ANALYSIS OF THE STRUCTURAL EMBEDDEDNESS OF A SCIENTIFIC RESEARCH LABORATORY’S SUPPLIER NETWORK

Objective: The objective of this research is to map a scientific research laboratory’s supplier network using Ucinet software and analyze its structural network. Theoretical Reference: Supply chain management is currently becoming increasingly important in companies as a decision-making strategy. The software tool based in Social Network Analysis (SNA) Ucinet allows modeling of a supply network based on its actors. By analyzing the supplier network, it is possible to identify opportunities for improvement that may not have been seen in previous diagnoses. Method: Data were collected using a questionnaire. Ucinet software and its integrated module NetDraw were used for data analysis, which allowed the data to be visualized in graphical format, to model the network and obtain structural indicators. Results: From the analysis of the supply network, improvements for the management of the laboratory’s logistics purchasing process were suggested. Implications of the research: The management of the laboratory’s logistics purchasing process helps researchers and managers and adjust its sizing parameters to the results obtained. Originality/Value: Ability to model the supplier network and obtain structural indicators, contributing to a better clarification of the exchange relationships between the authors.


INTRODUCTION
In recent years, a great deal of attention has been paid to studies on internal and external processes of suppliers integrated into the supply chain.The network structure allows organizations to analyze their suppliers more critically due to access to new knowledge, increasing the quality of the products and services provided (Tsai, 2001).
The laboratory has been considered an important instructional medium in science teaching since the beginning of the 20 th century (Fay, 1931).For a research laboratory to be able to serve its researchers, it needs to have the resources available through its suppliers.And through network analysis it can configure all the parties involved in the supply chain, directly or indirectly, input suppliers, manufacturers, distributors and end customers (Chopra & Meindl, 2011).According to Gerônimo et al. (2020), it is necessary to adopt actions that bring consistent responses to the management processes, as well as to build an efficient management of available resources and inputs, besides, it is also important to minimize waste generated by operational processes.By mapping and analyzing the research laboratory's suppliers, it is possible to make visible the patterns of information flow and collaboration in important groups of actors with strategic value as a competitive advantage (Balestrin et al., 2014;Cunha, 2004;Schmidt, 2010).
When analyzing an organization, i.e. an organizational network, its information serves as an important tool to understand the structure of exchanges between the authors, and thus visualize the position of the suppliers in order to more quickly achieve the objectives proposed in the formal channels or influence organizational behavior (Bastos;Santos, 2007).
Thus, the aim of this study is to map a network of suppliers to a scientific research laboratory and analyze it from a structural perspective.By investigating the patterns of ties that are established between suppliers, it seeks to identify opportunities or restrictions that sustain the structure, thus contributing to a better clarification of the exchange relationships between the authors.
The supplier network was modeled and the structural indicators were obtained using Ucinet 6.718 software and its integrated model NetDraw 2.174, which has characteristics to analyze the network structure.

NETWORK PROPERTIES
Generally speaking, a network is basically made up of two elements: the nodes and the relationships between each of these nodes.A company or individual takes on the role of a node within this network.In this way, each participant has a responsibility in the network in which they are inserted, varying according to their position, be it supplier or customer (Thoben & Jagdev, 2001).Figure 1 shows the most common types of networks.

Types of nodes
Source: Thoben & Jagdev, 2010.The complexity of the network lies in its number of nodes and the number and type of relationships between each of these participants.Consequently, the greater the number of nodes, the greater its complexity and difficulty to manage it (Thoben & Jagdev, 2001).
Besides, Jagdev & Thoben (2001) define some types of relationships between the nodes of a network, which are: commercial transactions; non-contractual agreements; contractual agreements; joint ventures and integrated companies.
The commercial transaction relationship occurs between two organizations in which their relationships are entirely based on commercial negotiation, such as payment methods, credibility and agreements.Non-contractual agreements, on the other hand, are negotiations based on trust between the two parties, and generally this type of relationship tends to last longer than the commercial relationships mentioned above.In contrast, contractual agreements are also common between organizations, where one company is inserted into another's network as a step in the process and flow.The joint venture relationship implies a relationship in which a group of organizations come together to benefit each other, usually in these relationships a new organization emerges.Finally, the integrated company is defined by the network relationships formed through the different sectors and hierarchies formed within the same organization (Thoben & Jagdev, 2001).
As for the type of network, each configuration shown in the previous figure has its own characteristics.In the discussion and analysis section, it will be shown that the network in this study has a tree-type configuration.This type of arrangement can take on a convergent or divergent pattern, the former being used to control network processes and transitions and the latter being generally used in distribution networks (Jagdev & Thoben, 2001).

STRUCTURAL ANALYSIS OF NETWORKS
The structural network analysis approach describes the overall architecture of the network, covering the properties of the links between companies as a whole (Dacin, 1999).
Positions and roles are perceived by observing the similarities of the connections that exist between the nodes of a network (Makagon, 2012).The variables of this concept include closure, density, connectivity and hierarchy (Wasserman & Faust, 1994).
According to Lemieux & Ouimet (2008), the several pieces of information that are collected about the links between actors can be measured, i.e. scores are assigned at various levels of measurement and these measures can determine the quantity at multiple levels of analysis, but are usually used at the individual and group or subgroup levels.In the context of this study, metrics were selected to assess the centrality of suppliers in a network, which are: degree centrality, closeness centrality and betweenness centrality.The degree centrality measure shows how the network is structured (Lazzarini, 2008).
The centrality degree indicates the intensity with which a company is connected to others.(Balestrin & Verschoore, 2016;Cerqueira, 2014).Closeness centrality is related to the proximity or distance of an actor in relation to the other actors in a network.The proximity of a vertex is calculated by dividing the total number of other vertices by the sum of all the distances between that vertex and each of the other vertices, as it can be seen in Equation 02( Borges & Mourão, 2013).
Betweenness centrality is related to the number of geodesic paths that a node must take to reach other nodes.The higher the number of geodesic paths that a source node must take to a destination node, the lower the closeness centrality of this source node is (Borges & Mourão, 2013;Cerqueira, 2014).
Network density represents the proportion of possible lines that are actually present in the network.The density measure shows whether the network has high or low connectivity, i.e. 6 when networks are dense, they allow a greater flow of information between actors.In terms of size, it corresponds to the number of actors in the network, while stability represents the frequency of actors entering and leaving the network (Wasserman & Faust, 1994).

SUPPLIER NETWORK
Suppliers are a crucial component of the supply chain, which in turn can be defined as a set of three or more entities (whether organizations or individuals) that are directly involved in the upstream and downstream flow of products, services, finances and information, whether from a supplier or a consumer (Mentzter, 2001).In other words, the members who take part in this process form a network, even though they are independent, they are connected to each other.
In order to contribute to the study of relationships in supply chains, network theory plays a role in helping to analyze relationships (Borgatti & Li, 2009).For Sacomano Neto & Truzzi (2009), the network as an analytical tool is based on the structure of relationships to understand a wide range of aspects present in the relationships of the actors.
An analysis of the supply/supplier network is of paramount importance for a better understanding of how each component behaves within the network and what impacts and levels of importance each organization or individual exerts according to their position in the network (Capioto et al., 2019).Understanding this dynamic of relationships contributes to the management of partnerships between different organizations and the opportunity to systematize excellence practices aimed at transferring knowledge and technologies between companies in the same network, which is fundamental for the evolution of a sector (Barbosa & Musetti, 2012).

RESEARCH METHOD
This research was carried out in two stages.The first comprised a complete bibliographic review of the relevant subjects.In the second stage, the necessary data was collected using an interview questionnaire divided into two modules, the first characterizing the laboratory and the second characterizing the laboratory's supplier network.
The study was carried out in a laboratory that conducts research into mutagenesis and environmental monitoring using tumor and non-tumor cell culture test systems, mice (Wistar lineage), fish (Astyanax Altiparanae) and Allium Cepa, evaluating cytotoxicity, genotoxicity, Data collection was carried out through two interviews with the person responsible for purchasing the laboratory's supplies.These interviews collected the necessary data on the laboratory's suppliers, as well as providing an understanding of the degree to which the site's supply process is managed.
Based on the information provided by the laboratory, a questionnaire was formulated using the roster call method (Morrison & Rabellotti, 2009), which consists of providing a list of all the members of each actor in the network, so they can mention who they have relationship with.
Connections between the actors in the network were classified according to the degree of interaction, using a 3-point likert scale, where 1 point represents low interaction, 2 points a moderate interaction and 3 points a high interaction between the actors.This degree of interaction was obtained through a questionnaire answered by the person responsible for the laboratory's input purchasing process.
Three structural indicators were chosen to evaluate the network studied: degree centrality, mediation centrality and proximity centrality.According to Lazzarini (2008), these measures indicate how the network is structured, in other words, the network can be characterized by how the relationships between the actors are established.The indicators were established using Ucinet 6.718 software and its integrated module NetDraw 2.174, which allowed the data to be visualized in graphical format, in order to model the network and obtain the structural indicators.Finally, through observations on the results of the quantitative analysis and the characteristics of the network, the main problems were listed and, with some references, improvements were suggested.

ANALYSIS AND DISCUSSION
Products consumed by the laboratory and supplied through the network under study can be classified into the following categories: reagents, chemical compounds, glassware, equipment, guinea pigs, cell lines, hospital products and general consumables.It is important to note that there are cases in which a supplier can supply more than one product category.
Twelve suppliers to the research laboratory were identified.Of these, just one is considered indispensable to the laboratory, as it is an exclusive supplier of a line of products for cell analysis used in the research carried out in the laboratory.
Due to the low level of management observed in the laboratory's purchasing process, there is no change of suppliers in the process of purchasing inputs.As demand for products arises, requests for quotes are sent to all suppliers who sell the product line required.
Mechanisms or criteria for selecting and evaluating suppliers were also not identified.
On account of the activity's nature carried out in the laboratory, i.e.Scientific research, the most important factor in the relationship with suppliers is the quality of the product supplied.
It is extremely necessary that all products and materials used in the research stages have a highquality level, which can be expressed in terms of precision, for instance, in the case of equipment and the composition of chemical reagents or standardization.
The supply chain consists of the laboratory in question, represented in Figure 2 as node L, and twelve suppliers, represented by nodes F1 to F12, resulting in 13 actors in the network.The modeled network presents a total of 13 actors, 12 of which are suppliers of inputs to the laboratory, the network's focal actor.In the supply chain studied, there are seven suppliers who act exclusively as first-tier suppliers to the laboratory, two in the second tier, and the remaining three act as both first and second-tier suppliers.The thickness of the vertices connecting the first-tier suppliers and the laboratory represents the degree of interaction 9 between the actors in the network, so that the thickest vertices represent the greatest degree of interaction between the nodes.
Considering the number of vertices existing in the network and the total possible connections among its 13 actors, it was possible to calculate the network's density.The result obtained was 10.9% which indicates a network with low connectivity between its actors, and consequently a reduced flow of information exchange.Table 1 presents the role of each actor and its degree of interaction with the laboratory studied: for these indicators.

Table 2
Result of the centrality degree, proximity degree and intermediation degree indicators According to the results of the centrality degree shown on Table 2, the actor with highest centrality in the network is L, which means that he is the actor who holds the greatest participation in the network.He has this high degree of centrality because he maintains a relationship with all the suppliers.Among the suppliers, F4 has the higher degree of centrality because, besides acting as a first-tier supplier to the laboratory (L), it has links with two other second-tier suppliers, F2 and F8.F4 also has the highest degree of interaction with the network's focal company, supplying reagents, glassware, equipment and consumables in general.
F10 and F11 are the actors with the lowest degree of centrality.Supplier F10 is the only actor in the network to work with cell cultures, so it only has a connection with the laboratory, since the other suppliers don't work with this type of product.F11, on the other hand, is a second-tier supplier of hospital products which is only connected to F9, which in turn is connected to the laboratory.
Supplier F11 also has the lowest closeness centrality of all the actors in the network, followed by supplier F12, both of whom are second-tier suppliers to the laboratory, which is 11 the actor with the highest closeness centrality in the modeled network, with greater power to disseminate information compared to the other nodes.
In terms of centrality of intermediation, L has the highest degree of centrality, thus possessing the greatest capacity to connect the companies in the network, followed by actor F9, a first-tier reseller of hospital products, who has a connection with F11, the second tier supplier of this type of product.On the other hand, suppliers F1, F6, F8, F10, and F11 have a null degree of intermediation centrality, resulting in a low communication capacity with the other actors in the network.
The laboratory has a low level of management of the purchasing and supply process.
There is no internal stock control to help identify when supplies need to be replenished and there are no indicators or metrics related to supply chain performance that are monitored on site.Relationships with suppliers are informal, without any kind of efficiency assessment or even periodic exchange of suppliers.Estimates are requested from suppliers and the one with the best cost-benefit ratio, according to a subjective assessment by the requester, is selected to make the purchase.
Considering the current state of the laboratory's supply chain management process and the analysis of the centrality of degree, intermediation and proximity indicators, it is suggested reorganizing the process based on the information it generates, starting with the creation of a database that stores order and stock data, thus enabling the creation of basic performance indicators to evaluate each supplier individually and consequently the supply network as a whole.
Besides that, this improvement could be used to increase the flow of information between the actors in the network, resulting in an improvement in network density.This set of data would also enable a better analysis of the structural indicators of the supply network, its actors and consequently possible actions to be taken with the aim of developing the supply chain.

FINAL CONSIDERATIONS
The main objective of this article was to map a network of suppliers to a scientific research laboratory and to analyze this network from a structural perspective, involving centrality of degree, centrality of proximity, and centrality of intermediation.Applying analysis of social networks, through the software tool Ucinet and its module NetDraw, was the guiding ___________________________________________________________________________ Rev. Gest.Soc.Ambient.| Miami | v.18.n.7 | p.1-14 | e07953 | 2024.
___________________________________________________________________________ Rev. Gest.Soc.Ambient.| Miami | v.18.n.7 | p.1-14 | e07953 | 2024.7 mutagenicity and antimutagenicity, as well as histological changes.Located at a public university in the city of Maringá, approximately 12 people do their research in the laboratory, including professors and undergraduate and graduate (master's and doctorate) students from the institution.

Table 1
Identifying the role of the actors in the supply network