A Triple-Helix Approach for the Assessment of Hyperloop Potential in Europe

Hyperloop is a proposed high-speed surface mode of passenger and freight transport that has gained much visibility in recent years. Since its introduction in modern form in 2013, progress on Hyperloop has been thriving, with several companies involved in research and development of Hyperloop systems and subsystems. Some of them have planned testing tracks in Europe, anticipating what could be the start of commercial routes. Nevertheless, there are concerns that need to be addressed regarding the safety and serviceability performance, and further steps are necessary for the standardization and certification of the system. This study leverages the state of play of Hyperloop development, identifies issues and challenges from a European perspective, and provides policy insights towards testing and commercialization. To this end, it follows a two-tier approach that (i) addresses safety concerns related to the implementation of Hyperloop as a key enabler of the proposed technology and (ii) analyses Hyperloop technology developments using a triple-helix innovation approach, using a structured methodology based on consolidated data sources (i.e., the Scopus database of peer-reviewed literature and the Patstat patent database). The performed analyses highlight the significant amount of research and patent activity on several aspects of Hyperloop, while the mapping of activities carried out by industry in Europe, as well as from European Services, highlights the progress made towards the future Hyperloop implementation in Europe. Altogether, these findings provide factual information regarding the necessary research, policy, and industry steps taken so far that can help bring Hyperloop into the market. To the authors’ knowledge this represents a very systematic and extensive literature review on Hyperloop’s scientific and technological developments.


Introduction
Transport is one of the main pillars of society, comprising a spectrum of individual systems and their interconnections that are intended to cover the mobility demand of people and goods. Transport systems include physical and organizational elements and are characterized by an overall intrinsic complexity. These elements can influence each other directly and/or indirectly, linearly or nonlinearly, having also potential feedback cycles [1]. In this sense, the transport system can be considered as an intrinsically very dynamic complex, large-scale, interconnected, open, socio-technical (CLIOS) system [2]. Even though existing transport modes (rail, road, air, and waterborne transport) have experienced many incremental improvements regarding their efficiency, safety, and reliability, during the past few years, paradigm shifts are emerging, supported by the increasing automation and advances in information and communication technology (ICT) [3]. New technologies, social change, and regulatory pressures are creating innovation in industry segments and industry structure as technologies. Furthermore, while a lot has been made public on the technical feasibility of Hyperloop, literature on safety and security, including identification and assessment of risks (see for example [20]), is rather limited.
As a matter of fact, even if Hyperloop is in large part based on existing subsystems, already tested individually, reliability of the subsystems does not necessarily imply the reliability of the system as a whole, and safety assurance provided for a given system is specific to that system in its certified configuration [21]. New challenges will also arise from specific issues never tested before (e.g., evacuation of pods in partial vacuum conditions).
From a safety perspective, increasing (interactive) complexity, more complex relationships between humans and automation, and changing regulatory and public views of safety (the latter with various forms of oversight and regulation from the institutions) are challenging both traditional accident models and accident prevention and risk assessment techniques [22]. For this, it is common for new complex systems to define appropriate serviceability and structural integrity limit-states for the system life-cycle and link them to acceptable damage levels [23].
Considering the complexities mentioned above, there is a pressing need to inquire about the safety, security, availability, and reliability of Hyperloop. As extensive research has been done on the safety and risk assessment in other transport sectors, a suitable approach is to build on existing practices, rules, and knowhow, including the avail of existing codes and standards. These may cover transport modes (rail, aviation, but also to a lesser extent, road and waterborne), but also sectors that can provide solutions in specific areas, for example, civil engineering (for sub-structures and tubes), space technologies, or pipelines (tube-pod interfaces). Experience gained in previous studies, in particular on maglev trains, can be useful [24].
In addition, Hyperloop technologies are rapidly and dynamically evolving, something that adds to the complexity. Very recently, Hyperloop promoters presented (proprietary) technologies that alter the state of play in the field. These developments may lead to performance improvements, but also to possible vulnerabilities that need to be assessed.
Overall, as Hyperloop is a new technology operating under conditions that reach the edge of human endeavor, it inevitably raises concerns about the safety of its operation. There is no doubt that travelling for long distances at almost sonic speeds, under near vacuum conditions, and with very limited or zero maneuverability, constitutes a high hazard situation. Still, the level of risk can be reduced to acceptable levels, provided that the appropriate safety systems are put in place (see for example [25]).
Considering the above, this study recognizes that energy consumption, safety and serviceability, and financial feasibility are essential elements contributing to the sustainability of the Hyperloop technology, as well as to its importance in contributing to the global sustainable goals. Thus, it follows a two-tier approach for the analysis, which (i) addresses the safety concerns and provides preliminary risk-analysis considerations for the safe implementation of Hyperloop, and (ii) analyses Hyperloop technology developments, applying the methodology co-developed by one of the authors for the European Commission's Transport Research and Innovation Monitoring and Information System (TRIMIS) [26].
To achieve this objective, a triple-helix innovation approach is used [27], in which efforts from academia (scientific publications), the industry (patents), and the government (in the form of possible initiatives linked to future policy actions) are assessed ( Figure 1). The reasoning behind this approach, is that in the case of Hyperloop, which is fundamentally a new system, it creates the need for new basic research that focuses on new processes and challenges. Likewise, as the technology becomes more mature, adequate governance needs to address it from a policy perspective, focusing on safety and accounting for other issues such as standardization and certification. The whole process is anything but linear, and several links between the three elements are possible. By identifying and clustering efforts from the academia, the research interest is highlighted, focusing on individual technical issues, while the patent activity shows the trend of the interest from industry, investing in Hyperloop technologies. Finally, efforts from the government show the interest from the people's perspective in fostering Hyperloop developments.

Methodological Background
In order to study the problems linked to the deployment of Hyperloop systems, an inclusive and holistic system of systems approach is necessary, accounting for interdependencies and interactions between the different systems and subsystems that support and interact with the Hyperloop ecosystem. To this end, this study starts with a system decomposition that can be a starting point for risk-analysis considerations and helps identify issues and areas of Hyperloop research. The rest of this section focuses on the methodology behind the scientific research analysis and the intellectual property (patent) analysis.

System Decomposition
While Hyperloop is a new transportation mode in which there is limited or no experience on its operation as a whole, there is sufficient experience in the operation of many of its subsystems, as well as experience in similar systems from other sectors, which, after thorough analysis of the risks, can address all safety challenges and guarantee that risks remain low at all times.
A first step is to perform a system decomposition of Hyperloop into different parts and analyze their interdependencies (including interactions with external systems). This is because the decisions taken, especially in the design process, are based on the knowledge obtained about the object of study [28].
This can be done using a system approach in which the lowest aggregation breaks down the Hyperloop system. Four macro areas are identified: • Substructure (including foundations and bridge work) • Tube Figure 1. A triple-helix approach to assess Hyperloop developments (source: own elaboration).
By identifying and clustering efforts from the academia, the research interest is highlighted, focusing on individual technical issues, while the patent activity shows the trend of the interest from industry, investing in Hyperloop technologies. Finally, efforts from the government show the interest from the people's perspective in fostering Hyperloop developments.

Methodological Background
In order to study the problems linked to the deployment of Hyperloop systems, an inclusive and holistic system of systems approach is necessary, accounting for interdependencies and interactions between the different systems and subsystems that support and interact with the Hyperloop ecosystem. To this end, this study starts with a system decomposition that can be a starting point for risk-analysis considerations and helps identify issues and areas of Hyperloop research. The rest of this section focuses on the methodology behind the scientific research analysis and the intellectual property (patent) analysis.

System Decomposition
While Hyperloop is a new transportation mode in which there is limited or no experience on its operation as a whole, there is sufficient experience in the operation of many of its subsystems, as well as experience in similar systems from other sectors, which, after thorough analysis of the risks, can address all safety challenges and guarantee that risks remain low at all times.
A first step is to perform a system decomposition of Hyperloop into different parts and analyze their interdependencies (including interactions with external systems). This is because the decisions taken, especially in the design process, are based on the knowledge obtained about the object of study [28].
This can be done using a system approach in which the lowest aggregation breaks down the Hyperloop system. Four macro areas are identified: • Substructure (including foundations and bridge work) Interface (pod-tube) Figure 2 provides a schematic of the decomposition.  This decomposition is used to identify possible risk issues, as well as cluster research issues from scientific literature and patent topics from patent activity in Section 3.

Scientific Research and Patent Analysis
Even though Hyperloop is a relatively new system, research has already been performed on its performance and safety challenges, as well as on other issues (e.g., commercial feasibility). In addition, Hyperloop promoters have registered proprietary technologies through patent filings.
The objectives of the methodology presented below are to capture research findings and to identify international patent activity, in order to identify trends and key research organizations and countries.
To achieve the first objective, the Scopus database was used, which has strict indexing rules, something that makes the quality of the results relatively valid. For the analysis, the following steps were taken:

•
A search was carried out focusing on the keywords in "Hyperloop" or "tube transport" in the abstract, title, or keywords. Results were limited to those after 2013 (when the modern concept of Hyperloop was introduced) and to conference or journal papers (articles). The exact query used was: TITLE-ABS-KEY ("Hyperloop" OR "tube transport" OR "tube transportation") AND PUBYEAR > 2013 AND (LIMIT-TO (SRCTYPE, "j") OR LIMIT-TO (SRCTYPE, "p")). This search, performed in September 2020, provided 162 results.

•
A further manual filtering of the documents based on the title or abstract one-by-one limited the results to 91 documents. The aim of this filtering was to eliminate those documents that were not relevant to the field due to lexical ambiguity and those that simply outlined Hyperlooprelated aspects. This left 52 articles and 39 conference papers.

•
After this step, all papers were read and analyzed. Following that, an attempt was made to assess quantitatively research under several clusters. The decomposition from Section 2.1 was adopted, together with four additional vertical clusters (energy, safety, aerodynamics, and operations) and two that focused on general discussions and Hyperloop network development.
To achieve the second objective, the most recent Patstat dataset was used [29], which covers patents published up to early spring 2020. The dataset is currently being implemented in the European Commission's Transport Research and Innovation Monitoring and Information System (TRIMIS) database [30]. While data accuracy in Patstat is not without concerns [31], the performed analyses aimed to overcome this issue with adequate filtering and by checking the results one-byone. For the analysis, the following steps were taken: • First, the relevant cooperative patent classification (CPC) codes were identified. In this case, CPC code B61B13/10 ("other railway systems, tunnel systems (pneumatic tubes conveyors)") was the most relevant to Hyperloop. This decomposition is used to identify possible risk issues, as well as cluster research issues from scientific literature and patent topics from patent activity in Section 3.

Scientific Research and Patent Analysis
Even though Hyperloop is a relatively new system, research has already been performed on its performance and safety challenges, as well as on other issues (e.g., commercial feasibility). In addition, Hyperloop promoters have registered proprietary technologies through patent filings.
The objectives of the methodology presented below are to capture research findings and to identify international patent activity, in order to identify trends and key research organizations and countries.
To achieve the first objective, the Scopus database was used, which has strict indexing rules, something that makes the quality of the results relatively valid. For the analysis, the following steps were taken: • A search was carried out focusing on the keywords in "Hyperloop" or "tube transport" in the abstract, title, or keywords. Results were limited to those after 2013 (when the modern concept of Hyperloop was introduced) and to conference or journal papers (articles). The exact query used was: TITLE-ABS-KEY ("Hyperloop" OR "tube transport" OR "tube transportation") AND PUBYEAR > 2013 AND (LIMIT-TO (SRCTYPE, "j") OR LIMIT-TO (SRCTYPE, "p")). This search, performed in September 2020, provided 162 results.

•
A further manual filtering of the documents based on the title or abstract one-by-one limited the results to 91 documents. The aim of this filtering was to eliminate those documents that were not relevant to the field due to lexical ambiguity and those that simply outlined Hyperloop-related aspects. This left 52 articles and 39 conference papers.

•
After this step, all papers were read and analyzed. Following that, an attempt was made to assess quantitatively research under several clusters. The decomposition from Section 2.1 was adopted, together with four additional vertical clusters (energy, safety, aerodynamics, and operations) and two that focused on general discussions and Hyperloop network development.
To achieve the second objective, the most recent Patstat dataset was used [29], which covers patents published up to early spring 2020. The dataset is currently being implemented in the TRIMIS database [30]. While data accuracy in Patstat is not without concerns [31], the performed analyses aimed to overcome this issue with adequate filtering and by checking the results one-by-one. For the analysis, the following steps were taken: • First, the relevant cooperative patent classification (CPC) codes were identified. In this case, CPC code B61B13/10 ("other railway systems, tunnel systems (pneumatic tubes conveyors)") was the most relevant to Hyperloop. • Second, the earliest filing year was set as 2014 or later. • Third, the patents belonging to the same patent family (i.e., set of patents taken in various countries to protect a single invention) were counted once, keeping the earliest entry.

•
Finally, patent descriptions were checked one-by-one to assert that they were relevant to Hyperloop technologies, leading to the identification of 46 patents. Table 1 reassumes the search coverage of the two tasks. For additional details on the methodology, the reader is referred to the TRIMIS methodology for the identification and assessment of new and emerging technologies and trends (NETT) in the transport sector [32,33].

Analyses and Results
Focusing on the considerations made in Section 1 and the methodology presented in Section 2, the following paragraphs present and discuss the main findings of the analysis.

Preliminary Considerations for Risk Analysis
As Hyperloop is a new technology operating under conditions that reach the edge of human endeavor, it inevitably raises concerns about the safety of its operations. There is no doubt that travelling for long distances at almost sonic speeds, under near vacuum conditions and with very limited or zero maneuverability, constitutes a high hazard situation.
As a first step, starting from the decomposition performed in Figure 2, it is possible to identify possible legislative, technological, and other challenges and enablers for the introduction of Hyperloop, considering what has been done in aviation, rail, and road, as well as in other sectors ( Table 2). Regarding the substructure for example, there are strong similarities with highway and railway bridge structures. However, there are points for further investigation compared to the state of art: the serviceability and safety limit-states need to be addressed within Hyperloop requirements, and the loading system has different characteristics. The European Standards for Structural Design (Eurocodes) provide the framework for the structural analysis [34]. Hyperloop substructure design can be facilitated by recent work on the standardization needs for the design of underground structures in Europe ( [35]).
The tube itself, shares similarities with tunnel structures. Minimum superstructure requirements for safety and serviceability need to be considered. In this case, the Hyperloop system is more confined and controlled, something that makes less likely the occurrence of catastrophic fires, which were a major issue for road tunnels in particular.
The pod structure is very innovative, yet it shares many similarities with rail and aviation. Acceleration and deceleration limits (induced by passenger stability and wellbeing limits), both in normal service and in emergencies, are crucial for the definition of minimum safety distances (safety headways) that have a direct impact on the system capacity. The redundancy and reliability of life support systems in the case of a catastrophic event need to be tested, and an adequate evacuation time needs to be guaranteed, considering the internal pod geometry. Acceptable tube life conditions and evacuation routes within the tube (including safety exits) need to be guaranteed for the evacuation.
Finally, regarding the pod-tube interface, considering the low pressure in the tube, there are strong similarities with space applications. Standards for docking could be applicable [36].
Addressing risks within a structured risk acceptability criterion and using advanced modelling methods (see for example [37]) is essential. Risk analysis as defined in many International Codes and Standards (e.g., ISO 31,000 [38]) or guidelines ( [39,40]) usually incorporated in a global process of risk management.
Furthermore, evacuation procedures need to be carefully planned, especially when considering the network aspect of Hyperloop, which means having switches for diverging tracks. Failure modes of bridges, tunnels, and pipelines (e.g., from the thermal expansion) need to be considered.
On top of that, natural events triggering technological accidents (Natech) need to be accounted for [41], including multi-hazard risks [42][43][44]. Natural events can also have a short-or long-term effect on the track stability and can lead to performance drops.
Natural events include: • Landslides/geotechnical issues (related to the soil morphology and the soil-structure interaction).

•
Floods, which can also cause corrosion and displacement of the substructure and eventual damage to the tubes system. • Earthquakes (for earthquake prone areas).
Critical infrastructures, including the transport network, are in general prone to cyberattacks. Security risks (including cybersecurity) are important, as terrorist attacks can have severe consequences, while the long distances make for difficult surveillance.
All of the safety-related concerns, which were briefly accounted above, need to be addressed through a thorough risk assessment, performed rigorously according to the ISO 31,000 standard and taking into account the engineering standards for the design and operation of the different subsystems. This standard defines the steps of a systematic approach to assess the risks and provide insights to the triplet of fundamental questions: • What can go wrong (and what can cause it)? • How frequently can it happen? • What will be the consequences?
This analysis should cover all hazards (i.e., structural, electro-mechanical failures, Natech, security) and should be accompanied by risk evaluation and assessment of the need to introduce additional safety measures (risk treatment), with the goal to keep risk as low as reasonably practicable (ALARP). Focusing on the proposed macro-level Hyperloop decomposition, there is an indicative connection between the hazards discussed above and their linking to the Hyperloop system (Table 3). For each hazard-part, the level of interaction is identified from indirect/weak (+) to direct/strong (+++). An example of the first steps of such an approach is given in [20], where an established hazard identification method (FMECA-failure mode, effects and criticality analysis) is applied to identify and assess the most safety-critical failures in order to conclude that the door failure is the most critical failure of the pod. Following this step, it is expected that a systematic assessment of the preventive and mitigation measures would take place, for example through a bow-tie approach, to identify the measures necessary to ensure a desired safety integrity level (SIL).
As Hyperloop will operate in extreme conditions, with very limited margin for errors, it is important to address all safety issues in the design phase. Nevertheless, human errors cannot be excluded, especially when linked with maintenance operations, changes of procedures, and so forth., as it was tragically demonstrated at the Lathen Maglev train accident in 2006 [45]. These potential errors need to be properly taken into account in the systems' design phase.
Indeed, it has been demonstrated in several technological sectors that accidents trigger both new research and new regulation. Research is triggered immediately after an accident, in order to investigate the causes and to develop the necessary prevention and mitigation measures to prevent its occurrence in the future. However, regulatory needs also manifest themselves after an accident as a direct reaction of the society to the losses. For example, the Three Mile Island nuclear accident in 1979 triggered research that led to a completely new field, the quantitative risk assessment (QRA), with numerous applications in all technological sectors, from nuclear to chemical and to aerospace. It also gave birth to a new philosophy of regulation, the risk-informed safety regulation. Similarly, the Seveso accident in 1976 gave birth to the Seveso Directive for the regulation of the chemical industry, while the big tunnel fire at the Mont Blanc tunnel in 1999 resulted in dedicated legislation for the safety of road tunnels, as well as in the development of tools for the reduction of risks. Therefore, a triple-helix approach and relevant interlinks can be observed among accidents, R&I, and regulation, aiming at a continuous improvement of safety and finally safer and more sustainable solutions.
It is important for a new and emerging technology to ensure and demonstrate that all risks have been taken into account and are managed to be as low as possible. The role of regulators, standardization bodies, and the insurance industry are essential, especially during the first steps of the new technology (see, for example [46]). Finally, a system for sharing lessons needs to be established amongst the Hyperloop promoters as soon as possible to ensure that lessons from failures are identified and learned.
Finally, user acceptance is very important. Hyperloop should have high safety standards, and for this reason individual safety target risk limits should be aligned with the values currently implemented in railway and civil aviation. The railway fatality risk (fatalities per million train-km) in Europe (EU-28) in the period 2012-2016 was lower than 0.6 and the passenger fatality risk (passenger fatalities per billion passenger train-km) was lower than 0.11 [47]. In aviation, the fatality risk, measured as the number of full-loss equivalents per 1 million flights, is constantly declining and was 0.11 in 2019 [48].

Academic Research on Hyperloop
Following the steps taken in Section 2.2, Figure 3 shows the outcome of this exercise. As can be seen, the publication trend is positive, considering also that results for 2020 cover only a few months. Following an assessment of the 91 abstracts, the clusters are further defined in Table 4 using a bottom-up approach, and each document is tagged with one cluster from the vertical and one from the decomposition or with one generic cluster (discussion or network). As data are limited to Scopus scientific research, the research topics identified in literature can act as a baseline for future developments, as it is plausible that many Hyperloop promoters carry out research on specific topics without publicly communicating the results. With the exception of 20 documents on Hyperloop discussions and eight documents on Hyperloop network developments, the one-to-one relationship between vertical and decomposition clusters in the remaining documents is shown in the chord diagram of Figure 4.  Following an assessment of the 91 abstracts, the clusters are further defined in Table 4 using a bottom-up approach, and each document is tagged with one cluster from the vertical and one from the decomposition or with one generic cluster (discussion or network). As data are limited to Scopus scientific research, the research topics identified in literature can act as a baseline for future developments, as it is plausible that many Hyperloop promoters carry out research on specific topics without publicly communicating the results.
With the exception of 20 documents on Hyperloop discussions and eight documents on Hyperloop network developments, the one-to-one relationship between vertical and decomposition clusters in the remaining documents is shown in the chord diagram of Figure 4.
This kind of visualization can help identify the most researched topics and their relationships, and at the same time identify deficiencies in research. As can be seen, most research is dedicated to pod and operations (18 documents), pod and aerodynamics (8 documents), and interface and energy (8 documents). Scientific research on safety issues is rather limited, addressed only in 6 documents (three focusing on the pod, two on the tube, and one on the pod-tube interface). Table 5 provides a selection of recent technology related topics and approaches from the above review, restricted to the last 4 years (from 2017 and onwards) and technology related issues. Although the literature sample considered is not very large, and thus findings cannot be necessarily generalized, they still provide an insightful view on Hyperloop research.  This kind of visualization can help identify the most researched topics and their relationships, and at the same time identify deficiencies in research. As can be seen, most research is dedicated to pod and operations (18 documents), pod and aerodynamics (8 documents), and interface and energy (8 documents). Scientific research on safety issues is rather limited, addressed only in 6 documents (three focusing on the pod, two on the tube, and one on the pod-tube interface).

Authors and Year Issue Principal Method
Lim et al. 2020 [49] Optimization of a high-temperature superconducting magnet for subsonic transportation.

Simulation
Nick and Sato 2020 [50] Computational fluid dynamics simulation of Hyperloop pod predicting laminar-turbulent transition; aerodynamic analysis for two pod designs (short and optimized).

Simulation (FEM)
Connolly and Costa 2020 [52] Interaction between moving vehicle and vibration characteristics of the underlying soil-guideway system. Influence of the vehicle axis position.

Simulation
Chaidez et al. 2019 [53] Minimization of rolling and contact levitation friction. Analytical Guo et al. 2019 [54] Design and analysis of a plate type electrodynamic suspension structure for a proposed high-speed maglev/tube system.

Computational (FEM)
Guo et al. 2019 [55] Study of a null-flux coil electrodynamic suspension structure for evacuated tube transportation.  Figure 5 shows the outcome in terms of number of patents granted by applicant sector and by the country of origin of the applicant. The majority of patents (27) come from private companies, while nine are filed by government/not profit organizations, nine from Universities, and one by an individual. China leads with 21 patents, followed by the US (13) and Korea (nine), while there are three patents granted from EU Member States, namely France, the Netherlands, and Spain. Focusing on the sector-country split, it is interesting to observe that all 13 US patents come from the private sector (including six from Hyperloop Technologies Inc. and four from the Boeing company), and all nine Korean patents come from government/non-profit organizations (six from the Korea Railroad Research Institute and three from the Korea Institute of Construction Technology). CRRC Yangtze Co. leads the entities with patents in China with five cases.   There are still some concerns regarding the analyses, considering that it is a very recent technology, so patents are filed under various patent codes and data issues are common. Furthermore, a similar search on the Espacenet [69], a free online service for searching patents and patent applications, revealed a much larger volume of patent activity, especially from China. However, Espacenet also covers patent applications (not necessarily granted) and cannot be considered as reliable as Patstat, while other reasons (e.g., linked to political decisions) may explain the high number of patent filings from China.

Patent Activity on Hyperloop
In a similar way to what has been done for scientific research, an effort is made to map these patents within the clusters identified in Section 2.1. Table 6 provides the results. An additional cluster has been added (other) that gathers mostly patents on complete systems and logistics and freight operations. Table 6. Hyperloop clusters and corresponding patent topics from Hyperloop patent analysis (source: own elaboration).

Clusters
Patented Topics Decomposition clusters 1 Substructure concrete tube construction; tube sealing construction; tube concrete slab; tube support; underground construction;    There are still some concerns regarding the analyses, considering that it is a very recent technology, so patents are filed under various patent codes and data issues are common. Furthermore, a similar search on the Espacenet [69], a free online service for searching patents and patent applications, revealed a much larger volume of patent activity, especially from China. However, Espacenet also covers patent applications (not necessarily granted) and cannot be considered as reliable as Patstat, while other reasons (e.g., linked to political decisions) may explain the high number of patent filings from China.
In a similar way to what has been done for scientific research, an effort is made to map these patents within the clusters identified in Section 2.1. Table 6 provides the results. An additional cluster has been added (other) that gathers mostly patents on complete systems and logistics and freight operations. Table 6. Hyperloop clusters and corresponding patent topics from Hyperloop patent analysis (source: own elaboration).

Clusters
Patented Topics Decomposition clusters 1 Substructure concrete tube construction; tube sealing construction; tube concrete slab; tube support; underground construction; There are still some concerns regarding the analyses, considering that it is a very recent technology, so patents are filed under various patent codes and data issues are common. Furthermore, a similar search on the Espacenet [69], a free online service for searching patents and patent applications, revealed a much larger volume of patent activity, especially from China. However, Espacenet also covers patent applications (not necessarily granted) and cannot be considered as reliable as Patstat, while other reasons (e.g., linked to political decisions) may explain the high number of patent filings from China.
In a similar way to what has been done for scientific research, an effort is made to map these patents within the clusters identified in Section 2.1. Table 6 provides the results. An additional cluster has been added (other) that gathers mostly patents on complete systems and logistics and freight operations. It should be noted that there are several (seven in total) earlier patents (from 2015 and 2016) that focus on a Hyperloop complete system. In recent years, and as Hyperloop research has become more mature, patents have focused on specific aspects.
This "intellectual property mapping" helps identify possible areas for synergies within Hyperloop promoters, and identify technology spillovers to other transport modes (rail in particular). The identification of common areas of research, designated by patent activity, is also crucial for the collaboration of different Hyperloop promoters in research and testing toward the interoperability of the system. Finally, new technological developments lead to new sources of risk that need to be accounted for in a continuous risk assessment and monitoring process.
Furthermore, it is possible to identify company-specific innovation and possibly identify areas for further research. As an example, Hardt Hyperloop presented a patented technology, "the switch", a track changing mechanism, which allows the vehicle to pass from one track to another and opens up the concept of networks. This adds additional complexities, as the absence of lane switches made Hyperloop an inherently safer mode. Hyperloop Transportation Technologies also filed patents that have the potential to overcome propulsion challenges within the tube.
It is difficult to draw specific conclusions on the connection between academic research and patent activity. In a modern triple-helix context, funded academic scientific activity leads to research results or applications, which in turn can lead to further development and commercialization from companies [70]. That said, the relationship is far from linear, considering the increasing role of universities as technology hubs and thus their tendency to capitalize research results through the creation of incubators, technology hubs, university spin-offs, and so forth [71]. This is also evident in the analyses, with nine patents coming from Universities. Nevertheless, focusing on the funding sponsor in the scientific analyses, only the Korea Railroad Research Institute is acknowledged as such on two occasions. Notably, Hyperloop Technologies Inc. and The Boeing company (present with 10 patents in the analysis) are not acknowledged as funding sponsors in any research papers. When focusing on the patent descriptions and the scientific research papers, no apparent connections are found among them. A possible explanation is that much of the research from private companies is done in-house or under disclosure agreements.
There are still some limitations of the method applied in this study. Data sources are limited to Scopus scientific research (articles and conference papers) and patents found in the Patstat database. More importantly, it is plausible that many Hyperloop promoters carry out research on specific topics without communicating the results. Consequently, these results should be interpreted with attention.

Hyperloop Developments in Europe
This section provides an overview of recent Hyperloop developments in Europe that are not covered either by basic research or patent activity and may call for the need of a policy framework. Even though there are no official policy developments at an EU level, initiatives are taking place in Europe by Hyperloop promoters, which are expected to have an impact and trigger policy developments.

Initiatives from Hyperloop Promoters
Initiatives focus primarily on (i) the creation of testing sites (also creating a connection with local communities), (ii) initiatives to raise funds (regional and private in particular), and, more recently, (iii) initiatives aiming for the standardization of the Hyperloop ecosystem.
Regarding the first point, several Hyperloop promoters are planning testing centers in Europe.   [75].
Finally, on standardization, a big challenge for Hyperloop is the standardization (and certification), including infrastructure. In July 2020, TÜV SÜD and Hyperloop TT published a set of generic guidelines for design, operation, and certification of Hyperloop [76].
The above-mentioned initiatives, culminated in the last two years, are an indication that policy developments are needed to address important issues such as standardization, certification, and above all, safety, both for the testing period but also for future deployment.

Developments from European Services
Although Hyperloop is not addressed in current European policies, there are EU initiatives regarding Hyperloop research and innovation potential, while two more recent initiatives focus on operational aspects and the standardization of Hyperloop.
The 2019 Strategic Transport Research and Innovation Agenda (STRIA) roadmap for connected and automated transport (CAT) identifies mergers between rail and air in Hyperloop as technology and experience transfer potentials from CAT research [77].
In January 2020, the Shift2Rail Joint Undertaking (S2R JU) launched a coordination and support action (CSA) focusing on the concepts of operations, safety cases, functional specification, and operational conditions [78].
In February 2020, the European Committee for Standardization (CEN) and the European Committee for Electrotechnical Standardization (CENELEC) launched a new joint technical committee (TC), CEN/CLC/JTC 20, dedicated to Hyperloop system standardization [79]. The new Joint TC will also benefit from the experience of the rail, space, and pressure equipment standardization in which CEN and CENELEC are already active and in particular with the CEN/TC 256 'Railways applications'. Hyper Poland, Hardt Hyperloop, TransPod, and Zeleros Hyperloop support this initiative.
Beyond Europe, policy developments take place that could trigger policy collaborations. In 2019, the U.S. Department of Transportation (DOT) created the Non-Traditional and Emerging Transportation Technology (NETT) Council, an internal deliberative body tasked with identifying and resolving jurisdictional and regulatory gaps that may impede the deployment of new technology, including Hyperloop [80]. All this considering that new transport technologies may not always fit precisely into the existing regulatory structure.

Conclusions
Hyperloop is a new transport technology that has the potential to play a key role toward a future decarbonized transport system. This paper provides a first look at issues and challenges for the successful testing and commercialization of Hyperloop from a European perspective. It also maps Hyperloop intellectual property filings that can help identify both risk sources and research spillovers. The issues are presented in a structured manner and are commented on regarding the way forward, providing links to existing systems, analyses, and standards.
The proposed three-step process and the system decomposition provides support to the identification of Hyperloop technological developments and the deriving risks, thus offering insights into the sector's stakeholders, while backing the current transport systems' decarbonization through technological advances. To the authors' knowledge, this represents a very systematic and extensive literature review on Hyperloop's scientific and technological developments.
Regarding scientific research carried out on Hyperloop, a first analysis focusing on identified Hyperloop clusters (substructure-tube-pod interface) leads to the following conclusions: • most of the research is dedicated to pod and operations (18 documents), pod and aerodynamics (eight documents) and interface and energy (eight documents). • research on safety issues is rather limited, addressed only in six documents (three focusing on the pod, two on the tube, and one on the pod-tube interface).
Regarding patents granted on Hyperloop: • it is interesting to observe the differences between countries, with all 13 US patents coming from the private sector and all nine Korean patents coming from government/non-profit organizations. • patents granted earlier (2015-2016) focus on the entire Hyperloop system and therefore are more conceptual. In the last years, patents focus on specific technical issues, something that shows the higher maturity achieved.
These findings, despite the limitations in the availability of data (restricted to Scopus and Patstat datasets), represent a very systematic literature analysis on Hyperloop scientific and technological developments. In addition, the mapping of activities carried out by industry in Europe, as well as from European Services, highlights the progress made toward future Hyperloop implementation in Europe and the dynamics of the technology.
The findings can also provide a starting point for policy support for future decisions regarding Hyperloop testing and deployment in Europe. For example, policy makers can use these findings to obtain a snapshot of the research and patent activity, and consequently an indication of the maturity of Hyperloop subsystems and functionalities. This process will require, among other thing, continuous identification and analysis of issues, also beyond technology, that will help pave the way towards the demonstration and implementation of Hyperloop. Beyond the importance of this analysis for policy makers, this paper and the associated triple-helix assessment methodology can make several contributions to transport scholars and practitioners who are interested in the development of Hyperloop technologies, for example by identifying synergies between research topics identified for Hyperloop and research topics relevant to existing transport modes.