Courtesy : www.sciencedirect.com
Green e-vehicle design
Globalization and financial development stimulate international collaboration and removing trade constraints, thereby those amplify the intensity and extensity of activities in transportation . However, the fossil fuel dependent system in the industry brings far-reaching environmental issues such as vehicle exhaust emissions and climate change, affecting ecological balance, living conditions, and human health . The air quality was improved through fuel-efficient technology substitution but challenges in climate change remain. Given 24% emissions of greenhouse gases come from transport section, from which ground transport account for 72% and keep growing, the stakeholders need more engagement and efforts for green transportation transition . Hence, electric vehicles (EVs) born as the strategic solution to achieve the goals of decarbonization, ecological balance, commercialization, and technology innovation in transportation sector.
EVs are referred to road-used vehicles rely on electric powertrain and plug-in charging approach, including battery electric vehicles (BEVs), plug-in hybrid electric vehicles (PHEVs), and fuel cell electric vehicles (FCEVs) . The sustainable development of the EV industry aims at ecological and economic benefits in ecosphere for long-term scope, but the widespread adoption of electric driving is affected by salient uncertainty, e.g., geographic heterogeneity of policy preference, upstream and operational emissions, end-of-life waste, and technological bottlenecks in supply chain .
Scholars have found factors in several parties (e.g., government, industry, and market), that influence the commercialization and development process of the EV industry, such as patent and prototype counts, research and development (R&D), fuel economy, subsidy policy, social impact, and keen interest of investment . The relevant studies mainly analyze specific topics of the EV industry, but only a few have comprehensively reviewed the industry development. Publication in scientometrics field mounts for the past decade, and researchers have unveiled diversified research fronts as technology innovation goes on. The collected findings in publishing outputs, patterns of publication, topic categories, and international productivity were used to comprehensively review the trends of industry development . Thus, scientometrics-based method may work facing the complexity, multiformity, and uncertainty in the emerging EV industry development.
In the future, the success of EV industry sustainable development will highly depend upon the question whether key stakeholders (e.g., governments, the public, and suppliers) fulfill their commitment, involvement, and efforts . Thus, there should be further research to systematically analyze the salient stakeholders amid and to support the sustainable pathway of the industry. Stakeholder is defined as any group or individual influenced by or can affect the achievement of an organization’s objectives, and this concept is established to address three business problems: sustainable value creation, capitalism ethics, and managerial mindset . Although stakeholder theory originates from traditional shareholder theory, the goal of stakeholder theory is to set up a more equilibrated distribution of the benefits from operating results between non-shareholders and shareholders . Scholars widely use stakeholder theory to evaluate business evolution, ethical dilemmas, corporate governance, corporate social responsibility, mergers and acquisitions, and other organizational issues in a comprehensive level . Given development trends and the significance of stakeholders, the stakeholder engagement system deepens the research from a systematic perspective to support the sustainable development of the industry. Regarding the global industry development, the system explores the driving forces behind (e.g., environmental concerns, energy security, corporate strategy, and industrial strategy), and analyzes the salient stakeholders and their functional roles for sustainability in this emerging industry.
Generally, this study combines the scientometrics analysis on existing research to explore the development trend, and a stakeholder engagement system to support the sustainable development of the EV industry. It firstly explores the EV industry development trends by a scientometrics-based data evaluation system (SDES), as well as forward-looking analysis. Based on the SDES, three extracted keywords depict the development trajectory and research fronts, as well as help detect the potential forces driving the industry. It tries to provide a well-rounded view of the development trends in the past and possible directions in the future. Based on detected trends and significance of stakeholders in the emerging industry development, the stakeholder engagement system is established to support EV industry sustainable development. The remainder of this study organizes as follows. Through analysis of SDES the development trends are detected and visualized in Section . The results of SDES by keywords retrospect, trend analysis, and forward-looking analysis are illustrated in Section . Then, Section proposes the stakeholder engagement system, and the important elements within the system are analyzed to support the sustainable development of the EV industry. Finally, Section 5 draws the conclusions.
2. Methodology
As the EV market develops at an unprecedented pace, the demand for analyzing the trend and direction in the blooming business is rising. Because the renewable energy and transportation electrification sectors are subject to periodic technological revolutions, identifying the significant and cutting-edge research can be challenging. Consequently, this study uses SDES to gain insights into the connection and interaction between the prominent scientific papers. The scientometrics mining tool is built by period, historical keywords, and selected fields, to uncover the evolutionary trajectory and key areas in the EV market.
2.1. SDES
The process of proposed SDES includes data collection, data analysis, result visualization, and data update, as the system shown in Fig. 1. The SDES is based on the Web of Science (WoS) database and CiteSpace. The data collection starts at WoS, and the search continues both at the Science Citation Index Expand (SCIE) and Social Science Citations Index (SSCI) databases, the most reliable and widely used sources for scientometrics research, for all these associated literature reviews. Thereafter, the data was standardized in the format of Institute for Scientific Information (ISI), and the data update and result visualization were executed by CiteSpace. The results are conducive to understand the dynamic networks, unveil the potential tendency, and furtherly predict the development of the industry.
Most pertinent research is acquired from ISI WoS through advanced search in three dimensions, that is “Xi”, “Yi” and “Zi”. New energy and electric are included in the “Xi” keyword; “Yi” relates to vehicles, automotive, etc.; and “Zi” contains the keywords such as corporation and company. From 1990 to 2020, 660 publications in total are found in the data analysis.
As shown in Table 1, with the help of Jaccard similarity coefficients to measure the similarity between the keywords, the networks can be built by the pruning approach to “Minimum Spanning Tree”, therefore, keywords can provide objective definition as the nodes. Then, the term frequency-inverse document frequency (TF*IDF) algorithm was applied to extract cluster labels. The tendency of the EV can be illustrated by different views of results visualization, i.e., spectral clusters, timeline, and time zone.
Table 1. A review of selected clusters and associated keywords.
| Cluster ID | Size | Silhouette | Mean (Year) | Label (TFI*DF) | Keywords in clusters |
|---|---|---|---|---|---|
| 6 | 27 | 0.778 | 2006 | Vehicle Exhaust Emissions | 18 policy; 12 hybrid electric vehicle; 10 fuel consumption |
| 9 | 25 | 0.658 | 2010 | Climate Change | 30 renewable energy; 26 performance; 23 sustainability |
| 10 | 25 | 0.628 | 2015 | Integration | 46 optimization; 41 management; 19 demand response |
2.2. Scientific mapping
Scientific mapping is a powerful scientometric methodology for examining the conceptual structures in targeted fields . This method can monitor certain fields and delimit study regions to detect formation and development. Science mapping, which is based on domain knowledge, shows the process and structure of scientific knowledge growth, exposing complex linkages between network structure, knowledge groups, concepts, and overall evolution . According to Kuhn]; iterative revolutionary process is the science advance while a paradigm is a well-known model or pattern. Kuhn also produced the scientific development theory, the model that introduce the CiteSpace for its predictive function in scientific knowledge mapping .
Scientific progress is a historical scientific revolution In the pre-scientific period, science was not distinguished by a congruent paradigm, however, a new paradigm is required for a scientific revolution since science evolves via continuous problem resolution, which brings the revolution of science. When it enters a new paradigm, a new normal comes forth. As a result of the combination of CiteSpace and the method of scientific advancement, clustering accumulation, diffusion, and evolution of scientific citations upbuild the framework of a knowledge map, revealing the development of knowledge emergence and cutting-edge domains.
CiteSpace can track and detect the development trends via identifying co-citation clusters in the literature. CiteSpace’s core technologies include its spectrum clustering and features selection algorithms, as well as the visualized display of findings to assist users in comprehending research patterns and finding research fronts. The timeline view of Cite Space visualizes and recognizes the changes in the electric cars sector across publication time . CiteSpace also generates the clusters to model the trajectory of the EV industry development trend.
With the help of algorithms of clustering and feature selection for standardized data, CiteSpace is capable to recognize research trends with co-citation clusters. The visualization of clusters can track the pathway of EVs while the timeline view provides insight into the temporal evolution of the industry.
2.3. Data clustering and visualization analysis
The following part of this study moves on to describe in greater detail the views of visualization. Fig. 2 shows the view of clusters related to the EV industry by the SDES.
Modularity and silhouette measures are two important indicators of CiteSpace, defining the features of overarching structural networks. As a relatively high value (0.9452) of the modularity is shown in Fig. 2, it indicates that clusters are reasonably divided in the network. Thereafter, silhouette measures can examine the clustering effect of the processed data. The data are appropriately clustered when the value of silhouette measure approaches 1, otherwise, if it is close to −1, the data are not correctly clustered. If the silhouette equal to zero, the data should be located at the junction of boundaries of two natural clusters. The silhouette measure tells that the trend of the EV is high because cluster homogeneity is located at 0.6377.
74 clusters are found by auto-labeling, thereafter, three proper clusters are found through size and silhouette measures, that is, “Vehicle Exhaust Emissions”, “Climate Change” and “Integration”, each of those has a large size and silhouette (Cluster ID 6, 9, and 10; size 27, 25, and 25; silhouette 0.778, 0.658, and 0.628). Because the three keywords are distinctive, they are seen as representative elements of the EV industry. Besides, relevant keywords are allocated in Group I-III, as shown in Table 1 and Fig. 2, and the word frequency is the number before keywords.
Fig. 3 and Fig. 4 illustrate the relevant research in the EV industry in the view of timeline and time zone. The timeline view graph depicts the clusters in horizontal timelines from 1990 to 2020, and each cluster is displayed from left to right. On top of the view is a legend indicating the publishing time. The time zone graph has an array of vertical strips from left to right as time zones and reveals temporal patterns between a research front and its intellectual basis.
The layout method is a modified spring embedder technique in which an item’s horizontal movement is limited to its time zone, but its vertical mobility is decided entirely by its connections to objects in other time zones . Both timelines view and time zone view aims to highlight the particularity, but the difference is that the timeline covers the sustainability of clusters over years while the time zone views emphasize another feature of clusters, the temporal barycenter.
3. Development trends of the electric vehicle industry
To promote EV industry sustainable development and contribute to the literature in the field of the EV industry, the SDES involving the data collection, data analysis, and result visualization is constructed. Based on the SDES visualized results, the three EV development phases are determined: Vehicle Exhaust Emissions (2006), Climate Change (2010), and Integration (2015), which are the basis for further analysis.
3.1. Environmental concerns
Based on SDES results, vehicle exhaust emissions and climate change are main environmental concerns that attract multiple stakeholders’ attention at the early development stage.
3.1.1. Vehicle exhaust emissions
From Section , vehicle exhaust emissions are the earliest distinct cluster being identified. Motor traffic and coal-fired electricity are the main source of ambient air pollution, which has a serious environmental impact, increases the prevalence of respiratory symptoms, and even leads to premature death ; Chambliss et al., 2013;. With the surge of population, traffic, and commercial activity, vehicle exhaust emissions become an urgent problem in the development of industrialized society.
Ground-level ozone and particulate matter (PM) are the two most concerning air pollutant caused by vehicular exhaust emissions of the modern generation. Ground-level ozone is caused by the reaction of sunshine (UV rays) and emissions such as volatile organic compounds (VOCs), nitrous acid (HONO), hydrocarbons (HCs), and nitrogen oxides (NOx), and it helped toxic smog formation and damaged human health consistently [29,30]. Besides, from the assessment of the Environmental Protection Agency (EPA), 5 to 20% of the US population are vulnerable to the effect of ozone, mainly children, the elderly, pregnant, and others with medical history . As for PM2.5, the mixture of particles with aerodynamic diameters less than 2.5 μm and liquid droplets, can be easily inhaled by humans into the bloodstream and lungs, causing lung irritation and heart attacks. The resource of this pollution can be broken into primary and secondary pollutants, where the primary pollutant is emitted directly to the atmosphere by nature and humans while the secondary as precursors comes from the chemical reaction across different types of environmental pollutants, such as VOCs, SOx and NOx.
3.1.2. Climate change
Climate change was the second cluster indicating the elevated awareness towards environmental impacts. Since the early 19th century, climate change was referred to as human-induced impacts that drives global warming and extreme weather patterns, which threats human society with rising sea level, flooding, displacement, drought, wildfires, and food security, and it is defined as the greatest risk facing humankind in the 21st century by World Health Organization.
The greatest contributor to climate change is greenhouse gases (GHGs) because GHGs prevent the energy escape atmosphere and the temperature keeps rising with the escalated emissions. Within GHGs emissions composition, 80% and 10% are carbon dioxide (CO2) and methane (CH4), both are highly related to fossil fuels burning, such that in the production of coal, natural gas, and oil, and the use of gasoline and diesel-powered transports .
In 2019, 36.44 billion metric tons of CO2 equivalents are emitted globally, where power sectors account for 39%, industrial production for 28%, then 24% and 10% for transportation and residual consumption respectively . Because the burning of fossil fuel and other carbon-based uses are the largest single contributor to GHGs emissions, the shift towards low-carbon energy sources is vital to achieving the goals of stabilization and sustainability against global warming.