4.3.1 Per Year Publication and Citation Count

Development in a particular area of research can be estimated by looking into the annual growth trend in publication count related it. At the same, the number of times the research work is followed by other sources can provide a brief about its quality. The higher the citations of a particular work, better is the work [16]. Figure 4.2 reveals the annual voluminous publications with citation count of research work using fuzzy logic from 1990 to June 2021 in a chronological order. However, it is observed that none of the years from 1990 to 2004 could come up with more than 10 publications yearly but from year 2005, more than 10 publications related to fuzzy logic were published.

The records retrieved from the database gained 5125 citations in total with an average cite score of 11 citations per document. The first two articles related to fuzzy logic was published in year 1990 but failed to gain any citation but the article published in the next year gained a total citation of 75. In year 1993 and 1997, there was not a single publication related to fuzzy logic. In the year 2018, highest number of articles related to fuzzy logic was produced (n=44) with a maximum count of citations (n=917). The article that gained maximum citations in year 2018 was “Awasthi A. (2018)” with 184 citations. The second highest publications was observed in year 2016 with 38 documents gaining a total citations of 479. The article that gained maximum citations in year 2016 was “Nilashi M. (2016)” with 71citations. Till June 10, 2021, 32 documents were published with a total citations of 23. Figure 4.2 represents the line graph drawn by using Microsoft Power bi that comprises two curves representing total publication count (red) and citations in total (blue). For overall count of documents, there is significant correlation between publication count in year and total citations of documents (p<0.001, r=0.720). Nevertheless, there is no significant correlation observed between the total publications and citations of the top sources (p=0.088, r=-0.538) and also there is no significant correlation observed between the total publications and citations of the top authors (p=0.142, r=0.674). This is evident that with the passage of time, the use of fuzzy logic is coming into limelight and is gaining citations, which depicts large interest in the research area.

4.3.2 Prominent Affiliations Contributing Toward Fuzzy Logic

North China Electric Power University produced maximum number of documents (n=11) with a total citation of 195 followed by two universities of Turkey, i.e., Galatasaray Üniversitesi with eight documents and 93 citation count and then by Istanbul Teknik Üniversitesi with six documents and 97 citation count. Two eminent authors naming Büyüközkan, G. (PC= 5, TC=85) and Göçer, F. (PC=4, TC=83) publishing highest number of publications are currently working in Galatasaray Üniversitesi. North China Electric Power University has started publishing research work on fuzzy logic in 2010 and has already acquired maximum citations of 195 and Galatasaray University has started publishing research work related to fuzzy logic in 2006 and acquired 93 citations, Istanbul Teknik University has started publishing research related to fuzzy logic in 2010 and acquired 97 citations, University of Tehran has started publishing research related to fuzzy logic in 2004 and acquired 119 citations, National Cheng Kung University has started publishing research related to fuzzy logic in 2002 and acquired 124 citations. Table 4.1 puts light on the top 5 universities focusing on fuzzy logic on the basis of publication and citation count.

Publication count (PC), Total citations (TC), First publication year (FPY), Last publication year (LPY).

4.3.3 Top Journals Emerging in Fuzzy Logic in Major Subject Areas

In academics, more number of publications as well as research work on a specific topic signifies the necessity of it and the quality of the same represents better understanding on the subject matter encouraging thrust in the area of research work [17]. On one hand, the volume of research work can be predicted from the growing trend of publications in the particular area of research and on the other hand quality of research work in different journals can be judged from cite score (calculated on the basis of total citations acquired by the journal in the last 4 years divided by total publication count in the same time frame), total citations (number of times a particular idea is followed or preferred while writing a research work), Scimago journal rank (measures the prestige earned by the journal in a given year based on its weighted citations received by the articles published divided by its total publications in last three years), and source normalized impact per paper of the journal (evaluates the contextual citation impact via total number of citations in a specific theme) [18]. The 3 prominent journals that contributed toward eminent publications in the field of fuzzy logic are “Advances in Intelligent Systems and computing” with a count of 15 publications and 13 citations, followed by “Lecture Notes in Computer Science Including Subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics”, with a count of 14 publications and 19 citations and “Communications in Computer and Information Science” with a count of 12 publications but unfortunately no citations for it. Further, Table 4.2 portrays the major sources of publications with their Total citations, Journal cite score, Scimago journal ranking, h-index, and Source normalized impact per paper.

CS (Cite score), TC (Total citations) SJR (Scimago journal rank) SNIP (Source normalized impact per paper of the journal) PC (Publication Count), TC (Total Citations).

However, the 8 major subject areas using fuzzy analysis for publishing maximum articles are computer sciences with 34.02%, followed by Engineering with 28.06 %, mathematics with 12.39%, environmental science with 6.68% and business management with 4.98, Energy with 4.86%, Social science with 4.61% and Decision Science with 4.37 which is depicted in Figure 4.3.

4.3.4 Major Contributing Countries Toward Fuzzy Research Articles

On the basis of data collected through bibliometric analysis from Scopus database, a list of prominent countries contributing voluminously toward fuzzy logic is projected in Table 4.3.

China tops the list with maximum publication count (n=99) and citation count (n=902) followed by Turkey (PC=37, TC=633) India (PC=36, TC=331), Taiwan (PC=34, TC=647), United States (PC=29, TC=438), Iran (PC=22, TC=447), Canada (PC=19, TC=501), Australia (PC=16, TC=227), United Kingdom (PC=16, TC=172), Italy (PC=11, TC=263). It is observed that even if both Australia and United Kingdom have similar number of publications but Australia exceeds United Kingdom in Citation count by 55. Further, even if Taiwan ranks 4th in publication count but has gained higher citations for its documents in comparison to Turkey and India. India ranks 3rd in publication count but its total citation is less than Canada (ranked 7^th) which has a total of 19 publications. Table 4.3 puts light upon the international collaboration of these 9 prominent countries on fuzzy logic. It was observed that China had major collaborations with Canada, Australia and France. However, Australia has major collaboration with China but Canada has limited it with Australia and Germany. International collaboration with foreign countries enhances country wide team up to access greater knowledge, skills and expertise in vivid themes of research work, which has every potential to build a strong problem solving strategy to build a robust research base.

On the basis of the data collected from Scopus database, the co-authorship of countries, network visualization is created (Figure 4.4) via Vosviewer. In the process of mapping, minimum number of documents of a country and citations is limited to 2 each. Minimum number of documents of a country limited to 2, minimum number of citations limited to 2, out of 73 countries 44 met the threshold. For each 44 countries the total strength of co-authorship link with other countries is calculated. The countries with greatest total link strength is then selected. Only 35 countries are connected to each other. 35 countries were segregated into eight clusters. Cluster 1 (comprising of six items) Colombia, Finland, France, Italy, Spain, Taiwan followed by cluster 2 (comprising of six items) Czech republic, Egypt, Pakistan, Saudi Arabia, South Korea, United states then cluster 3 (comprising of six items) Cyprus, Germany, Greece, Hong Kong, Romania, Turkey, and cluster 4 (comprising of five items) Brazil, India, Morocco, Portugal, Russian federation, then cluster 5 (comprising of four items) Bangladesh, Iran, Lithonia, Malaysia, then cluster 6 (comprising of three items) Canada, China, Singapore, and cluster 7 (comprising of three items) Mexico, Poland, United Kingdom and ultimately cluster 8 (comprising of two items) Australia and Sri Lanka.

The size of circle represented in Figure 4.3 shows the number of documents published by the particular country. Larger the size of the circle, more is the number of documents published by the country. The countries that are represented with same color implies that these countries have more co-authored documents. Through overlay visualization it was found that Russian federation, Morocco, Saudi Arabia had started publishing articles related to fuzzy logic with an average publication year 2018.

4.3.5 Prominent Authors Contribution Toward the Fuzzy Logic Analysis

After taking into consideration all the records retrieved from Scopus database, it was discovered that approximately 1178 authors have contributed toward fuzzy analysis in vivid dimensions with a mean of 4.35 to at least 5 authors per publication. Büyüközkan, G. from Galatasaray University in Turkey has published maximum number of research articles (n=5) with highest citation count (n=85), document h-index (n=3) and author h-index (n=43) by publishing his first paper in 2006 and his last paper was published in 2021. 5 authors have produced 4 articles on fuzzy analysis in their research work. Out of those 5 authors, it was observed in Table 4.4 that Göçer, F. from Galatasaray University in Turkey has highest number of citation (n=83) by publishing his first article in 2017 and last article in 2021. Two authors from Università degli Studi di Napoli Federico II in Italy naming Acampora, G. and Loia, V. have equal number of publications (n=4) and citations (n=74) and h-index of documents (n=3) but Loia, V. exceeds Acampora, G. in author’s h-index count but surprisingly both of them had published their first article in 2009 and lastly they published article related to fuzzy in 2011. Similarly, two authors from Brazil naming Barin, A. and Canha, L.N. have equal number of publications (n=4) and citations (n=9) and h-index of documents (n=2) but Canha, L.N. exceeds Barin, A. in author’s h-index count whereas both are similar in first publishing year (2009) and last publishing year (2011).

4.3.6 Coauthorship of Authors

Figure 4.5 represents network visualization of the co-authorship of authors contributing toward fuzzy analysis. In this process, the minimum number of documents of an author is limited to 2 and minimum number of citations is limited to 10. However, out of 1178 authors only 60 met the threshold. For each of the 60 authors the total strength of co-authorship link with other authors is calculated. The authors with greatest total link strength is selected. Out of these 60 authors, the largest set of connected items were 9, these 9 authors are visible through network visualization. These nine items were further divided into three clusters. Cluster 1 comprises four authors naming LI, J; Liu, P; Wang, F; Zhang, J. Second, cluster 2 comprises three authors naming Lin z, Yin J and, Zhang L. Third, cluster 3 comprises two items authors naming Zhou, D; Zhou, P. Bigger the circle, higher is the number of documents produced by the author, such as Wang F, with highest number of publications is represented with a bigger circle in comparison to others. Different shades of the circle in over lay visualization helps in discriminating authors on the basis of their publication count and average publication year. The deep blue represents average publication year as 2016, the green color represents average publication year as 2018, and red color average publication year as 2017.

4.3.7 Cocitation Analysis of Cited Authors

The minimum number of citations of an author gained was fixed to 25 and as a result, out of 21,250 authors only 55 meet the threshold. For each of the 55 authors the total strength of co-citation links with other authors is calculated. The authors with greatest total link strength is selected. It was found that 54 items were largely connected. Further, those 54 items were divided into 5 clusters, with 1256 links and 999.75 as total link strength. Authors in similar cluster color in Figure 4.6 represents they have cited each other’s works. It can be understood by documents in same cluster is linked to one specific area/one authors work is extension of others. Cluster red (23 items), green (12 items), blue (11 items), yellow (6 items), and purple (two items) represents top down order in terms of publication year and citations, i.e., cluster 5 in purple color with two authors Govindan K, and Smith S.F are new to the filed compared to other authors in other clusters.

4.3.8 Cooccurrence of Author Keywords

The bibliometric analysis conducted on fuzzy analysis identified 1314 keywords in the research work. The keyword fuzzy logic is recognized as the most used keyword with an occurrence of 54 times in the persisting literature followed by Decision making with 17 times, fuzzy Ahp by 15 times, Ahp by 12 times and decision support system by 11 times. To elucidate the research hotspots in fuzzy analysis, keywords cooccurrence is investigated via Vosviewer by using fractional counting. The co-occurrence threshold for the keywords is limited to minimum number of occurrence to five times. Out of 1314 keywords, only 32 meet the threshold. For each 32 keywords the total strength of co-occurrence link with other keywords is calculated. Those 32 keywords were divided into six clusters, having 106 links, and a total link strength of 84.50. It is evident that the new keywords in the latest research area are: dematel, industry 4.0, fuzzy vikor. These 32 keywords are illustrated in a network visualization in Figure 4.7. The relative size of circle in Figure 4.5 denotes the co-occurrences of the keywords. The larger the circle is, more the times a keyword is selected in the publications related to fuzzy analysis. The keywords fuzzy logic has total link strength of 25 which has linked o other 19 keywords. However, the strength and similarity between the themes is exhibited by the distance between the two keywords. Same colored circle shows the similarity of topic among diverse published articles. The co-keyword network visualization shown in the figure exhibits six different clusters. As a result, six main clusters are appropriately labeled on the basis of the node circles. Cluster 1 comprised 10 items (middle left in red color), such as analytic hierarchy process, decision making, expert systems, fuzzy, fuzzy sets, fuzzy theory, genetic algorithm, performance evaluation, supplier selection, and uncertainty. Next, cluster 2 comprised eight items (upper middle in green color), such as decision support system, decision making, fuzzy decision making, fuzzy inference system, group decision making, Mcdm, multi criteria decision making, and sustainability. Following, cluster 3 comprised five items (lower right in blue color), such as advanced manufacturing technology, fuzzy ahp, fuzzy topsis, fuzzy vikor, and multicriteria decision making. Further, cluster 4 comprised four items (middle right in yellow color), such as Ahp, dematel, industry 4.0, topsis. Additionally, cluster 5 comprised three items (center in purple), such as fuzzy logic, madm, network selection. Ultimately, cluster 6 comprised two items (lower left in sea blue), such as cloud computing and Internet of things.

4.4.1 Bibliographic Coupling of Documents

Out of 481 documents contributed by over all countries toward fuzzy analysis in Scopus database, a threshold limit of documents with a minimum of 10 citations is taken into consideration. 132 documents matched the threshold. For each 132 documents the total strength of bibliographic coupling links with other documents is calculated. The documents with greatest total link strength is selected. The largest set of connected items were 99. However, these 99 items were divided into 11 clusters. Cluster 1 (19 items in red color), cluster 2 (13 items in green color), cluster 3 (11 items in blue color), cluster 4 (nine items in yellow color), cluster 5 (9 items in purple color), cluster 6 (nine items in sea blue), cluster 7 (nine items in orange color), cluster 8 (eight items in brown color), cluster 9 (six items in pink color), cluster 10 (four items in light red), cluster 11 (two items in light green) is presented in Figure 4.8. The overlay visualization of the country wide documents is portrayed where the bigger circles represents comparatively higher citations than the ones with smaller circles. Highly cited documents are Colak M (2020) with 184 citations and total link strength of 13; followed by Wang X (2020) with 170 citations and total link strength of 4; Morente-Molenera J.A (2019) with 112 citations and total link strength of 3; and Addae, B.A (2019) with 110 citations and total link strength of 6 as shown in Figure 4.8.

4.4.2 Bibliographic Coupling of Sources

A total of 481 documents are published in 323 sources in Scopus database. A threshold limit of 2 documents with a minimum of 10 citations is taken into consideration. Out of 323 sources, 39 met the threshold. For each of the 39 sources, total 4l strength of bibliographic coupling with other sources is calculated. The sources with greatest total link strength is selected. The largest set of connected items were 37. These 37 items were divided into five clusters. Cluster 1 (10 items in red), cluster 2 (8 items in green), cluster 3 (7 items in blue), cluster 4 (7 items in yellow), cluster 5 (5 items in purple) is shown in Figure 4.9. Top sources are Brazilian power electronics conference Cobep 2009 with 1 documents, 463 citations and a total link strength of 27 followed by Chinacom 2010 with 1 documents, 184 citations and a total link strength of 3; then by with WCCI 2010 with 1 document, 156 citations and a total link strength of 7, further by with ICMCE 2010 with 1 document, 124 citations and a total link strength of 24, and ultimately wit heeeic.eu with 1 document, 119 citations and a total link strength of 8. However, Correlation is not significant between number of documents published in a source and its citation (p=0.718, r=-0.023 for n=323), but there is a significant relation between publication count and its total link strength (p<0.001, r=0.412 for n=323). Further, Correlation is also not significant between citation and its total link strength (p=0.873, r=0.010 for n=_323_).

4.4.3 Bibliographic Coupling of Authors

It was discovered that 1178 authors have contributed approximately 491 documents toward fuzzy analysis. A minimum number of documents for an author is fixed as 2 with a minimum number of citations as 10. Out of 1178 authors, only 60 meet the threshold. For each of the 60 authors, the total strength of bibliographic coupling links with other authors is calculated. The authors with greatest total link strength is selected. The largest set of connected items were 59. These 59 items were divided into 11 clusters. Cluster 1 (15 items in red), cluster 2 (9 items in green), cluster 3 (8 items in blue), cluster 4 (6 items in yellow), cluster 5 (4 items in purple), cluster 6 (three items in sea blue), cluster 7 (three items in orange), cluster 8 (three items in brown), cluster 9 (three items in pink), cluster 10 (three items in light red ), cluster 11 (two items in light green) are presented in Figure 4.10. On one hand most productive authors with highest number of publications as well as highest number of citations are Acampora g with 5 documents and 152 citations (154 as total link strength), followed by Alahakoon D with four documents and 123 citations (189 as total link strength); Aviso K.B. with 4_documents and _113citations (86 as total link strength); Beskese A. with 4 documents and 108 citations (255 as total link strength). However, it is found that Correlation between total publications and citations count is significant (p<0.001, r=0.880 for n= 60), Correlation between total publications and total link strength is not significant (p=0.080, r=0.228 for n=60 and correlation between citation count and Total link strength is also not significant (p=0.569, r=0.075 for n=60).

4.4.4 Bibliographic Coupling of Countries

A total of 73 countries have contributed toward fuzzy analysis. A minimum number of documents per country is fixed as 2 with a minimum number of citations of 10. Out of 78 countries, 38 countries met the threshold. For each of the 38 countries, the total strength of bibliographic coupling links with other countries is calculated. The countries with greatest total link strength is selected. A total of 38 items were divided into 8 clusters links. Cluster 1 (seven items in red color), cluster 2 (seven items in green color), cluster 3 (six items in blue color), cluster 4 (five items in yellow), cluster 5 (five items in purple), cluster 6 (three items in sea blue), cluster 7 (three items in orange), cluster 8 (two items in light brown) is exhibited in Figure 4.11. Countries with Highly cited documents are Australia with 902 citations and total link strength of 610 followed by Azerbaijan with 647 citations and total link strength of 5; Brazil with 10 documents, 633 citations and total link strength of 363; and Canada with 18 documents, 484 citations and total link strength of 885.

1. 1. Bellman, R.E. and Zadeh, L.A., Decision-Making in a Fuzzy Environment. Manage. Sci., 17, 4, B–141, 1970, https://doi.org/10.1287/mnsc.17.4.B141.
2. 2. Zadeh, L.A., Fuzzy sets. Inf. Control, 8, 3, 338–353, 1965, doi: https://doi.org/10.1016/S0019-9958(65)90241-X.
3. 3. Dubois, D. and Prade, H., Fuzzy sets and systems: Theory and applications. Math. Sci. Eng., 33, 2, p. 198, 1980, https://doi.org/10.1057/jors.1982.38.
4. 4. Zadeh, L.A., Is there a need for fuzzy logic? Inf. Sci., 178, 13, 2751–2779, 2008, https://doi.org/10.1016/j.ins.2008.02.012.
5. 5. Blanco-Mesa, F., Merigó, J.M., Gil-Lafuente, A.M., Fuzzy decision making bibliometricbased review. J. Intell. Fuzzy Syst., 32, 3, 2033–2050, 2017, https://doi.org/10.3233/JIFS-161640.
6. 6. Merigó, J.M., Gil-Lafuente, A.M., Yager, R.R., An overview of fuzzy research with bibliometric indicators. Appl. Soft Compu., 27, 420–433, 2015, https://doi.org/10.1016/j.asoc.2014.10.035.
7. 7. Laengle, S., Lobos, V., Merigo, J.M., Herrera-Viedma, E., Cobo, M.J., De Baets, B., Forty years of fuzzy sets and systems: A bibliometric analysis. Fuzzy Sets Syst., 402, 155–183, 2021, https://doi.org/10.1016/j.fss.2020.03.012.
8. 8. Takagi, T. and Sugeno, M., Fuzzy identification of systems and its applications to modeling and control. IEEE Trans. Syst. Man Cybern, SMC-15, 1, 116–132, 1985, https://doi.org/10.1109/TSMC.1985.6313399.
9. 9. Yu, D., Xu, Z., Wang, W., Bibliometric analysis of fuzzy theory research in China. Knowl. Based Syst., 141, C, 188–199, 2018, https://doi.org/10.1016/j.knosys.2017.11.018.
10. 10. López-Herrera, G. and M. J. C., Visualization and evolution of the scientific structure of fuzzy sets research in Spain [Text], T.D. Wilson (Ed.), University of Boras, Sweden, 2009, Retrieved September 15, 2021, from http://informa-tionr.net/ir/14-4/paper421.html.
11. 11. Moher, D., Liberati, A., Tetzlaff, J., Altman, D.G., Group, T.P., Preferred Reporting Items for Systematic Reviews and Meta-Analyses: The PRISMA Statement. PloS Med., 6, 7, e1000097, 2009, https://doi.org/10.1371/journal.pmed.1000097.
12. 12. Cobo, M.J., López-Herrera, A.G., Herrera-Viedma, E., Herrera, F., Science mapping software tools: Review, analysis, and cooperative study among tools. J. Am. Soc Inf. Sci. Technol., 62, 7, 1382–1402, 2011, https://doi.org/10.1002/asi.21525.
13. 13. van Eck, N.J. and Waltman, L., Visualizing Bibliometric Networks, in: Measuring scholarly impact: Methods and practice, Y. Ding, R. Rousseau, D. Wolfram (Eds.), Cham, pp. 285–320, Springer International Publishing, 2014, https://doi.org/10.1007/978-3-319-10377-8_13.
14. 14. Burnham, J.F., Scopus database review. Biomed. Digit. Libr., 3, 1, 2006, https://doi.org/10.1186/1742-5581-3-1.
15. 15. Van Eck, N.J. and Waltman, L., Text mining and visualization using VOSviewer, in: ISSI Newsletter, vol. 7, pp. 50–54.
16. 16. Guo, Y.-M., Huang, Z.-L., Guo, J., Li, H., Guo, X., Nkeli, M., Bibliometric Analysis on Smart Cities Research. Sustainability, 11, 3606, 2019, https://doi.org/10.3390/su11133606.
17. 17. Sevinc, A., Web of science: A unique method of cited reference searching. J. Natl. Med. Assoc., 96, 7, 980–983, 2004.
18. 18. Aytac, S., & Tran, C. Y. Authorship Trends of Research Articles Published in Seven Scientific, Technical, Engineering, and Medical (STEM) Library Journals: Analysis of STEM Library Research from 2016 to 2019 Update. Science & Technology Libraries, 40, 2, 117–132, 2021. https://doi.org/10.1080/0194262X.2021.1871702