Industrial Plant Layout Analysing based on SNA
M. L. R. Varela, G.D. Putnik, Madureira, A.M.
Department of Production and Systems, School of
Engineering, University of Minho
Guimarães, Portugal, leonilde@dps.uminho.pt,
putnikgd@dps.uminho.pt, amd@isep.ipp.pt
V. K. Manupati, K. V. Anirudh
School of Mechanical Engineering,
Division of Manufacturing, VIT University,
Vellore, Tamil Nadu, India.
manupativijay@gmail.com, anirudhk13@gmail.com
Abstract — Social network analysis (SNA) is a widely studied
research topics, which has been increasingly being applied for
solving different kind of problems, including industrial
manufacturing ones. This paper focuses on the application of SNA
on an industrial plant layout problem. The study aims at analysing
the importance of using SNA techniques to analyse important
relations between entities in a manufacturing environment, such
as jobs and resources in the context of industrial plant layout
analysis. The study carried out enabled to obtain relevant results
for the identification of relations among these entities for
supporting to establish an appropriate plant layout for producing
the jobs.
Keywords – Manufacturing systems, plant layout; social network
analysis; case study.
I. INTRODUCTION
Social network analysis (SNA) is the mapping and
measuring of relationships and flows between people, groups,
organizations, computers, URLs, and other connected
information/ knowledge entities. The nodes in the network are
the people and groups while the links show relationships or
flows between the nodes [1].
The growth of interest in the techniques of social network
analysis has been considerable since the 1970s and has been
especially marked in the last decades. Recent growth has been
sparked partly by the increasing emphasis on the importance of
“networking” in practical management guides and partly by the
proliferation of “social networking” websites such as Facebook
and Twitter. This has encouraged many to see the advantages
of using social network analysis [1].
In this paper in intended to analyse data through SNA, about
a study for analysing the relationship between a set of jobs that
incorporate a set of tasks, which have to be performed on a set
of resources, with the aim of preparing a plant layout, by
checking the influence that may exist between these entities
through typical measures of SNA techniques, such as centrality,
closeness and betweeness, among other descriptive statistical
analysis associated that might be of relevance to be able to
propose an appropriate layout for arranging the resources,
according to a given production process for the jobs.
This paper is organized as follows: besides this section 1 of
the introduction, a brief overview about Social Network
Analysis (SNA) is presented next, in section 2. Section 3
presents a brief literature review about some more or less
closely related work about the application of SNA techniques.
Section 4 describes the case study carried out in this work for
the supporting the establishment of an industrial plant layout,
along with the social network analysis and some important
measures of network structure. Finally, in Section 6 are
presented some conclusions and directions for future work are
discussed.
II. SOCIAL NETWORK ANALYSIS
Research on social networks has grown significantly over
the last few years [2]. A social network consists of a finite set
of actors and the ties between them [3]. The three basic
elements in social networks are actors, ties and graphs [4].
Actors are network members that can be distinct individuals or
collective units. Ties, which can be formal or informal, link
actors within a network. Graphs are visual representations of
networks, displaying the actors as nodes and the ties as lines
[5].
Social network analysis (SNA) is the study of social
structure [6]. SNA describes a group of quantitative methods
for analyzing the ties among social entities and their
implications [7]. An important aspect in social network analysis
is to identify key players in a network [8].
A. SNA measures
The main measures considered in SNA are cohesion
measures and centrality measures. Cohesion describes the
interconnectedness of actors in a network [4]. The main
measure of cohesion is the density of the network, which
corresponds to the total number of ties divided by the total
possible number of ties.
To calculate the density of the network Equation 1 was used.
D = T/PT
(1)
Where D, T, and PT refer to Density, Ties, and Possible
Ties, respectively.
Centrality measures identify the most prominent actors, i.e.
those extensively involved in relationships with other network
members [8]. The most commonly used centrality measures
are: degree, betweenness and closeness [9]. The centrality
measures of degree, betweenness and closeness were calculated
in this work through the UCINET tool.
Degree centrality is the number of actors with whom a
particular actor is directly related. Betweenness centrality and
closeness centrality are related to the distance (neighbourhood)
between actors in a network. Betweenness centrality is the
number of times an actor connects pairs of other actors [4].
Closeness centrality presents distances between actors and
describes how closely actors are connected to the entire network
population [10].
III. Literature Review
According to the authors in [11], Social Network Analysis
(SNA) provides tools to examine relationships between people.
Text Mining (TM) allows capturing the text they produce in
Web 2.0 applications, for example, however it neglects their
social structure.
In their paper [11] they apply an approach to combine the
two methods named “content-based SNA”. Using the R mailing
lists, R-help and R-devel, they show how this combination can
be used to describe people’s interests and to find out if authors
who have similar interests actually communicate. As stated by
the authors, they found that the expected positive relationship
between sharing interests and communicating gets stronger as
the centrality scores of authors in the communication networks
increases [11]. Moreover, they refer that the paper shows how
content-based SNA can be used to find people’s interests in
mailing list networks.
Additionally, by comparing communication graphs and
networks showing who has similar interests, a relationship
between the correlation of these two and node centrality could
be found. Accordingly, the authors conclude that the expected
relationship between sharing interests and communicating
exists only for very active authors while less active authors do
not answer everyone who has similar interests. Thus, they refer
that the communication efficiency can be regarded to be high
for very active mailing list authors while it is moderate for mid-
active authors. The paper also suggests using only the subjects
to find the relationship between communicating and sharing
interests because the content contains more noise [11].
Another interesting contribution is presented in [12], were
a case study examines infrastructure planning in the Swiss
water sector. According the authors, water supply and
wastewater infrastructures are planned far into the future,
usually on the basis of projections of past boundary conditions,
but they affect many actors, including the population, and are
expensive. Therefore, their objective consisted on investigating
fragmentation in water infrastructure planning, to understand
how actors from different decision levels and sectors are
represented, and which interests they follow [12]. The network
analysis they did obtain confirmed their hypothesis of strong
fragmentation, as they sated that they found little collaboration
between the water supply and wastewater sector (confirming
horizontal fragmentation), and few ties between local, cantonal,
and national actors (confirming vertical fragmentation).
Moreover, according the authors infrastructure planning was
clearly dominated by engineers and local authorities, and little
importance was given to longer-term strategic objectives and
integrated catchment planning, which was perceived as more
important in a second analysis carried out by the authors, that
went beyond typical questions of stakeholder analysis. In their
study, the authors concluded that linking a stakeholder analysis,
comprising rarely asked questions, with a rigorous social
network analysis is very fruitful and did enable to generate
complementary results. Moreover, this combination gave them
deeper insights into the socio-political-engineering world of
water infrastructure planning, which according their opinion is
of vital importance to the general welfare [12].
As stated in [13] coordination increasingly occurs through
networks of informal relations rather than channels tightly
prescribed by formal reporting structures or detailed work
processes. However, while organizations are moving to
network forms through joint ventures, alliances, and other
collaborative relationships, executives generally pay little
attention to assessing and supporting informal networks within
their own organizations. Moreover, the authors refer that social
network analysis is a valuable means of facilitating
collaboration in strategically important groups such as top
leadership networks, strategic business units, new product
development teams, communities of practice, joint ventures,
and mergers. Moreover, according to them, by making informal
networks visible, social network analysis helps managers
systematically assess and support strategically important
collaboration [13].
Another interesting example of application of SNA is
provided in [14], were is presented a mixed evaluation method
that combines traditional sources of data with computer logs,
and integrates quantitative statistics, qualitative data analysis
and social network analysis in an overall interpretative
approach. The authors propose the use of several computer
tools to assist in this process, integrated with generic software
for qualitative analysis. The authors applied their evaluation
method and tools incrementally and validated them in the
context of an educational and research project that has been
going on during three years. The use of their proposed method
was illustrated on their paper through an example consisting of
the evaluation of a particular category within their project.
Moreover, their proposed method and tools aimed at providing
an answer to the need for innovative techniques for studying
new forms of interaction emerging in Computer-supported
Collaborative Learning (CSCL); at increasing the efficiency of
the traditionally demanding qualitative methods. The authors
concluded that their methods can be used by teachers in
curriculum-based experiences; and at the definition of a set of
guidelines for bridging different data sources and analysis
perspectives [14].
In paper [15], the authors provide another kind of
application of SNA, for supply chain researchers with an
overview of social network analysis, covering both specific
concepts (such as structural holes or betweenness centrality)
and the generic explanatory mechanisms that network theorists
often invoke to relate network variables to outcomes of interest.
As stated by the authors, one reason for discussing mechanisms
is to facilitate appropriate translation and context-specific
modification of concepts rather than blind copying. Therefore,
they have also taken care to apply network concepts to both
“hard” types of ties (e.g., materials and money flows) and “soft”
types of ties (e.g., friendships and sharing-of-information), as
according to them, both are crucial (and mutually embedded) in
the supply chain context. Moreover, the authors did also aim to
point to areas in other fields that they think are particularly
suitable for supply chain management (SCM) to draw network
concepts from, such as sociology, ecology, input–output
research and even the study of romantic networks. According
to their statement, they believe that the portability of many
network concepts provides a potential for unifying many fields,
and a consequence of this for SCM may be to decrease the
distance between SCM and other branches of management
science.
As we can realise through the overview presented in this
section about applications of SNA, this kind of approaches and
underlying techniques can be applied to many different
domains, in general, and also to some more specific areas, for
instance, in the context of industrial management, and in this
paper the aim is its application for establishing an industrial
plant layout.
IV. CASE STUDY
In this study a set of 25 jobs (jobs, J1 ... J25), including a set
of tasks, varying from 1 up to 5, that have to be processed on
manufacturing resources, among a set of 5 available (R1, ...,
R5), to analyse, through the application of a SNA technique,
measures of centrality, closeness and betweeness, among other
relevant descriptive statistical analysis for supporting the
establishment of an appropriate plant layout for producing the
jobs.
In a social network analysis first a network has to be
modelled. Therefore, it was created a matrix with all ties
identified between the jobs and the corresponding
manufacturing resources, for producing the jobs. This data is
presented in the affiliation matrix expressed in Table 1 below,
were 1 is assigned in cases when a given manufacturing
resource processes an operation on a given job and 0 for the
opposite situation. The matrix was uploaded in the software
UCINET, which was the software tool for the SNA execution.
Table 1. Jobs affiliation matrix
R1
R2
R3
R4
R5
J1
0
1
0
0
0
J2
0
0
0
1
0
J3
1
1
0
0
0
J4
0
0
1
0
0
J5
0
1
0
0
0
J6
0
1
0
1
0
J7
0
0
1
1
0
J8
1
0
0
0
1
J9
0
0
1
0
0
J10
1
1
1
1
1
J11
1
1
1
1
1
J12
0
0
0
0
0
J13
0
1
1
0
1
J14
0
0
0
0
0
J15
0
1
1
1
1
J16
1
1
0
1
1
J17
1
1
0
1
0
J18
0
1
1
0
1
J19
0
1
0
0
0
J20
0
1
1
1
0
J21
1
1
0
0
1
J22
1
0
1
1
1
J23
0
0
0
0
1
J24
0
1
1
0
1
J25
0
1
0
1
0
After that, using the same software, the social network
graph was created (Figure 1). In this Figure are represented all
the ties that occurred between the different entities (jobs and
manufacturing resources) for accomplishing the underlying
tasks. With this social network graph it is possible to realize
with how many manufacturing resources the jobs interact with,
and also to observe the entities that do have more intense
activity in this production scenario.
Figure 1. Social network graph of the production scenario.
V. EXPERIMENTATION WITH SOCIAL NETWORK ANALYSIS
MODEL
The method describes how the manufacturing execution
data can be extracted and viewed as a network with nodes. Here,
from the considered case study the data has been taken which
has been analysed through various SNA tools. Finally,
identifying different characteristics of the obtained network in
detail. The SNAM is categorized into two steps: (a) network
modelling, and (b) network analysis and it is mentioned in the
following sections:
A. Network modelling
Network consists of set of nodes connected with ties
indicating interaction. This section presents how the
manufacturing system execution data can be represented as
networks. In this step we have fed the input data which is in the
form of affiliation matrix into the Ucinet software package.
Using Netdraw we have represented the matrix in the form of
collaboration network. The network is made much interesting
and meaningful by showing a variation in the representation of
the resources and jobs in terms of shape, size, and colour. The
considered resources are represented with a blue coloured
square shape and the resources are shown as a red diamond. The
arrows connecting the jobs and resources explain us that the
particular resource is required to complete the job. This
collaborative network can thus be effectively used to
understand which resource is highly influential and the inter
relations among them. With the data that has been collected
from the case study and their details with three centralities has
been shown in Table 2. The relationship between the attributes
(jobs and resources) is represented in Figure 1.
B. Network analysis
The main objective of network analysis is to breakdown and
comprehend the complex information of the structure in to
collaboration networks for potential synergies. In order to
obtain the information of the structure, crucial features of
descriptive statistics such as degree centrality, betweeness
centrality, and closeness centrality of the network to examine
the complexity, interdependencies and interrelationships
involved in it.
In this research, we have considered the three most popular
centrality measures such as Freeman’s degree, closeness and
Freeman’s betweenness centrality of each attribute. The
centrality is used to find how influential a node is in the network
and also the interrelations among them for its complete
analysis. In the below Table II, values of the three centralities
of the collaboration networks for the above referred scenarios
of 5 resources and 25 jobs is presented.
The degree centrality measures influence on the node from
and to its closest neighbour with complexity of O(n) to linearly
scale the nodes in the network, where n is the number of nodes.
The jobs and resources with higher degree centrality represents
strongly connected, whereas the jobs and resources with lower
degree centrality exhibit very less connections. Thus, we have
identified the key resources which are having higher degree
centrality can act as hubs and they can serve as the central
elements of the industrial plant.
The other two centrality measures betweenneess and
closeness give us information about the shortest path involved
among the various attributes of the network. The betweenness
centrality has the higher level of control on the information
floating between different nodes in the network while the
Closeness centrality is a measure of how closely the nodes are
connected with each other. This data can be used to analyse the
order in which the resources of the plant need to be arranged
and how close each resource need to be from each other.
From the above analysis we have gained complete
understanding of the various relations among the resources to
complete the assigned jobs effectively. The resources that are
key machines as hubs of the industrial plant used for completing
maximum number of jobs are identified by degree centrality.
Also, the arrangement of the resources in the plant based on
their interrelations is also understood through the betweenness
and closeness centrality. In this way the complete industrial
layout can be designed through the obtained statistical data.
Table 2. Centrality measures for the resources
DEGREE CENTRALITY
BETWEENESS
CENTRALITY
CLOSENESS
CENTRALITY
R2
16
142.896
27.619
R3
11
76.180
25.217
R4
11
64.360
25.217
R5
11
60.580
25.217
R1
8
24.977
23.967
J10
5
20.686
26.606
J11
5
20.686
26.606
J15
4
13.928
26.126
J16
4
10.613
25.217
J22
4
11.284
24.786
J18
3
7.478
25.217
J24
3
7.478
25.217
J13
3
7.478
25.217
J20
3
7.600
24.786
J17
3
5.292
24.370
J21
3
4.730
23.967
J3
2
1.706
23.200
J25
2
2.057
23.577
J6
2
2.057
23.577
J7
2
2.095
22.481
J8
2
0.831
21.481
J5
1
0
22.137
J1
1
0
22.137
J19
1
0
22.137
J2
1
0
20.567
DEGREE CENTRALITY
BETWEENESS
CENTRALITY
CLOSENESS
CENTRALITY
J9
1
0
20.567
J23
1
0
20.567
J4
0
0
20.567
J14
0
0
0
J12
0
0
0
V. CONCLUSION
This paper focused on the application of SNA on an
industrial plant layout problem. The study aimed at analyzing
the importance of using SNA techniques to analyse important
relations between entities in a manufacturing environment, such
as jobs and resources in the context of industrial plant layout
analysis. The study carried out enabled to obtain relevant results
for the identification of relations among these entities for
supporting to establish an appropriate plant layout for
producing the jobs.
Future work is planned for continuing to explore the
application of SNA techniques to this kind of industrial plant
layout problem, for instance, by exploring this problem in the
context of extended manufacturing environments, namely
including a big data analysis, and also to compare the
application of SNA techniques to some other existing methods
for industrial plants layouts establishment.
ACKNOWLEDGMENT
This work has been supported by COMPETE: POCI-01-0145-
FEDER-007043 and FCT – The Foundation for Science and
Technology within the Project Scope: UID/CEC/00319/2013.
REFERENCES
1. Schott, J. “Social Network Analysis”, Sage, 2013.
www.orgnet.com/sna.html
2. Borgatti, S. (2006). Identifying sets of key players in a social network.
Computational & Mathematical Organization Theory, 12(1), 21-34.
3. Borgatti, S. & Halgin, D. (2011). On network theory. Organization
Science, 22(5), 1168-1181.
4. Hawe, P., Webster, C. & Shiell, A. (2004). A glossary of terms for
navigating the field of social network analysis. Journal of Epidemiology
and Community Health, 58(12), 971-975.
5. Barabási, A. (2002). Linked: The new science of networks. Perseus Books.
6. Wellman, B. & Berkowitz, S. (1988). Social structures: A network
approach. Cambridge University Press.
7. Wasserman, S. (1994). Social network analysis: Methods and
applications. Cambridge University Press.
8. Freeman, L. (1979). Centrality in social networks conceptual clarification.
Social networks, 1(3), 215-239.
9. Opsahl, T., Agneessens, F. & Skvoretz, J. (2010). Node centrality in
weighted networks: Generalizing degree and shortest paths. Social
Networks, 32(3), 245-251.
10. Hanneman, R. & Riddle, M. (2005). Introduction to social network
methods. University of California, Riverside. URL:
http://faculty.ucr.edu/~hanneman/nettext/.
11. Bohn, A. Feinerer, I., Hornik, K., and Mair, P. (2011). “Content-Based
Social Network Analysis of Mailing Lists”, The R Journal, Vol. 3/1, June
2011, pp. 11-18.
12. Lienert, J., Schnetzer, F., & Ingold, K. (2013). Stakeholder analysis
combined with social network analysis provides fine-grained insights into
water infrastructure planning processes. Journal of environmental
management, 125, 134-148.
13. Rob Cross, Stephen P. Borgatti, Andrew Parker (2002). “Making
Invisible Work Visible: Using Social Network Analysis to Support
Strategic Collaboration”. California Management Review, Vol. 44 No.
2, Winter 2002; (pp. 25-46) DOI: 10.2307/41166121.
14. Martı
neza, A., Dimitriadisb, Y., Rubiac, B., Gómezb, E., Fuentea, P.
(2003). “Combining qualitative evaluation and social network analysis for
the study of classroom social interactions”. Vol. 41, Issue 4, December
2003, pp. 353–368. Elsevier.
15. Borgatti, S. P. Li, X. (2009) . “On Social Network Analysis in a Supply
Chain Context”. Journal of Supply Chain Management, Vol. 45, Issue 2,
pp. 5–22. Wiley Online Library.
DOI: 10.1111/j.1745-493X.2009.03166.x.