The competitiveness of companies is determined by institutions, policies and other factors, determining their continued presence in the market, profitability and ability to adapt production to demand (Chao et al., 2015) while reducing vulnerability to competition and being resistant to erosion by substitute products (Porter, 1999; Subramanian et al., 2014). The resource-based view emphasizes that internal resources and productive capacity are crucial to maintaining organizational performance (Huang, 2016). The strength of organizations is determined by their competitiveness and the strategies they adopt to gain competitive advantage and strengthen their positions in the market (Nara et al., 2013). The potential of a resource to be widespread can be decisive for a company to sustain its competitive advantage (Liao et al., 2015) that may still be directly linked to demands and work-related stress and how to deal with them (Dijkhuizen et al., 2016). In SMEs (Small and Medium Sized Enterprises) focused on high technology launching new products is an important way to maintain their competitive advantage (Pan et al., 2018). It can be noticed that several factors influence in the competitiveness of companies and it is important to determine these factors and the areas in which they can gain competitive advantage.

In 2016, SMEs accounted for 99.8% of all companies, 57.4% of added value, and 66.8% of jobs generated in the European Union (European Union, 2016). This fact evidences how these companies are crucial to the global economy. SMEs are characterized by close relationships with customers, employees and suppliers (Mayr et al., 2017) and this proximity allied to entrepreneurship favors the long-term competitiveness of these companies (Pérez-Luño et al., 2016). SMEs operate under more difficult conditions with less access to modern solutions, that leads them to seek new paths with innovative solutions to be competitive (Staniewski et al., 2016) and to take certain risks with proactivity and external links being key factors in moving SMEs to success through innovative attitudes (Gunawan et al., 2016). The strength of SMEs is in faster responses to situations through decision making that results in better product quality and innovation, while their weaknesses are in the areas of marketing, capital generation, technology, and finance (Gunasekaran et al., 2011). Another area of weakness results from SMEs’ failure to implement strategies and systems to measure organizational performance (Taylor and Taylor, 2014). Despite the difficulties, SMEs need to interact with those in their business environment to eliminate weaknesses and obtain improved competitiveness, as well as to establish strategies according to the market where they operate. For that purpose, it might be advantageous to propose the monitoring of the competitiveness level by measuring it.

The measurement of competitiveness is an important strategic and motivational tool in management that enables organizations to monitor and optimize actual performance (Kozená and Chládek, 2012). Such evaluation should be guided by the use of uniform and reliable benchmarks (Zinnes et al., 2001). To enable the measurement of competitiveness, the adoption and use of key performance indicators (KPIs) is suggested.

KPIs enable managers to assess realistically how far goals and objectives have been achieved (Rebelo et al., 2014; Personal et al., 2014) and should be defined according to the objectives to be achieved by the companies (Andres and Poler, 2016). KPIs are a set of metrics that help companies to manage and seek continuous improvement in productivity, quality assurance, operational performance, and efficiency (Kang et al., 2016). Thus, KPIs provide benchmarks and fundamental information to gauge companies’ competitiveness, assist in analysis and decision on strategies to be adopted, to enable companies to adjust to evolving environments.

However, SMEs often experience difficulties in translating knowledge into opportunities, mostly due to a lack of research activity and systematic development (Agostini and Nosella, 2017). In SMEs, most decisions are based on the capacity and expertise of its owners (Roy and Dangayach, 2015), and this makes the subject competitiveness important for the academy, because it is from the academy that studies arise to facilitate organizations management.

Networks formed by relationships between words can involve both relations between words that form a phrase, and the network of words in a dictionary, for example. The first complex networks to be studied were the networks of citations (Newman, 2001a; Newman, 2001b). Networks can be represented by electrical networks, links between web pages, representation of integrated circuits, road and rail networks. A social network can also be considered as a group of people interacting with each other (Scott, 2000). They are characterized by having a broad user base with various differences, with participation rates that vary widely and therefore have a high degree of uncertainty. This needs to be considered when modelling a large-scale decision-making process (Herrera-Viedma et al., 2017; Dong et al., 2018; Liu et al., 2016b; Zhang et al., 2018).

The networks and the relationships between them present two problems that merit attention. The first is that the large number of agents and background can lead to uncertainty or even inconsistency of information, making it difficult to assess the quality and the union of the information provided. The second problem is that it is desirable, and even indispensable, to achieve a solution that is accepted by the majority of the members or at least to assess the degree of agreement between the parties. One of the great challenges in any decision-making scenario within SME is precisely to obtain a complete and unanimous solution agreement between all those involved (Pérez et al., 2010; Urena et al., 2019). In this article we tried to present that there are several authors that research on the same theme, forming a network, and that through network analysis it is possible to find the most appropriate author in each subject.

Among the articles published that make use of statistical software for the study and analysis of science, the following can be highlighted: Bornmann and Ozimek (2012) made use of the Stata commands to import bibliometric data and processes of information about authors. Elango et al. (2013) used Sci2 Tool and Ucinet to perform a scientometric analysis on nanotribology research. Jamali (2013) used Pajek, HisCite, Sci2 Tool and VOSViewer for analysis and visualization of the citations on the human behavior information theories. Wood and Khan (2015) used the technique of social network analysis to understand the semantic and knowledge networks related to international trade. Badar et al. (2015) used Publish or Perish software tool in addition to social network analysis, and Poisson regression techniques to explore a network of co-authorship in chemistry, in Pakistan. Liu et al. (2016a) have built networks of co-citations and co-occurrences of keywords in the area of magnetic nanoparticles using CiteSpace III software tool. Bernabò et al. (2016) conducted geospatial and temporal analyses using Citespace and Sci2 Tool software tool in the area of reproductive biology. Biglu et al. (2016) applied Sci2 Tool and CiteSpace to create co-autorship networks and co-organizations using breast cancer as the theme. Boyack (2017) created synonym terms maps in the area of Astronomy with the help of the Public Library of Science (PLOS) and Pajek software tool. Guan et al. (2017) constructed knowledge networks through Sci2 Tool. Yao et al. (2017) used Sci2 Tool for geospatial, topical and collaborative analysis of neuroimaging initiatives of Alzheimer´s disease. Baier-Fuentes et al. (2018) presented an overview of Academic Research on International Entrepreneurship using VOSViewer.

Nerur et al. (2008) traced the evolution of the intellectual structure of the area of strategic management, using a co-citation analysis and a Pathfinder analysis of the main authors of the field. For Abbasi et al. (2012), scholars with more co-authors and those who exhibit higher levels of distance centrality perform better in terms of research. So, it is possible to notice the use of statistical software such as Sci2 Tool to study and map Science, but its use for mapping authors’ networks in certain areas of Science, which becomes the proposal of this article, is still an unexplored area.

This study aims to present a scientometric analysis, based on authors’ network maps made by Sci2 Tool software, used to detect who are the main authors, their research networks and interconnections in published articles, on topics such as competitiveness, and its measurement, key performance indicators (KPIs) and small- and medium-sized enterprises (SMEs). From this analysis, it will be possible to identify research gaps to be explored. Additionally, the knowledge about the main authors and research networks will facilitate the search on what has already scientifically produced with respect to the searched areas and authors.

To obtain a quantitative overview of articles already published on the subjects of relevance, two bibliometric surveys were carried out, taking into consideration articles from the publications indexed to Scopus and WoS databases. Table III presents the results from the bibliometric survey of Scopus database held on August 30 and 31, 2018. Table IV presents the results from the bibliometric survey of WoS database held on August 30 and 31, 2018.

The initial bibliometric analysis revealed that while retrieving individual search terms, especially the search term “measurement,” from Scopus and WoS databases, the number of resulting articles was large. However, when combinations of search terms were used, the number of published articles decreased discernibly. It can be seen that the number of articles that deal with KPIs related to SMEs and competitiveness is very small.

The next stage of this study is the construction of authors’ networks maps. In these maps, the nodes represent the authors and the bigger and the darker the black colour is, the larger the number of articles published by the author. The edges present the relations of co-authorship of articles and, the thicker and darker the edge, the greater the co-authorship relation between the authors interconnected. The number of records retrieved in each database to elaborate the authors’ networks, considering the respective search terms and the minimum number of articles an author should have published, are presented in Table V.

Fig. 2 shows a map with networks of authors with more than 10 articles published, indexed by Scopus database, retrieved using the search term “competitiveness”. This image shows fourteen large independent networks of authors, two of them interlinked: the networks with the authors Phusavat K. and Takala J. at the center, and the networks with the authors Gunasekaran A. and Skitmore M. Also seen are smaller networks interlinked with networks of other more influential authors.

Fig. 3 presents a map with networks of authors with more than 8 articles indexed by WoS database, retrieved using the search term “competitiveness”. This image shows twelve large independent networks of authors, two of them interlinked: the networks with the authors Mazzanti M. and Costantini V., and the networks with Gunasekaran A. and Cheng T.C.E.

The size of the resulting node in the networks elaborated from the Sci2 Tool correctly reflects the importance of the authors on the topic, since they also have high h-index and a considerable number of articles published on the topic. Table VI presents a ranking with the 5 authors with the highest node size in the authors’ networks maps on competitiveness. It can be seen that only Gunasekaran A. is present as one of the 5 main authors in both maps. This shows the importance of carrying out surveys covering more than one database, given the diversity of research present in each of them.

Fig. 4 presents networks of authors on SMEs from Scopus database.

It is evident from Fig. 4 that there are seven networks of authors with significant research work on SMEs. Four of them are made up of interconnected networks, showing that there are several related research groups researching together on SMEs. Fig. 5 presents the map of authors’ network on SME from WoS database.

It can be seen that the map originating from Scopus is larger than that of WoS, and the authors with more relevance are not the same. Table VII presents the data of these authors.

Fig. 6 illustrates the networks map of the main authors constructed from data retrieved using the search string “key performance indicator” from Scopus and WoS databases. On the left side, from Scopus database, we can see one bigger network with more than 20 relevant authors interconnected, and five other networks with median relevance. On the right side is the map from WoS where Yin S. appears as one of the leading authors on the subject in both maps.

Table VIII presents data about the authors on the subject “key performance indicator” in both databases.

The next maps of authors’ networks were drawn up using two search terms as a filter. Fig. 7 illustrates the maps of authors’ networks obtained from the search terms “measurement” and “competitiveness” covering Scopus and WoS databases.

The presented image shows, from Scopus maps side, three large networks and other less influential ones. Again, Gunasekaran A., Phusavat K. and Siluk J.C.M. are at the center of three major networks of authors with published research work on competitiveness and measurement. However, in the map concerning published articles about these two themes, the nodes sizes are much smaller when compared to the map on the subject of competitiveness. Gunasekaran A., which had 72-size node, passes to a node 13-size, Phusavat K. moves from a node size 30 to 12, and Siluk J. C. M. which had 31-size node, passes to a node 16-size. Around Siluk J.C.M., are Nara E. O. B. with node size 11 and Soliman M. with node size 9, being this network composed by 17 authors in all. Considering this fact, it can be inferred that these researchers publish several articles on competitiveness, but few ones with aggregation to the “measurement” topic. On the map side with WoS data appear 4 larger networks: the first one centralized by Gunasekaran A. with node size 17; the second with the authors Sadeghi L. and Rezaei G. with nodes size 10; the third with Kipper L.M. with node size 7 and composed by the authors of the network centralized by Siluk J.C.M. on the Scopus map; the fourth network centralized by Choy K.L. with node size 7.

On Fig. 8 are presented the authors’ networks using the search terms “competitiveness” and “SME” extracted from Scopus (left side) and WoS (right side) databases.

Fig. 8 shows, on the left side, 4 networks of research that address competitiveness and SMEs from Scopus database: the first centered by Mahmood A. K. with a node size 10; the second formed by Tudisca S., Di Trapani A. M., Sgroi F. and Testa R., all of them with node size 6; the third network with Gunasekaran A. as the central author; and the fourth with Martinez-Martinez D. with node size 6. There are other smaller networks complementing the map. On the right side it can be seen the authors’ network from WoS and the name of Martinez-Martinez D. appears, since it is the only node greater than 4. This reflects his importance for the themes addressed in this article.

For the next map, presented in Fig. 9, the authors with two or more articles published on “key performance indicator” and “measurement” indexed to the database Scopus were used.

The use of KPIs as an appraisal tool is a topic addressed by several authors and, as shown in Fig. 9, was represented by the formation of four major and several smaller networks. Of the larger networks, one is centered around Lanza G. with node size 12; the second is centered around Bauer M. and Schalke J. C., both with node size 9; the third around Camanho A. S. with node size 7; the fourth around Fernandes T.R. and Caldeirinha R.F.S. with node size 8; and several smaller networks. The records retrieved from WoS did not map authors’ network with relevant information, given the low number of authors’ indexed articles.

The search terms “key performance indicator” and “SME,” when used together for all the research papers indexed in Scopus database, retrieved 12 articles. Searching on WoS no articles were retrieved. As such, the creation of network maps to show most relevant authors was not possible, which reveals an area that offers scope for research and a gap that should be exploited. Another combination of search terms that retrieved a dwindling number of articles was “key performance indicator” and “competitiveness”. Searches using this combination, retrieved 48 research papers from Scopus database, 2 papers from WoS, and all articles retrieved were written by different researchers, not allowing the creation of maps of relevant authors.