pearson Education.
The research onion illustrates different research philosophies which are not superior to another but rather serve different aims and are thus more appropriate for specific research topics and aims. Based on the aim of the project and the research question, pragmatism is considered the most appropriate research philosophy. This philosophy guides the research methodology in a practical and flexible manner which allows addressing the key research problem in the most beneficial way (Klenke, 2016).
[...]
[1] A disruptive technology is defined to be an innovation that replaces existing technologies and are causing significant shifts in business and social environments (Christensen, 2013).
[2] An emerging technology is a term that is relatively defined as a new technology that is currently being developed (Congress of the United States Office of Technology Assessment, 1995).
[3] Within this paper the term or phrase newcomers includes tech firms such as Waymo and Apple that are aiming to enter the automotive industry.
[4] Since AD-level 5 is perceived as a long term future prospect, in the following the single term AV refers to AD-level 4 vehicles.
[5] The term specially designed AV refers to a vehicle that is developed for AD by purpose, while the term conventional vehicle with AD-ability refers to a traditional vehicle that is upgraded with AD-abilities.