SERVICE NETWORKS PERFORMANCE ANALYTICS
A Literature Review
Noel Carroll
Lero - The Irish Software Engineering Research Centre, Department of Computer Science & Information Systems
University of Limerick, Limerick, Ireland
Yan Wang
Eriss – European Research Institute in Service Science, Department of Information Management
Tilburg University, Tilburg, The Netherlands
Keywords: Business process management, Business process modelling, Service network, Simulation, System dynamics,
Performance metrics.
Abstract: The success of developing service networks rely on obtaining a correct understanding of the end-to-end
business processes. However, there are major concerns as to the lack of research efforts to examine
methods to successfully manage the complexity of service networks. The insufficient communication efforts
between business and technical experts results in a dissatisfactory service delivery and the inability to
predict and measure the service network performance. This literature survey is initiated with purpose of
finding a novel way to represent business processes in service networks and analyses the process
performance. Specifically, we discuss the need to conceive tools and techniques to manage the complexity
of service networks without jeopardising the performance of service networks and provide an overview of
current simulation-based modelling approaches and optimising business processes.
1 INTRODUCTION
The business and engineering world has transformed
from an object-orientation view towards a service-
orientated view. Many resources are co-created in
the service systems, including people, software
systems, computing devices and sensor networks,
organisations and shared information. In such
increasingly complex and dynamic markets and
operating environments, it requires an innovative
smart service network to place equal emphasis on
the business domain and technical domain. Thus it is
crucial to have a transparent communication
network between business modellers and technical
modellers and between business design and
Information Technology (IT).
The starting point of successfully developing a
smart service network is to have a comprehensive
picture of the process in which all the required
services are delivered and all the stakeholders are
involved, as well as to find a novel way to explain
the picture to both business and technical experts.
This enables us to integrate knowledge on people,
process and systems which make u a smart service
system. In the past, business process modelling and
simulation has been widely used to improve the
understanding of the business picture and to observe
the impact of process changes in business process
reengineering (Low et al., 2007). In this literature
survey, we look into the complex environment of
service networks and adopt the process-orientation
view to provide an overview of current simulation-
based approaches in modeling and optimising the
business processes.
2 SERVICE ENVIRONMENT
The growth in ‘service science’ as a discipline has
underscored the need to investigate the contributory
value of business processes and its influence on how
a service system (including people, technology, and
organisations) affects the delivery of organisational
performance. Within organisational and
technological management theory, understanding
301
Carroll N. and Wang Y..
SERVICE NETWORKS PERFORMANCE ANALYTICS - A Literature Review.
DOI: 10.5220/0003386903010304
In Proceedings of the 1st International Conference on Cloud Computing and Services Science (CLOSER-2011), pages 301-304
ISBN: 978-989-8425-52-2
Copyright
c
 2011 SCITEPRESS (Science and Technology Publications, Lda.)

and measuring value (i.e. application of
competences) of service networks is considered one
of the key problems which prevent the sustainability
of organisational growth. Service science explores
the value co-creation of interactions between service
systems (Spohrer and Magilo, 2008). ICT
contributes towards organisational “flattening”
(Friedman, 2006) which adds to the complexity and
evolvement of service systems (Chesbrough and
Spohrer, 2006). Technological advances continue to
act as a driving force for ‘making new patterns and a
new elevated level of value creation possible’
(Normann, 2001; p. 8).
As service networks continue to grow,
understanding the dynamic exchange of resources
which creates “value”, determined through specific
relationships and interactivity between service
systems and specifically business processes is of
significant importance. Within a service system,
measurement of performance, i.e. performance
analytics, plays a fundamental role, to inform
management of quantify activities and reduce
uncertainty by mapping business processes and their
influence on service performance. This places more
importance on the need to simulate service network
behaviour and a means to analyse, predict, and even
measure service performance.
3 BUSINESS PROCESS
ORIENTATION
Since 1990s, the concept of business process
orientation has been introduced and reported to
improve the organisation performance in terms of
faster time cycle, reduced cost, and less duplication
of work across functions (Galbraith, 2001). In
process representation, there are six common
perspectives (Lin et al. 2002): functional,
behavioural, organisational, informational,
verification and validation, and modelling
procedure, which are essential for managers to
organise the business activities, and for technical
experts to clarify the cross-functional interactions
within the business system. Within a business
system, different modelling techniques are required
to represent one or more of the aforementioned
perspectives.
3.1 Business Process Modelling
A business process may involve multiple service
providers and service users, and numerous
information systems to process the information
exchange among those stakeholders. Business
process modelling maps the business activities into a
visual representation. A business process model is a
simplified representation of a system in certain
business domain, which is used for improving the
understanding the essence of the core business logic.
Common business process modelling techniques
include (Giaglis, 2001):
• Flowcharting: static graphical representation of
process flows;
• IDEF0: models what activities a system
performs;
• IDEF3: models how the system operates;
• Petri nets: models parallel dynamic systems
and their behaviour;
• Knowledge-based techniques: links process to
organizational rules and objectives
• Role activity diagramming: models roles with
their associated activities.
3.2 Business Process Simulation
Simulation is a powerful, rigorous yet practical suite
of methods and tools that not only helps to better
understand and manage service systems at large, but
also the processes that embody them as well as their
supporting information systems. In doing so
simulation allows us to iteratively discover, define,
refine and improve our knowledge of the principles
and laws of such systems, and make more informed
and accountable decisions. Regardless of the
application domain, we may discern the following
dimensions of simulation models (Seila, 1995):
Tabel 1: Dimensions of simulation models.
Dimensions Characteristics
Stochastic
Allowing to randomly selecting some
parameter values
Deterministic
Predictable behaviour and excluding
probabilistic nature of real-world events
Steady-state
Time-invariant data aggregations or
consolidations
Dynamic Dynamic system behaviour over time
Continuous Changes occur continuously
Discrete (event)
Stable systems between two events/ time
intervals
There are three mainstream paradigms in simulation
modeling, namely Discrete Event Simulation (DES),
System Dynamics (SD), and Agent Based Modeling
(ABM), all of which have been widely used in
various areas from business and supply chain to
healthcare and urban planning (Borshchev and
Filippove, 2004). SD is a successful system thinking
approach that captures the causal relationships
within large scale systems at top management levels.
It analyses feedback loops and the emerging
CLOSER 2011 - International Conference on Cloud Computing and Services Science
302

behavioural effects, such as exponential growth or
decline, which result from them. It is appropriate for
decision making at aggregated levels, with a
comprehensive integrative perspective and relatively
minimal data requirements (Schieritz and Milling
2003). Compared with SD, DES has a narrow scope
in modelling the system at operational levels. It
models the system behaviour as its states evolve
over time by following sequential system events
(Robinson 2004), and is appropriate for detailed
analysis of a specific system or linear process. In
order to achieving the insights of how the system
performs, DES requires accurate data on the system
operation history or estimation for the proposed
system (Borshchev and Filippove, 2004). ABM
simulates the operation and collocations between
autonomous agents. Instead of global system
behaviour, ABM defines the system behaviour at
individual level, and the behaviours of many
individual agents together perform the global
behaviour. Each agent has its own individual
perception and incomplete information of an end-to-
end process, and they are able to communicate and
share information with other agents by following
their own behaviour rules.
4 PERFORMANCE ANALYTICS
A service network comprises of a complex system
which relies on the harmonisation of numerous
actors. Service performance is often influence by
external entities causing structural variability across
a service eco-system which impacts of the networks
characteristics and ultimately, its performance.
Therefore, performance analytic is critical in order
to gain a thorough understanding of what influence
service performance for two main reasons; firstly to
enhance service management decision-making tasks
with simulation results, and secondly, to feed this
information into service requirements engineering
(service computing) within a BPM lifecycle.
Figure 1 depicts the BPM lifecycle; basic BPM
view (model, simulate, implement and test, deploy
and execute) and the need to analyse performance;
BAM and service network analytics (analyse,
monitor, measure, and optimise). We encapsulate
this lifecycle view to service networks.

In alignment with performance analytics, the
Information Technology Infrastructure Library
(ITIL) has suggested to answer four important
questions. The first question is “where do you want
to be?” This suggests that organisation must be
committed to service transformation and cooperated
to meet the business objectives, mission, and vision.
The second question, “where are we now?” may be a
difficult question to answer but managers must
identify where changes are needed, for example,
people, process, practice, technology/technical
infrastructure, and data (i.e. metrics) to steer the
service towards the service vision. The third
question asks, “how do we get to where we want to
be?” which requires a more detailed plan including a
top-down (process-orientated technical
infrastructure) and bottom-up (influence the
development of processes) of a service system. The
fourth and final question is “how do we know when
we have arrived?” This is a critical question as it
determines the success criterion (which is a major
factor within service science). Therefore, it is
paramount that management focus on a number of
performance metrics.

Figure 1: BPM Lifecycle (S-Cube, 2009).
In order to implement an approach to service
network analytics, one must adopt a generic view of
the activities which are performed within a service
in order to understand how a service provides value
to the stakeholding business(es). The objective of
implementing a performance analytics strategy is
typically a means to improve the business processes
which underpin their value propositions which they
serve. This acts at the motivation to develop a
performance analytics strategy.
Figure 2 above illustrates the five tiers which
form the service network anatomy; the human and
software infrastructure and the software and human
services governed by service level agreements
(SLA) and Quality of Service (QoS); the atomic
services monitored controlled by process metrics;
the service processes managed by participant
metrics; and the business transactions managed by
network key performance indicators (KPIs). These
five abstracted levels are interconnected, and the
value of the indicators at different tiers are
influenced or co-created via metrics at other tiers.
SERVICE NETWORKS PERFORMANCE ANALYTICS - A Literature Review
303


Figure 2: Service Network Anatomy (S-Cube, 2009).
5 SUMMARY & FUTURE WORK
This paper offers a platform which provides a
general overview of the need to develop methods of
service analytics through the experimentation of
simulation techniques and summarises the
fundamental techniques to simulate service
interaction to determine service analytics. In
addition, we anticipate the service network
performance analytics offer greater transparency,
which is considered a critical factor within service
deployment and innovation to discover the service
enabling or inhibiting factors of business process
behaviour across service networks. Thus, we
propose that employing service network analytics
facilitates managers ability to (re)configure service
networks to (re)construct reusable methods and
process patterns or blueprints to support service
networks through the visualisation of dynamic
business process to open up new possibilities on the
generation of service innovation. As part of our
future work, we will examine the affordance of
various simulation techniques in analysing service
performance through a number of case studies to
report how service behaviour impacts on service
performance.
ACKNOWLEDGEMENTS
The research leading to these results has received
funding from the European Communities Seventh
Framework Programme FP7/2007-2013 under grant
agreement 215483 (S-Cube). For further information
please visit: http://www.s-cube-network.eu/. This
work was supported, in part, by Science Foundation
Ireland grant 03/CE2/I303_1 to Lero - the Irish
Software Engineering Research Centre
(www.lero.ie).
REFERENCES
Borshchev, A., and Filippov, A., 2004. From system
dynamics and discrete event to practical agent based
modeling: Reasons , techniques , tools. The 22nd
International Conference of the System Dynamics
Society.
Chesbrough, H., and Spohrer, J., 2006. “A research
manifesto for services science”, Communications of
the ACM, July, Volume 49, Number. 7, pp. 35-40.
Galbraith, J. R., 2001. Designing organizations: An
executive guide to strategy, structure, and process.
Jossey-Bass, San Francisco (December).
Giaglis, G. M., 2001. A Taxonomy of Business Process
Modeling and Information Systems Modeling
Techniques. The International Journal of Flexible
Manufacturing Systems, Volume 13, 209-228.
Forrester, J. W., 1969. Urban Dynamics. Pegasus
Communications.
Friedman, T. L., 2006. The world is flat. New York,
Penguin Books.
Lin, F.-R., Yang, M.-C., and Pai, Y.-H., 2002. A generic
structure for business process modeling. Business
Process Management Journal, Volume 8, Number 1,
pp. 19-41.
Low, M. Y. H., Turner, S. J., Ling, D., Peng, H. L., Chan,
L. P., Lendermann, P., 2007. Symbiotic simulation for
business process reengineering in high tech
manufacturing and service networks. Proceedings of
the 2007 Winter Simulation Conference, pp. 568-576.
Normann, R., 2001. Reframing business: when the map
changes the landscape. Chichester, New Sussex:
Wiley
Robinson, S., 2004. Discrete-event simulation: from the
pioneers to the present, what next? Journal of the
Operational Research Society, Volume 56, pp. 619–629
S-Cube, 2009. Documents available from Website:
http://www.s-cube-network.eu/
Schieritz, N., & Milling, P. M., 2003. Modeling the Forest
or Modeling the Trees A Comparison of System
Dynamics and Agent-Based Simulation. Proceedings
of the 21st International Conference of the System
Dynamics Society.
Schwaninger, M., Ulli-Beer, S., and Kaufmann-Hayoz, R.,
2008. Policy Analysis and Design in Local Public
Management: A System Dynamics Approach. In:
Handbook of Transdisciplinary Research. Heidelberg:
Springer, pp. 205-221
Seila, A. F., 1995. Introduction to simulation. In:
Proceedings of the 1995 Winter Simulation
Conference, pp. 7-15
Spohrer, J., and Maglio, P. P., 2008. The emergence of
service science: Toward systematic service
innovations to accelerate co-creation of value.
Production and Operations Management, Volume 17,
Number 3, pp. 238-246.
Sterman, J. D., 2000. Business Dynamics: Systems
Thinking and Modeling for a Complex World.
McGraw-Hill.
CLOSER 2011 - International Conference on Cloud Computing and Services Science
304