Servitization – Scopus – Bibliometrix

Main Information

Overall Scope and Growth:

• Timespan: 1988-2025: This is a substantial period covering 37 years. This suggests a long-term view of the research area. The inclusion of 2025 data likely means these are articles that have been accepted but not yet fully published/indexed at the time the data was extracted.
• Sources (Journals, Books, etc.): 932: The collection draws from a diverse range of sources, indicating a broad coverage of the research field. Analyzing *which* specific sources are most prominent could reveal key journals and publishers in the area.
• Documents: 2731: This represents a moderately sized collection. It is substantial enough to perform meaningful analysis, but not so large that it becomes unwieldy.
• Annual Growth Rate %: 14.98: This is a very high annual growth rate! This strongly suggests that the research area represented by this collection is rapidly expanding and attracting increasing attention. This could be due to emerging technologies, pressing societal needs, or breakthroughs driving further investigation. You should investigate *when* this growth is occurring. Is it consistent across the entire timespan, or has it accelerated in recent years?

Impact and Influence:

• Average citations per doc: 30.07: This is a fairly good average citation rate. It indicates that, on average, the documents in the collection have had a reasonable level of impact and influence within the scientific community. However, keep in mind that citation counts can vary widely based on the field, document type, and age of the publication. We would want to look at the *distribution* of citations to understand if this average is skewed by a few highly cited articles.
• References: 111288: A high number of references suggests that the research in this area is built upon a solid foundation of prior work and that there is strong cross-referencing among publications.

Content and Focus:

• Keywords Plus (ID): 6603; Author’s Keywords (DE): 5572: The difference between Keyword Plus and Author’s Keywords is important. Keyword Plus are keywords automatically generated by the indexing database (SCOPUS in this case), while Author’s Keywords are provided by the authors themselves. A significant difference between the number of these keywords may indicate a difference between the author’s perspective and the database’s classification of the research topics. Analyzing the overlap and differences in these keyword sets can provide insights into the evolving language and framing of the research area. What are the top keywords in each set? Are there emerging keywords not captured in earlier publications?

Authors and Collaboration:

• Authors: 4781: A large number of authors contributing to the collection demonstrates a wide community involved in this research area.
• Authors of single-authored docs: 201; Single-authored docs: 249: The small number of single-authored documents suggests that collaboration is a prevalent practice in this field.
• Co-Authors per Doc: 3.22: This reinforces the collaborative nature of the research. An average of over 3 authors per document highlights the importance of teamwork and interdisciplinary approaches.
• International co-authorships %: 29.99: Almost 30% international co-authorship indicates a significant level of global collaboration in this research area. This suggests that the research challenges being addressed are of international concern, or that expertise is distributed across different countries.

Document Types:

• article: 1555; book: 27; book chapter: 243; conference paper: 711; conference review: 39; editorial: 29; erratum: 4; note: 5; retracted: 4; review: 113; short survey: 1: The distribution of document types reveals the primary modes of knowledge dissemination in this field. Articles and conference papers are the dominant forms, indicating a strong emphasis on journal publications and presentations at academic conferences. The number of reviews (113) indicates that there are a good number of publications synthesizing and summarizing the state of the field. The few retracted articles (4) should be looked into to understand if there is a common cause or author.

Overall Interpretation and Discussion Points:

Based on these statistics, we can infer the following about the research area represented by this SCOPUS collection:

• A Growing and Dynamic Field: The high annual growth rate points to a vibrant and expanding area of research.
• Collaborative and Interdisciplinary: Collaboration is a key characteristic, with multiple authors per document and significant international co-authorship.
• Reasonable Impact: The average citations per document suggest a reasonable impact, but further analysis of the citation distribution is needed.
• Well-Established Foundation: The large number of references indicates a solid base of prior knowledge.
• Diverse Knowledge Dissemination: Articles and conference papers are the primary modes of communication, but books, book chapters, and reviews also play a role.

Recommendations for Further Analysis:

To gain a deeper understanding, I recommend exploring the following:

• Citation Analysis: Examine the distribution of citations to identify highly influential articles and assess the impact of specific authors or institutions.
• Keyword Analysis: Conduct a more detailed analysis of the keywords (both Author’s Keywords and Keywords Plus) to identify emerging trends and research themes. Use co-occurrence analysis to see which keywords appear together most frequently.
• Source Analysis: Identify the most prominent journals and conferences contributing to this collection. This will highlight the key venues for publishing research in this area.
• Author/Institution Analysis: Identify the most prolific and influential authors and institutions in this field. This can reveal centers of excellence and key players.
• Trend Analysis: Analyze the evolution of research topics and keywords over time to identify emerging trends and shifts in focus. How has the growth rate changed over the years?
• Network Analysis: Visualize collaboration networks among authors and institutions to understand the structure of the research community.

By delving deeper into these aspects, you can develop a more nuanced and comprehensive understanding of the research landscape represented by your bibliometric collection. Remember to connect these findings to your specific research question and use them to support your arguments and interpretations. Good luck!

Annual Scientific Production

Average Citations Per Year

Three-Field Plot
Overall Structure and Interpretation

This Three-Field Plot visualizes the relationships between:

• Left Field (CR): Cited References – the academic papers that the analyzed publications are referencing.
• Central Field (AU): Authors – the authors of the analyzed publications in your SCOPUS dataset.
• Right Field (KW_Merged): Merged Keywords – the keywords associated with the analyzed publications, likely combined from author-provided keywords and/or index keywords.

The connecting lines (also called edges) indicate the co-occurrence of these elements within the dataset. For example, a line connecting Author A to Keyword B means Author A has published papers that use Keyword B. Similarly, a line connecting Author A to Cited Reference C means Author A’s publications cite Reference C.

Key Observations and Potential Insights

Here’s what we can infer from the visual structure, keeping in mind we don’t have the *number* of publications associated with each element, just the connections:

1. Prominent Authors: Authors with many connections (lines radiating from their name) are influential and/or prolific within the scope of your analyzed data. From the image, “Parida V”, “Kohtamaki M”, “Pezzotta G” appear to be prominent in the analyzed collection. They connect to numerous cited references and keywords, indicating they’ve likely worked on various aspects within the general research area.

2. Key Research Themes: The keywords in the “KW_Merged” field represent the major themes and concepts covered in your analyzed publications. The most frequent keywords are likely at the top of the bar. From the image, “Servitization”, “Digital Servitization”, “Product-Service Systems” and “Manufacture” appear to be frequent.

3. Influential Cited References: The “CR” field shows the papers that have had a significant impact on the research area covered by your dataset. The more lines connecting to a specific cited reference, the more influential that reference is. “Wise R, Baumgartner P” and “Vandermerwe S, Rada J” appear to be influential works.

4. Author-Keyword Associations: The connections between the “AU” and “KW_Merged” fields are particularly important. They reveal which authors are working on which specific topics. For example, “Parida V” is associated with the keyword “Servitization”, suggesting this author’s research focuses on this area.

5. Author-Cited Reference Associations: The connections between “AU” and “CR” indicate the intellectual foundations of an author’s work. If an author cites a particular reference frequently, it suggests that the author’s research builds upon or directly relates to the ideas presented in that reference.

6. Keyword-Cited Reference Associations (Inferred): While not directly visualized, you can *infer* connections between keywords and cited references by tracing connections *through* the authors. For example, if Author A frequently cites Reference C *and* frequently uses Keyword B, you can infer that Reference C is relevant to research on Keyword B.

Specific Examples from the Image

• “Parida V” and “Servitization”: The strong connection between Parida V and the “Servitization” keyword suggests that Parida’s research is heavily focused on this topic.
• “Wise R, Baumgartner P” and “Go Downstream…”: This highly cited reference likely presents a fundamental concept or framework that is central to the research area.
• “Digital Servitization”: This keyword emerges, which suggests a trend towards the intersection of servitization and digitalization.

How to Use this Information for Research

• Identify Key Researchers: Focus on the most connected authors to find experts in the field.
• Understand Research Trends: Analyze the keywords to identify current research trends and emerging areas.
• Explore Intellectual Roots: Examine the cited references to understand the foundational works and key theoretical influences.
• Discover Research Gaps: Look for areas where connections are weak or missing. This might indicate under-explored topics or potential areas for future research.
• Contextualize Your Work: Use the plot to understand how your own research fits within the broader landscape of the field. Are you building upon established theories? Are you addressing emerging trends?

Critical Considerations and Next Steps

• Data Cleaning: “KW_Merged” is helpful, but ideally, you’d want to ensure the keywords are standardized (e.g., using controlled vocabulary) to avoid issues of synonyms or variations.
• Normalization/Weighting: The visualization doesn’t show the *strength* of the connections (e.g., number of publications linking an author and a keyword). More advanced visualizations might incorporate edge weights to represent the strength of the associations.
• Filtering and Zooming: Biblioshiny should allow you to filter the data (e.g., by year, document type) and zoom in on specific areas of the plot to explore relationships in more detail.
• Complementary Analyses: This Three-Field Plot is a good starting point. Combine it with other bibliometric analyses (e.g., co-citation analysis, keyword co-occurrence analysis) for a more comprehensive understanding.

In summary, this Three-Field Plot provides a valuable overview of the key authors, research themes, and influential works within your SCOPUS dataset. By carefully examining the connections and patterns in the plot, you can gain insights into the structure and dynamics of the research field and identify promising avenues for future investigation. Remember to explore the data in more depth using the filtering and analysis tools available in Biblioshiny.

Most Relevant Sources

Core Sources by Bradford’s Law

Sources’ Local Impact

Sources’ Production over Time

Most Relevant Authors

Authors’ Production over Time
Overall Trends and Observations:

The plot visualizes the publication history, productivity, and citation impact of leading authors in the field, likely “Digital Servitization” or a related domain given the article titles. Several authors show a surge in publications and citations in recent years, particularly between 2018-2022, indicating a growing interest and activity in this research area. The dominance of digital servitization as a research field is further confirmed by the large amount of co-authored papers that many authors published.

Individual Author Analysis:

• Parida V. and Kohtamäki M.: These authors demonstrate a similar trend, starting around 2016/2017 and experiencing a significant increase in both publications and citations in the years 2019-2021. They maintain a relatively high level of activity in recent years. The high TC/year of their co-authored papers, such as “DIGITAL SERVITIZATION BUSINESS MODELS IN ECOSYSTEMS: A THEORY OF THE FIRM” (TCpY 99.9) and “THE RELATIONSHIP BETWEEN DIGITALIZATION AND SERVITIZATION: THE ROLE OF SERVITIZATION IN CAPTURING THE FINANCIAL POTENTIAL OF DIGITALIZATION” (TCpY 88.5), highlights their influential contributions. Their research seems centered around business models in digital servitization.
• Baines T.: Baines has a longer publication history within this dataset. His publications show a steady stream of research. Baines’ early work, like “SERVITIZATION OF THE MANUFACTURING FIRM: EXPLORING THE OPERATIONS PRACTICES AND TECHNOLOGIES THAT DELIVER ADVANCED SERVICES” (2014) and “SERVITIZATION: REVISITING THE STATE-OF-THE-ART AND RESEARCH PRIORITIES” (2017), laid the groundwork for later research.
• Adrodegari F., Rapaccini M., and Saccani N.: These authors show similar publication trends. They are all focused on “DIGITAL SERVITIZATION IN MANUFACTURING” and “NAVIGATING DISRUPTIVE CRISES THROUGH SERVICE-LED GROWTH.”
• Kowalkowski C.: Similar to the previously mentioned authors, Kowalkowski’s publications gained popularity in the late 2010s.
• Pezzotta G.: Pezzotta’s publication history starts around 2018 and seems to be growing in productivity.
• Vendrell-Herrero F.: Vendrell-Herrero has an earlier start date and seems to be more consistent in their publication history.
• Zhang Y.: Zhang Y’s publication history starts later and seems to be consistent with their production.

Key Insights and Potential Discussion Points:

• Emerging Trends: The plot clearly indicates a surge in research interest in digital servitization around 2018-2022. This could be linked to advancements in digital technologies, increasing adoption of servitization models in industry, or specific funding initiatives.
• Collaboration and Impact: The co-authorship between authors like Parida and Kohtamäki suggests active collaboration within the research community. The high citation counts for their joint publications emphasize the benefits of collaborative research.
• COVID-19 Impact: The high citation counts in 2020 for articles addressing the impact of COVID-19 (e.g., Adrodegari, Rapaccini, Saccani, Kowalkowski “NAVIGATING DISRUPTIVE CRISES THROUGH SERVICE-LED GROWTH”) highlight the relevance of servitization strategies during times of crisis.
• Research Focus: Based on the titles of the most cited articles, the dominant themes within this field include digital servitization business models, the relationship between digitalization and servitization, the role of ecosystems, and the impact of digital technologies.
• Knowledge Transfer: A more mature researcher, such as Baines T., could have laid the foundations for later research done by the other researchers.

Further Analysis and Research Questions:

• Keyword Analysis: Perform a keyword analysis of the publications to identify the most frequently used terms and research topics. This can provide a deeper understanding of the evolution of the field.
• Co-citation Analysis: Analyze the co-citation patterns to identify the core literature and intellectual structure of the field.
• Journal Analysis: Identify the key journals where these authors are publishing. This helps understand the dissemination channels and the target audience.
• Geographic Distribution: Investigate the affiliations of the authors to identify the leading research institutions and countries in this field.
• Longitudinal Analysis: Examine the evolution of research topics over time to identify emerging trends and research gaps.

Limitations:

• The analysis is based on a single database (SCOPUS). Results might differ if other databases (e.g., Web of Science) were included.
• The TC/year metric can be influenced by the age of the publication. Newer publications may not have had sufficient time to accumulate citations.
• The analysis is limited to the top authors. A broader analysis of all authors in the field could provide a more comprehensive picture.

By considering these points and conducting further analysis, you can develop a deeper and more nuanced understanding of the research landscape in the field of digital servitization.

Author Productivity through Lotka’s Law

Authors’ Local Impact

Most Relevant Affiliations

Affiliations’ Production over Time

Corresponding Author’s Countries

Overall Observations

The plot provides a clear visualization of the publication output and collaboration patterns of corresponding authors from different countries within your specific research area (as defined by your SCOPUS search). It allows us to:

• Identify the leading countries in terms of research output.
• Assess the extent to which these countries engage in international collaboration.
• Compare the balance between domestic (SCP) and international research engagement (MCP).

Key Findings and Interpretation

1. Leading Countries in Research Output:

* China is by far the most productive country in this dataset, with 405 articles. However, a significant portion of its publications (326) are Single Country Publications (SCPs), indicating a strong domestic research focus.
* Italy and the United Kingdom are the next most productive countries, with 207 and 203 articles, respectively. They also have substantial numbers of SCPs, but their MCP numbers are significantly higher than China’s in absolute terms.
* Germany, Sweden, Finland, and Spain follows, each contributing a substantial amount of articles.

2. International Collaboration (MCP Ratio):

* Norway exhibits the highest MCP ratio (66.7%), indicating a strong inclination towards international collaborative research. While its total publication count is relatively low (21), the majority of its research involves international partners.
* USA (57.8%) and Serbia (57.9%) also demonstrate a high MCP ratio, suggesting a strong reliance on international collaborations. However, keep in mind the relatively small sample size for both countries, especially Serbia (19 articles).
* Australia (52.2%) and Sweden (49.7%) shows a strong pattern towards international collaboration.
* In contrast, Korea (12.5%) has the lowest MCP ratio, indicating a strong preference for domestic research. China and Japan also exhibit low MCP ratios, suggesting a focus on internal research efforts.

3. Balance Between Domestic and Global Research Engagement:

* Countries like China, Italy, and Germany, although producing a high volume of research, tend to lean towards domestic research, as evidenced by the large proportion of SCPs. This might reflect strong national funding initiatives, well-established domestic research institutions, or specific national research priorities.
* Countries like Norway, Sweden, and Finland demonstrate a more balanced approach, with a higher proportion of MCPs, indicating a greater emphasis on international partnerships. This could be due to factors such as smaller domestic research communities, access to specialized expertise in other countries, or participation in large international research projects.

Potential Discussion Points and Further Investigation

• Database Bias: The analysis is based on SCOPUS data. It is important to acknowledge potential biases in database coverage. Different databases may index different journals and publications, which could affect the representation of research output from different countries. Consider comparing the results with other databases like Web of Science.
• Field-Specific Differences: The observed collaboration patterns could be specific to the research field being analyzed. Some fields are inherently more collaborative than others. Consider how the nature of your research area might influence international collaboration.
• Funding and Policy: National research funding policies and international collaboration programs play a significant role in shaping research collaborations. Investigate the funding landscape in the most productive countries and how these policies might encourage or discourage international partnerships.
• Impact of Collaboration: It would be interesting to explore the impact of international collaborations. Do MCP articles from specific countries have higher citation rates or other indicators of research impact compared to SCP articles?
• Network Analysis: This analysis could be expanded to include a network analysis of country-level collaboration, revealing the specific partnerships between countries. This would provide a more detailed picture of the global research landscape in your field.
• Reasons for Low MCP in China: Considering China’s huge participation it would be interesting to investigate why the international collaborative publications are lower than other countries such as USA, or European countries. Is that a consequence of fundings, research interest or international interest?

In summary, this Corresponding Author’s Country Collaboration Plot offers valuable insights into the global research landscape in your specific field. By examining the publication output and collaboration patterns of different countries, you can identify key players, understand the dynamics of international research, and generate hypotheses for further investigation. Remember to consider the limitations of the data and the potential influence of field-specific factors and research policies when interpreting these results.

Countries’ Scientific Production

Countries’ Production over Time

Most Cited Countries

Most Global Cited Documents

Most Local Cited Documents

Vandermerwe (1988); Neely (2008)

Overall Observations:

• Focus on Industrial Marketing Management: The journal *Industrial Marketing Management* (IMM) appears very frequently in the top 20 locally cited articles. This suggests a strong concentration of research within your dataset focused on topics covered by IMM. It’s a key outlet for research in this area. In fact, by excluding duplicates of the same article, 6 out of the most 20 local cited articles have been published in this journal.
• Prevalence of Recent Publications: A significant portion of the top articles were published in the late 2010s (especially 2017, 2018 and 2019). This could indicate a relatively new or rapidly evolving area of research. It’s worth investigating what specific themes or topics gained prominence during this period.
• Varied Citation Profiles: Articles exhibit different relationships between local and global citations, suggesting varying degrees of niche versus broad appeal. Some have high local citation counts but relatively modest global counts, while others demonstrate strong influence both within the specific field and more broadly.

Analysis of Key Articles (highlighting different scenarios):

• *VANDERMERWE S, 1988, EUR MANAGE J: LC 930, GC 2019, NLC 1, NGC 1*: This paper stands out with by far the highest local citations (LC=930) and a strong global citation count (GC=2019). However, the normalized citations are equal to 1, which might indicate that it is an old article or that the impact of this article in the research field is no longer what it used to be, making this article less relevant for the current research being undertaken.
• *NEELY A, 2008, OPER MANAGE RES: LC 480, GC 1070, NLC 6.02, NGC 5.8*: A paper with high LC and GC, with the normalized citations that are greater than 1, indicating a consistent significance in its research area, but it might be less impactful due to its lower NLC and NGC values.
• *BAINES TS, 2009, J MANUF TECHNOL MANAGE: LC 463, GC 1216, NLC 7.19, NGC 7.28*: A paper with high LC and GC, with the normalized citations that are greater than 1, indicating a consistent significance in its research area, but it might be less impactful due to its lower NLC and NGC values.
• *MARTINEZ V, 2010, J MANUF TECHNOL MANAGE: LC 238, GC 453, NLC 13.88, NGC 12.32*: This is an interesting case. While the absolute GC and LC are not the highest, the normalized local citation (NLC = 13.88) is among the highest in the list, and the normalized global citation is also high (NGC = 12.32), indicating that within the context of publications from 2010, this article has a significant impact both locally and globally.
• *RADDATS C, 2019, IND MARK MANAGE: LC 235, GC 373, NLC 23.43, NGC 8.37*: This article is particularly noteworthy due to its exceptionally high normalized local citation count (NLC = 23.43), suggesting it’s a highly influential paper within your specific research area, especially considering its relatively recent publication date. Despite its strong local impact, its normalized global citation count (NGC = 8.37) is not as high. This could indicate that the article’s primary influence is within a specialized field of study rather than across broader academic disciplines.
• *PASCHOU T, 2020, IND MARK MANAGE: LC 232, GC 430, NLC 35.41, NGC 10.34*: This article, published in 2020, demonstrates the highest normalized local citation count (NLC = 35.41) among the listed publications. This exceptionally high NLC value suggests that it has had a very significant impact within the specific research area covered by your dataset, especially considering its recent publication date. Its normalized global citation count (NGC = 10.34) is also notable, indicating a broader influence, although not as pronounced as its local impact. The high NLC suggests that this article addresses topics or methodologies that are particularly relevant and timely to the research community focusing on the themes within your dataset.

Interpretation and Discussion Points:

1. Identify Key Research Themes: What are the dominant themes or topics addressed in these highly cited articles, especially those from *Industrial Marketing Management*? This will help define the core focus of your research area. Are there specific methodologies or theoretical frameworks that are frequently cited?
2. Trace the Evolution of the Field: How have the research priorities evolved over time? Compare the themes and approaches of the older articles (e.g., Vandermerwe, 1988) with the more recent ones. Are there shifts in focus, new methodologies, or emerging areas of interest?
3. Assess the Breadth vs. Depth of Impact: Compare articles with high local *and* global citations to those with high local but lower global citations. The former likely represent foundational or widely applicable research, while the latter may indicate specialized topics or niche areas.
4. Consider the Influence of *Industrial Marketing Management*: The strong presence of IMM articles suggests that this journal is a central hub for research within your area. Analyze the types of articles published in IMM that are highly cited within your dataset. This could reveal specific sub-areas or research paradigms that are particularly influential.
5. Investigate Recent Trends: The prominence of articles from the late 2010s indicates a period of significant activity. Explore the specific topics and research questions that gained traction during this time. Are there any emerging trends or debates within the field that these articles reflect? What could have caused this surge in publications?
6. Examine the Content of High-NLC Articles: Focus on articles with particularly high NLC values (e.g., Raddats, 2019; Paschou, 2020). These are the papers that have had the most significant impact *within your specific research community*. Understanding their content is crucial. What makes them so relevant to your dataset? Are they introducing new concepts, methodologies, or data that are particularly valuable to researchers in this area? This could also point to potential research gaps.
7. Limitations of the Data: Remember that this analysis is based on Scopus data alone. The citation counts might differ in other databases (e.g., Web of Science).

Next Steps:

• Read the Most Relevant Articles: Prioritize reading the articles with the highest local citation counts and/or normalized citation counts, especially those from recent years.
• Perform Co-citation Analysis: Explore which articles are frequently cited together. This can reveal intellectual clusters or schools of thought within your research area.
• Analyze the Keywords: Examine the keywords associated with these articles to identify the key concepts and themes.

By carefully considering these points, you can gain a deeper understanding of the intellectual landscape of your research area and identify promising avenues for future research. Remember to relate these findings back to your specific research questions and objectives. Good luck!

Most Local Cited References

Reference Spectroscopy
Overall Interpretation:

The RPYS plot visualizes the historical roots of the research area. The black line represents the overall citation activity, indicating how many references from a given year are being cited within the dataset. The red line identifies years with significantly higher-than-expected citation impact based on a 5-year moving median, flagging particularly influential publications. The key is to look at the peaks in the red line as these represent years that contained seminal works that continue to be cited heavily in the field.

Specific Observations & Implications:

1. Late Emergence: The plot shows very little citation activity before the 1970s. This suggests that the area of study is relatively new, or that the specific perspective adopted in this dataset emerged more recently.
2. Early Foundational Work (1972): The first peak in the red line is 1972. The list of most-cited publications for this year includes works by Levitt, Child, and Webster. This points to foundational research in general management and marketing concepts, specifically the application of production line thinking to services, organizational structure and environment, and organizational buying behaviour. These were important building blocks upon which subsequent research would be built.
3. Mid-1970s Psychology and Organizational Theory (1978): The 1978 peak contains works such as Nunnally’s *Psychometric Theory*, and Pfeffer and Salancik’s *The External Control of Organizations*. This suggests that methodological rigor (psychometrics) and strategic perspectives like resource dependence theory were integrated into the field during this time. Mintzberg’s work on strategy formation also highlights the importance of strategic thinking.
4. Strategic Management and Institutional Economics (1985): The 1985 peak includes Porter’s *Competitive Advantage* and Williamson’s *The Economic Institutions of Capitalism*, along with Lincoln & Guba’s *Naturalistic Inquiry*. This signals a shift towards applying strategic management frameworks and institutional economics to the understanding of organizational phenomena, as well as a recognition of qualitative research methods in the field.
5. Servitization Takes Center Stage (1988 onward): The 1988 peak is dominated by Vandermerwe and Rada’s work on “Servitization of Business.” This is a critical turning point, indicating the formal introduction of the servitization concept into the academic literature.
6. Continued Development (1999-2017): The peaks in 1999, 2003, 2008, 2013, 2015 and 2017 show a sustained interest in servitization. Papers in these peak years such as Wise and Baumgartner (1999) emphasize the profit imperative in manufacturing, Oliva and Kallenberg (2003) examine the transition from products to services, Neely (2008) and Vargo & Lusch (2008) further develop the topic, and Baines (2013), Cusumano et al. (2015), Coreynen et al. (2017) Kowalkowski et al. (2017) and Vendrell-Herrero et al. (2017) provide ongoing theoretical and practical insights into servitization. The concentration of peaks towards the latter end of the time-frame shows this is still an evolving field of research. The 2017 references focus on the interplay between servitization and digitization, pointing to a more modern application of the servitization model in the business world.

Key Insights and Research Questions:

• Evolution of the Field: The RPYS shows how the field evolved from general management and organizational theory to a specific focus on servitization and product-service systems.
• Theoretical Influences: The plot reveals the influence of strategic management, resource dependence theory, institutional economics, and service-dominant logic on the field.
• Methodological Development: The inclusion of psychometrics and naturalistic inquiry suggests a diversification of research methods employed in the field.
• Future Directions: The recent focus on digitization and servitization points to potential research opportunities in areas such as digital servitization, data-driven services, and the impact of Industry 4.0 on service models.

Further Exploration:

• Citation Network Analysis: Complement this RPYS analysis with a citation network analysis to visualize the relationships between key publications and identify influential authors and research clusters.
• Content Analysis: Perform a content analysis of the highly cited publications to identify key themes, concepts, and research questions.
• Database Coverage: Be aware that the analysis is based on SCOPUS data. Consider the biases inherent in this database (e.g., journal coverage, language).

By combining the information from the RPYS plot with your domain knowledge, you can gain a deeper understanding of the historical development and intellectual foundations of the research area. Remember to critically evaluate the limitations of the data and consider alternative interpretations. Good luck!

Most Frequent Words

Words’ Frequency over Time

Trend Topics

Overall Interpretation

This plot visualizes the evolution of research trends over time, using keyword frequency as a proxy for research interest. Several key observations can be made:

• Emerging Trends: The terms appearing at the top and towards the right of the plot represent relatively new and rapidly growing research areas. These indicate current “hot topics”.
• Established Trends: Terms that appear earlier in the timeline and have longer “tails” (the interquartile range lines) represent more established research areas. These have been consistently researched over a longer period.
• Frequency and Prominence: The bubble size provides an indication of the term’s prominence within the collection in a particular year. Larger bubbles signify higher frequency and, therefore, greater attention.
• Database Context (Scopus): The fact that this analysis is based on Scopus data provides some clues on the type of trends being captured. Scopus has a wide subject coverage, but it has a certain orientation toward STEM fields and Social Sciences.

Detailed Analysis

Let’s dissect some specific trends based on the visualization:

1. Recent and Emerging Trends (around 2023-2025):

* Green Development, Fintech, Decentralized Finance, Economic Development, Human, Digital Service Innovation, Digital Servitization, Circular Economy, China: The significant presence of these terms indicates a surge of research interest in the intersection of technology, sustainability, and digital transformation. The inclusion of “China” might also be indicative of a specific regional focus within the dataset. The presence of Fintech, Decentralized Finance, Digital Service Innovation and Digital Servitization may represent the digital transition of the economics landscape.

2. Trends with Sustained Interest (2019-2021):

* Manufacturing, Business Models, Digitalization, Servitization, Innovation: These terms show strong activity around 2019-2021. This suggests a sustained research interest in these areas. The large bubble size for Digitalization, Servitization and Innovation around 2021 is very important and suggest high attention towards them.

3. Established but Continuing Trends (2015-2017):

* Life cycle, Competitive Advantage, Product-Service System (PSS), Business Modelling, Industrial Engineering: The fact that these keywords appear earlier suggests they are more established areas. However, the length of their tails indicates that they continue to be relevant in the research landscape, albeit perhaps not with the same explosive growth as the newer trends. The focus on “Product-Service System (PSS)” and “Business Modelling” reveals interest in integrated solutions and strategic planning within industrial contexts.

4. Older Trends (2011-2013):

* Products, Product offerings, Management Science, Research, Research studies, Information Science, Information Technology, Competitive Strategy, Outsourcing, Industry, Service Supply Chains, Integrated Products, Interoperability, Empirical Studies, Integrated Solutions, Design: These terms are the earliest in the plot, appearing around 2011-2013. It’s important to note that the limited number of these terms that made it to the top k of the year may just be a reflection of the available data from Scopus, suggesting that the other trends were much more prevalent in the dataset.

Further Considerations & Critical Discussion Points:

• Keyword Selection: The analysis is based on “KW_Merged.” It’s important to understand *how* these keywords were merged. Were they author-supplied keywords, keywords extracted from titles/abstracts, or a combination? This affects the interpretation. If author-supplied, there may be biases in keyword usage.
• Database Bias: As mentioned earlier, Scopus has a certain subject coverage. If the analysis aims to represent *all* research trends, consider complementing it with data from other databases (e.g., Web of Science, specialized databases).
• Data Cleaning: Was any data cleaning performed (e.g., stemming, lemmatization, stop word removal)? This impacts the precision of the trend identification. For example, were “servitization” and “servitisation” deliberately kept separate (perhaps to reflect different regional spellings or nuances)?
• Normalization: It might be useful to normalize keyword frequencies by the total number of publications in a given year to account for the overall growth of the literature. This would provide a more accurate picture of *relative* trend prominence.
• Qualitative Analysis: This bibliometric analysis is a good starting point, but it should be complemented with qualitative analysis. Reading a sample of papers associated with the identified trends can provide a deeper understanding of the research questions, methodologies, and findings.
• Specificity of Trends: Some terms are very broad (e.g., “research,” “industry”). Consider more granular analyses to uncover specific sub-trends within these broader categories.

In Summary:

This trend topics plot provides a valuable overview of the evolving research landscape. It highlights the increasing prominence of topics related to digital transformation, sustainability, and specific regional contexts. By considering the nuances of the data source, keyword selection, and analysis methods, you can draw more informed conclusions and identify promising directions for future research.

Clustering by Coupling

Co-occurrence Network
Overall Structure and Interpretation

The network clearly displays two distinct communities (indicated by different colors): a red cluster and a blue cluster. This suggests two main research areas being explored in the documents in your collection. The size of the nodes indicates the frequency of the keyword within the dataset, and the thickness of the edges reflects the strength of the co-occurrence between keywords. The closer the keywords are in the network, the stronger their relationship.

Community Analysis

• Red Cluster: This cluster is heavily focused on “Servitization”. Other keywords within this community are:

* “Manufacturing servitization”
* “Business Model”
* “Business Model Innovation”
* “Digitalization”
* “Digital Transformation”
* “Industry 4.0”
* “Digital Servitization”
* “Innovation”
* “Smart Manufacturing”
* “Case Study”
* “China”
* “Internet of Things”
* “Ecosystems”
* “Supply Chain Management”
* “Circular Economy”

Interpretation of the Red Cluster: This community represents research into the *servitization* of manufacturing processes, exploring how manufacturing is being transformed through digital technologies and new business models. The prevalence of “Industry 4.0,” “Digitalization,” “Internet of Things,” and “Smart Manufacturing” strongly supports the notion of digital technologies as key drivers of servitization. The inclusion of “Business Model Innovation” and “Ecosystems” indicate research into new business paradigms emerging from servitization. The presence of “China” suggests a geographic focus, potentially reflecting the significant role China plays in manufacturing and its adoption of servitization strategies.

• Blue Cluster: This community revolves around “Manufacture” and “Product-Service Systems”. Related keywords include:

* “Product Service System”
* “Service Industry”
* “Manufacturing Industries”
* “Manufacturing Companies”
* “Product Design”
* “Supply Chains”
* “Costs”
* “Commerce”
* “Industrial Research”
* “Competition”
* “Life Cycle”
* “Knowledge Management”
* “Industrial Management”
* “PSS”
* “Competitive Advantage”
* “Product-Service Systems (PSS)”
* “Information Management”
* “Profitability”
* “Sales”
* “Decision Making”
* “Business Modelling”

Interpretation of the Blue Cluster: This community represents research into the *traditional aspects of manufacturing*, including design, production, supply chain, and product-service system (PSS). “Product-Service Systems” is strongly connected to manufacture. It also explores management principles within manufacturing, such as industrial management and knowledge management. The presence of “costs,” “profitability,” and “competitive advantage” indicates research focused on the economic performance and strategic positioning of manufacturing companies.

Key Observations and Potential Insights

• “Servitization” and “Manufacture” as Central Themes: These keywords are the most connected, indicating their importance in your dataset. The connection of the two main cluster via edges implies there are cross-cutting studies addressing both the traditional manufacturing aspects as well as the transformation of manufacturing via servitization.
• Walktrap Clustering: The *walktrap* community detection algorithm is a good one to use in this case since it looks at random walks within the network, therefore is a better approach to detect the underlying structure of the keywords within the topic. This can be useful when exploring how different keywords are connected to each other in your study.
• Association Normalization: Since the normalization was set to association, the results are focused on the number of co-citations between articles with different citation behaviors.

Next Steps and Further Analysis

1. Refine Keyword Search: Based on these initial findings, consider refining your search queries to focus on specific sub-themes within these clusters. For example, you could investigate the intersection of “Digital Servitization” and “Supply Chain Management.”

2. Content Analysis: Perform a deeper content analysis of the papers within each cluster. Read abstracts and, if feasible, full texts, to understand the specific research questions, methodologies, and findings.

3. Temporal Analysis: Conduct a temporal analysis to examine how these themes have evolved over time. Are there emerging trends or shifts in research focus?

4. Author and Institutional Analysis: Explore the authors and institutions contributing to each cluster. This can identify leading experts and research centers in these areas.

5. Expand the Analysis: Consider using other bibliometric techniques, such as citation analysis or co-citation analysis, to complement the word co-occurrence network.

Critical Considerations

• Keyword Limitations: Remember that keyword analysis is limited by the author’s choice of keywords. It may not capture the full nuances of the research.
• Database Bias: Scopus is a comprehensive database, but it may have biases towards certain journals or disciplines.
• Parameter Selection: The parameters used to generate the network (e.g., `edges.min`, `cluster`, `community.repulsion`) can influence the results. Experiment with different settings to see how they affect the network structure.

By critically evaluating the network structure, community composition, and key terms, you can gain valuable insights into the intellectual landscape of your research area. Remember to corroborate these findings with other bibliometric analyses and, most importantly, with a thorough reading of the relevant literature. Let me know if you’d like me to help you with refining the keyword search, creating another graph, or anything else.

Thematic Map

Overall Structure and Interpretation

The strategic map is a two-dimensional representation that positions research themes based on two key metrics:

• Centrality (Relevance Degree): Indicates the importance of a theme within the network of research. Higher centrality suggests the theme is strongly connected to other themes and acts as a central hub.
• Density (Development Degree): Indicates the internal development of a theme. High density suggests a well-developed and specialized research area with strong connections between its constituent elements.

The map is divided into four quadrants, each representing a different strategic role for the themes:

• Quadrant I (Upper Right): Motor Themes High centrality and high density. These are well-developed and important themes driving the research field. *Currently, this quadrant is empty in your map.*
• Quadrant II (Upper Left): Niche Themes High density but low centrality. These are specialized themes, well-developed internally but with limited connections to the broader research field. Represented by ‘internet of things’ cluster.
• Quadrant III (Lower Left): Emerging or Declining Themes Low density and low centrality. These are either new themes with limited development or themes that are losing relevance. *This quadrant is empty in your map.*
• Quadrant IV (Lower Right): Basic Themes High centrality but low density. These are fundamental and cross-cutting themes, important for the field but not highly developed in specific areas. Represented by ‘servitization’ cluster.

Cluster Analysis

The analysis has identified three clusters: “internet of things”, “manufacturing”, and “servitization”. Let’s examine each:

1. “internet of things” (Niche Theme):

* Located in the upper-left quadrant, this cluster has high density but low centrality.
* Interpretation: This suggests that “internet of things” is a well-defined and specialized area of research but not strongly connected to other major themes in the broader field as defined by the keyword co-occurrence network. The research is focused and self-contained, but perhaps not as influential in driving overall trends in the larger field represented by the dataset.
* Key Articles:
* HEINIS TB, 2018, RES TECHNOL MANAGE (PageRank: 0.204): Likely a core article within the IoT research area, focusing on technology management.
* RYMASZEWSKA A, 2017, INT J PROD ECON (PageRank: 0.186): Suggests a focus on production economics within the IoT context.
* BRITO G, 2017, PROC – IEEE INT CONF IND INF, INDIN (PageRank: 0.18): Points to research on industrial informatics and applications of IoT within industry.
* Discussion Points: Is the lack of broader connections a limitation? Could the IoT research benefit from integrating more with other areas like servitization or sustainable development? Is the cluster too specific, hindering cross-disciplinary impact?

2. “manufacturing” (Central):

* Located near the center of the map, but slightly above the x axis, this cluster has relatively central and medium density.
* Interpretation: This suggests that “manufacturing” is reasonably connected to other major themes in the broader field, it is not driving the field forward like ‘motor themes’ would but is contributing in relevant and important ways.
* Key Articles:
* BATLLES‐DELAFUENTE A, 2021, INT J ENVIRON RES PUBLIC HEALTH (PageRank: 0.202): Likely a relevant article that covers public health or environmental research within manufacturing
* XING Y, 2023, TECHNOVATION-a (PageRank: 0.187): Suggests this area is related to technological innovation within the manufacturing context.
* GUO A, 2015, TECHNOL SOC (PageRank: 0.185): Points to research on the relations between technology and society.
* Discussion Points: How is sustainability being integrated into manufacturing research? What specific technologies are driving these innovations? How are these innovations affecting society?

3. “servitization” (Basic Theme):

* Located in the lower-right quadrant, this cluster has high centrality but low density.
* Interpretation: This indicates that “servitization” is a fundamental theme strongly connected to other areas but is perhaps not as internally specialized or developed as the “internet of things” cluster. It’s a core concept that influences many areas but might lack specific, deeply explored sub-topics.
* Key Articles:
* WANG LP, 2010, PROC – IEEE INT CONF EMERG MANAGE MANAGE SCI, ICEMMS (PageRank: 0.24): Potentially an influential early paper on servitization in emerging management and management science.
* BAECKER J, 2021, ANNU AMERICAS CONF INF SYST, AMCIS (PageRank: 0.237): Suggests ongoing interest and research in servitization within the information systems field.
* TESO G, 2016, PROCEDIA CIRP (PageRank: 0.229): Likely focuses on servitization within the context of manufacturing or production engineering (CIRP).
* Discussion Points: How is servitization impacting different industries? What are the key challenges in implementing servitization strategies? Is there a need for more specialized research within specific aspects of servitization? How is servitization linked to product-service systems and what are the implications of these links?

Parameters and Data Source Considerations:

• SCOPUS: The analysis is based on SCOPUS data. Keep in mind that SCOPUS has a particular coverage profile. Results might differ slightly if using Web of Science or other databases.
• Keywords (KW\_Merged): The analysis uses merged keywords. This is useful for capturing related concepts, but it can also blur the lines between distinct themes.
• Parameters (n=250, minfreq=13, ngrams=1, stemming=FALSE, size=0.3, n.labels=3, community.repulsion=0, repel=FALSE, cluster=walktrap): These parameters influence the network construction and clustering. The `minfreq` parameter (minimum keyword frequency of 13) will exclude less common keywords, potentially overlooking niche areas or emerging trends. The `walktrap` clustering algorithm is a community detection method; different algorithms could yield slightly different cluster compositions.

Overall Conclusion and Further Research Directions:

The strategic map provides a snapshot of the research landscape based on keyword co-occurrence. It highlights the importance of “servitization” as a core concept, the specialized nature of “internet of things” research, and the more balanced position of “manufacturing” and its ties to sustainability.

Further research could explore:

• The relationships *between* these clusters. For instance, how is the “internet of things” being used to enable “servitization” in “manufacturing”?
• The evolution of these themes over time. Has the centrality or density of any of these clusters changed significantly in recent years?
• A more granular analysis of the “internet of things” cluster to identify specific sub-topics and their connections to other fields.
• Investigating the themes absent from the map. What topics, based on your knowledge of the field, *should* be present but are not captured by this analysis? Why might this be?
• The “manufacturing” cluster’s specific innovations affecting society.

By critically examining the map and the underlying data, you can gain valuable insights into the structure and dynamics of the research field and identify promising avenues for future investigation. Remember that this is *one* view of the research landscape, shaped by the data source and analysis parameters.

Factorial Analysis
Overall Structure and Dimensional Interpretation:

• Dimensions: The plot shows two dimensions, Dim 1 explaining 38.74% of the variance and Dim 2 explaining 17.23%. Dim 1 appears to capture a broad distinction between traditional manufacturing and more modern, digitally-oriented concepts. Dim 2 seems to separate ideas of “competitive advantage” and process oriented themes.

* Axes Interpretation:
* Dim 1 (Horizontal): Moving from left to right, we likely transition from traditional manufacturing-focused topics toward digital transformation and service-oriented perspectives.
* Dim 2 (Vertical): Moving from bottom to top, we transition from themes concerning sustainable development, product-service systems, and literature reviews towards those emphasizing competitiveness, manufacturing processes, and information management.

Cluster Identification and Interpretation:

The graph shows several potential clusters and themes:

1. Traditional Manufacturing Cluster (Top Left): Terms like “competitive advantage”, “manufacturing industries”, “competition”, “industrial management”, “manufacturing companies,” and “manufacture” are grouped together. This suggests a research stream focusing on traditional aspects of manufacturing, competitiveness, and management within established industrial contexts.

2. Digital Transformation & Servitization Cluster (Top Right): “Digital transformation”, “service innovation”, “digital servitization”, “digitalization”, “industry 4.0”, “servitization”, and “innovation” cluster. This clearly indicates a research area concerned with the integration of digital technologies into service offerings and manufacturing processes. “China” also appears in this cluster, hinting at research concerning digital transformation or innovation in the Chinese context.

3. Product-Service System and Sustainability Cluster (Bottom Left): “Product-service system (pss)”, “product design”, “sustainable development”, “product-service system”, and “literature reviews” form a cluster, indicating research focused on the design, analysis, and sustainability aspects of PSS. The presence of “literature reviews” suggests a focus on summarizing and synthesizing existing knowledge in this area.

4. Supply Chain and Business Models Cluster (Center): This central area contains terms like “information management”, “supply chains”, “decision making”, “business models”, “supply chain management”, and “ecosystems.” This suggests a research focus on the strategic and operational aspects of supply chains, business model innovation, and the role of information in these processes.

Key Terms and Their Relevance:

• “Competitive advantage”: Its location far in the upper left quadrant highlights its relative separation from newer concepts of digital transformation or sustainable product-service systems. This might indicate a focus on traditional competitive strategies.
• “Digital Transformation” & “Industry 4.0”: These terms prominently placed in the top right emphasize the strong trend of digitization in manufacturing and service industries. Their proximity suggests that “Industry 4.0” is being studied as an enabler of “Digital Transformation”.
• “Product-Service System (PSS)” and “Sustainable Development”: The positioning in the bottom quadrant suggests a strong connection between these concepts, indicating research into environmentally conscious and service-oriented approaches to product design and delivery.
• “Supply Chain Management”: Located centrally, it bridges the gap between traditional manufacturing and the newer digital/service paradigms, signifying its crucial role in both contexts.

Interpretation and Discussion Points for Researchers:

1. Evolution of Manufacturing Research: The map clearly illustrates the shift from traditional manufacturing research (focused on competition and industrial management) towards more contemporary themes like digital transformation, servitization, and sustainability.

2. Strategic Implications: The separation of “competitive advantage” from the digital transformation cluster might suggest that researchers are exploring how companies can achieve competitive advantage in the digital age. Are traditional models of competitive advantage still relevant, or are new approaches needed?

3. Sustainability Concerns: The prominence of the “Product-Service System” and “Sustainable Development” cluster suggests a growing awareness of environmental issues in the manufacturing and service sectors. This could reflect research into circular economy models, resource efficiency, and the social impact of industrial activities.

4. Role of Information and Supply Chains: The central location of the supply chain and business model cluster highlights its importance as a connector between different research streams. This could prompt investigations into how digital technologies are impacting supply chain operations, or how business models need to adapt to accommodate sustainable practices.

5. Geographical Considerations: The presence of “China” as a keyword suggests that research might be focusing on the unique challenges and opportunities of digital transformation and innovation in the Chinese context.

Suggestions for Further Analysis:

• Explore Different Time Slices: Divide the SCOPUS data into different time periods and generate separate MCA maps to visualize how research trends have evolved over time.
• Keyword Network Analysis: Complement the MCA with network analysis to identify key authors, institutions, and publications that are driving research in each of the identified clusters.
• Qualitative Review: Conduct a qualitative review of selected papers from each cluster to gain a deeper understanding of the research questions, methodologies, and findings.

This interpretation should provide a solid foundation for discussing the results of your bibliometric analysis and formulating further research questions. Remember to always critically evaluate the results and consider the limitations of the data and methods used. Good luck!

Co-citation Network
Overall Structure:

The network shows a structure of interconnected nodes (cited references) clustered into distinct communities. There appear to be three primary clusters, distinguished by color (red, blue, and green), indicating groups of papers that are frequently co-cited together. The presence of these communities suggests the existence of distinct subfields, schools of thought, or methodological approaches within the broader research area.

• Central Node: The node labeled “vandermerwe liva.r.198831” (interpreted as Vandermerwe and Liva, 1988) is the most central and highly connected node, given its size. This suggests it’s a foundational work, acting as a bridge between different research streams.
• Peripheral Nodes: Several nodes appear on the periphery (e.g., “vandermerwe s. 1988-2”), suggesting that these publications may be more specialized or tangential to the main research themes. These might represent emerging areas, niche applications, or studies with a narrower scope.

Communities:

The “walktrap” clustering algorithm was used, which is designed to identify communities based on random walks on the network. Here’s a likely interpretation of the communities:

• Red Cluster: Given the presence of nodes like “Ulaga W. 2011”, “Vargo S.L. 2008”, “Gebauer H. 2011”, “Baines 1. 2013”, “Tukker A. 2004” and “Martinez 2. 2010”, this cluster likely represents research focused on Service-Dominant Logic (SDL), servitization, or related business models. Vargo and Lusch’s work is foundational to SDL. Gebauer and Tukker are prominent in the field of servitization. The centrality of “Vandermerwe Liva 1988” could point to the early conceptualizations of value creation in service contexts, which are precursors to SDL. The presence of Baines suggests the cluster focuses on advanced service offerings and industrial service.
• Blue Cluster: Nodes such as “Kohtamaki m. Cortynen w. 2017”, “Kohtamaki M. 2020”, “Sklyar A. 2019”, “Paiola M. 2020”, and “Tronvoll 2020” are likely centered on Business-to-Business (B2B) marketing, Innovation Management, or sales. Given the newer dates for the publications present in this cluster, it may represent a more recent research thread, potentially building upon the theoretical foundations represented by the red cluster.
• Green Cluster: With nodes such as “Eisenhardt K.M. 1989”, “Eisenhardt K.M. 2007”, “Kowalkowski C. 2017-1”, “Kowalkowski C. 2017-2” and “Story V.M. 2017” this cluster appears to be related to Case Study Research, and Key Account Management.

Most Connected Terms (Central Nodes):

• Vandermerwe and Liva (1988): As mentioned earlier, its high connectivity indicates its foundational role. This highlights the enduring relevance of its core ideas within the analyzed research area. Further research into the context of the co-citation would be helpful.
• Other central nodes in each cluster: Identifying the most central nodes within each cluster (beyond the overall most central node) provides insights into the key publications driving each specific subfield. Understanding the core themes addressed in each of these highly cited articles gives further insight.

Interpretation Guidance & Further Steps:

1. Contextualize the Communities: Based on the journals where the publications in each cluster appear, and the keywords associated with these publications, you can refine your understanding of the specific research themes represented by each community.

2. Examine the Bridging Publications: Investigate the articles that cite publications from multiple clusters. These “bridging” publications can reveal interdisciplinary connections and knowledge transfer between the different research areas.

3. Consider Temporal Trends: The publication years of the cited references provide a rough indication of the evolution of the research area. Note any shifts in focus over time (e.g., a transition from theoretical foundations to more applied research).

4. Limitations: Remember that co-citation analysis reflects citation patterns, which are not necessarily direct measures of influence or quality. Factors such as journal visibility and author reputation can also influence citation rates.

By combining this network analysis with a thorough reading of the key publications identified, you can gain a deeper understanding of the intellectual structure and research dynamics within your chosen field.

Historiograph

Overall Structure and Temporal Trends:

The network spans from 1988 to 2020, suggesting a sustained interest in the topic of servitization and related areas. The structure indicates a foundational paper in 1988 (“vandermerwe s, 1988”), which serves as a starting point for subsequent research. The density of citations appears to increase from the mid-2000s onwards, especially around 2015-2019, indicating a surge in research activity during this period.

Key Citation Paths and Pivotal Works:

• Vandermerwe (1988) as a Foundation: The paper by Vandermerwe in 1988 appears to be a core foundational work. The network shows multiple citations radiating from it, meaning it probably introduced central concepts of the analyzed research field. The title “Modelling Condition-Based Maintenance To Deliver A Service To Machine Tool Users” suggests an early focus on servitization in the context of industrial equipment and maintenance.
• Baines et al. (2007-2009) as Consolidating Works: The cluster formed by “baines ts, 2007”, “neely a, 2008”, “baines ts, 2009” and “martinez v, 2010” acts as a crucial bridge between the foundational work and the later developments. “Baines ts, 2009” addresses the role of “Information And Communication Technologies Enabling Servitized Manufacture”, “neely a, 2008” examines “Service Supply Chain And Its ‘Bullwhip Effect'”. This suggests the cluster refines the industrial perspective from 1988 towards a more modern understanding of servitization.
• Mid-2010s: Diversification and Specialization: Around 2015-2017, we see a diversification of research themes. Papers like “opresnik d, 2015”, “reim w, 2015”, “lerch c, 2015” and “benedettini o, 2015” suggest the network start to cover different aspects of servitization. We see topics as business logic, service process mapping and the algorithm for resource search. This phase might be characterized by increased specialization and the application of servitization concepts to various industries and contexts.
• Late 2010s: Innovation, Strategy, and Implementation Challenges: From 2017 onwards, papers such as “coreynen w, 2017” (“Servitization Strategy”), “kowalkowski c, 2017” (“Development Of Total Innovation Management Approach For Service Innovation” and “Integration For Manufacturing Servitization And Informationization”) and “vendrell-herrero f, 2017” (“Management Accounting Roles In Supporting Servitisation”) point to a focus on strategy, organizational capabilities, and the practical implications of servitization. The presence of “raddats c, 2019” (“Innovation Capability Of Chinese Manufacturing Enterprises Based On Manufacturing Servitization”) indicates a geographical and contextual expansion of the research.

Temporal Evolution of Clusters:

1. Early Stage (Pre-2010): Characterized by Vandermerwe (1988) and the subsequent cluster, this phase is dominated by defining the core concepts of servitization in a manufacturing and industrial context.
2. Expansion Phase (2010-2015): A growing network with increased publications, signifying wider acceptance and exploration of servitization. Research diversifies into specific methodologies (e.g., simulation), business models, and industry-specific studies.
3. Consolidation and Refinement (2015-2020): Focus shifts to the strategic and organizational aspects of servitization. A trend towards understanding the challenges of implementation, innovation, and the financial consequences of adopting servitization strategies. The latest research has a clear focus on innovation, enterprise systems and financial consequences.

Limitations and Considerations:

• Scope of SCOPUS Database: The analysis is limited to the papers indexed in SCOPUS. Other relevant publications in different databases might be missing.
• Citation Bias: Citation networks can be influenced by citation biases, where highly cited authors or journals receive disproportionate attention.
• Granularity of Analysis: This analysis provides a high-level overview. A deeper dive into the content of each paper would be needed for a more nuanced understanding.

Further Research Directions:

• Investigate the specific content of the pivotal works (Vandermerwe, Baines, etc.) to understand their key contributions and how they shaped subsequent research.
• Analyze the abstracts and keywords of the papers in the 2015-2020 cluster to identify the emerging trends and research gaps in the field.
• Explore the influence of specific authors or research groups on the development of servitization research.
• Compare the findings with similar analyses conducted on different bibliographic databases (e.g., Web of Science) to assess the robustness of the results.

By addressing these points, researchers can gain a more complete understanding of the evolution of servitization research and identify promising avenues for future investigations.

Collaboration Network
Overall Structure and Key Observations:

• Network Topology: The network exhibits a clear modular structure, with several distinct clusters (communities) of authors visible. This suggests that research in this field is conducted within relatively well-defined groups, possibly reflecting different sub-disciplines, geographical locations, or research projects.
• Community Detection: The Walktrap algorithm identified these communities. The color coding visually separates these groups.
• Central Authors: “Parida V.” appears to be the most central and highly connected author in the network. “Pezzotta G.” and “Saccani N.” also appear to be important hubs within a separate cluster. These authors likely play a bridging role, connecting different research groups or areas within the field.
• Normalization: “Association” was used as the normalization method. This means that the strength of a link between two authors is based on how often they co-author papers relative to the overall publication frequency of each author. It is useful when you want to highlight co-authorships that are more frequent than expected by chance.

Community-Specific Analysis:

Let’s break down the visible communities:

• Pink Cluster (Pezzotta G., Saccani N., Macchi M., Boucher X): This group seems to form a cohesive community, likely focusing on a specific area within the broader field. The strong interconnections among these authors indicate frequent collaboration. The prominent position of Pezzotta G. and Saccani N. suggests they are leading figures within this particular sub-area.
• Blue Cluster (Parida V., Bigdelkowalkowski E., Ziee Bigdeli A., Nikkela, Jodin D, Wincent J): Centered around Parida V., this community is another distinct research group. The connections suggest a collaborative research focus that is possibly different from the Pink Cluster. This group is more connected with the Purple and Red clusters than the pink cluster, thus this group may be a hub for the other groups in the graph.
• Red Cluster (Ayala NF, Raddats C): This smaller cluster is connected to the Blue Cluster. It’s smaller size could signify a more focused research area or a less prolific collaboration history.
• Purple Cluster (Vendrell-Herrero F., Kamp B, Parry G): This cluster is connected to the Blue and Green clusters. It seems smaller in comparison to the others.
• Green Cluster (Liu Y, Zheng P, Wang Z): This cluster seems to be composed of the Chinese researchers.
• Orange Cluster (Zhang Y, Wang H, Li H): This cluster appears to be connected to the Green Cluster.
• Brown Cluster (Taisch M, Terzi S, Sasanelli C): This group seems isolated in comparison to the others.

Interpretation and Implications:

• Collaboration Patterns: The network reveals that collaboration is a significant feature of research in this field. The presence of distinct clusters suggests that there are different schools of thought, methodologies, or research questions being pursued by different groups.
• Bridging Researchers: Authors like Parida V., who are highly connected across clusters, play a crucial role in knowledge transfer and integration within the field. Their work likely bridges different sub-areas and facilitates the exchange of ideas.
• Potential Research Directions: Investigating the specific research topics and methodologies of the different clusters could reveal emerging trends, unresolved debates, and potential areas for future research. For example, analyzing the publications of the Pezzotta G. and Saccani N. cluster versus the Parida V. cluster could highlight different approaches or sub-fields.
• Impact of SCOPUS Database: It’s important to remember that this network is based on data from SCOPUS. The database’s coverage may influence the representation of different regions or research areas.

Critical Discussion Points:

• Scope Limitations: This analysis is limited to the data within the SCOPUS database. A broader analysis incorporating other databases (e.g., Web of Science) might reveal additional collaboration patterns.
• Collaboration Strength: The “association” normalization method can be sensitive to authors with very few publications. A different normalization method might highlight different relationships.
• Temporal Dynamics: This is a static snapshot of collaboration. Analyzing how the network evolves over time could provide insights into the changing landscape of the field.
• Qualitative Context: While the network provides a quantitative overview, it’s essential to supplement this with qualitative analysis. Examining the content of the publications and conducting interviews with key researchers could provide a deeper understanding of the collaboration dynamics.
• Community Interpretation: The Walktrap algorithm provides a useful starting point, but the identified communities should be critically evaluated. Do they align with known sub-disciplines or research groups? Are there alternative community structures that might be more meaningful?

Recommendations for Further Analysis:

1. Keyword Analysis: Analyze the keywords associated with each cluster to identify the specific research topics being addressed.
2. Citation Analysis: Examine the citation patterns between clusters to understand the flow of knowledge and influence.
3. Temporal Network Analysis: Create a series of networks over different time periods to track the evolution of collaboration patterns.
4. Content Analysis: Conduct a qualitative analysis of the publications to understand the nature of the collaboration and the research questions being addressed.

By combining this network analysis with a deeper understanding of the research context, you can gain valuable insights into the structure and dynamics of this field. Remember that bibliometric analysis is just one piece of the puzzle, and it should be complemented by other research methods.

Countries’ Collaboration World Map
Overall Observations:

• Major Scientific Hubs: The map clearly indicates that the United States, Europe (especially Western Europe), and China are the major hubs of scientific production. The deep blue shading signifies a high volume of publications originating from these regions.
• Collaboration Intensity: The density of connecting lines (representing co-authorship) is particularly high within Europe and between Europe and the US. This suggests strong and frequent collaborative research efforts among these countries.
• Global Collaboration Patterns: While the US and Europe are heavily interconnected, China appears to have a significant number of collaborations extending outwards, suggesting a rapidly growing engagement with the global research community. We can observe collaborative links to Australia, and Brazil as well.

Specific Country Insights:

• United States: The US appears as a strong hub, with significant connections to Europe and possibly some to East Asia. This likely reflects established research relationships and the historical strength of US-based scientific institutions.
• China: China stands out with high research output. The presence of collaborations with numerous countries suggests a strategic effort to integrate into the global scientific network. It would be interesting to analyze whether the strength of collaboration (measured in number of co-authored papers) is evenly distributed or concentrated with specific partners.
• Europe: The European countries show very intense collaboration within the continent, forming a densely interconnected network. Strong links also exist to the US and to a lesser extent to other parts of the world.
• Australia: Australia features moderate research output and connections, predominantly to Europe, the US and Asia.
• Brazil: Appears to have some collaboration with Europe and North America

Interpretation and Discussion Points:

1. SCOPUS Bias: Remember that this analysis is based on SCOPUS data. SCOPUS has a certain coverage profile, which may influence the observed patterns. For example, if SCOPUS has a stronger representation of English-language journals, collaborations involving countries that primarily publish in other languages might be underrepresented.

2. Historical and Political Context: International collaborations are often shaped by historical relationships, political alliances, and funding initiatives. The strong US-Europe link likely reflects long-standing research partnerships and transatlantic funding programs. China’s increasing collaboration reflects its growing scientific prowess and strategic international engagement.

3. Research Areas: The collaboration patterns could also be discipline-specific. Certain research areas might be more prone to international collaboration than others. Further analysis, possibly involving keyword analysis, could shed light on the subject areas driving these collaborations.

4. Data Granularity: The map provides a high-level overview. It would be useful to drill down to specific institutions or research groups to understand the micro-level dynamics of collaboration.

5. Collaboration Strength: The map indicates *existence* of collaboration but not its *strength* (e.g., the number of joint publications). A weighted network visualization or a table showing the number of co-authored papers per country pair could provide a more nuanced understanding.

Suggestions for Further Analysis:

• Temporal Analysis: Analyze how these collaboration networks have evolved over time. Is China’s engagement increasing rapidly? Are new collaborative links emerging?
• Network Metrics: Calculate network centrality measures (e.g., betweenness centrality) to identify countries that play a key “brokerage” role in the global collaboration network.
• Discipline-Specific Maps: Create separate maps for different research areas to see if collaboration patterns vary across disciplines.
• Compare Databases: Repeat the analysis using data from Web of Science or Dimensions to assess the robustness of the observed patterns.
• Funding Data Integration: Overlay funding data to see if specific funding programs drive international collaboration.

By considering these points and conducting further analysis, you can develop a deeper and more nuanced understanding of the global landscape of scientific collaboration.