AN AGENT-BASED COLLABORATIVE MODEL FOR SUPPLY CHAIN
MANAGEMENT SIMULATION
C. M. Vieira A. P. Barbosa-Póvoa C. Martinho
Centre for Management Studies Centre for Management Studies Dep. of Computer Science
Technical University of Lisbon Technical University of Lisbon Technical University of Lisbon
1049-001, Lisbon, Portugal 1049-001, Lisbon, Portugal 1049-001, Lisbon, Portugal
ESTG E-mail: apovoa@ist.utl.pt Email:carlos.martinho@ist.utl.pt
Polytechnic Institute of Leiria
2411-901 Leiria, Portugal
E-mail: carlos.vieira@ipleiria.pt
KEYWORDS
Supply Chain Management, Multi-Agent System,
Simulation.
ABSTRACT
In traditional supply chain (SC), planning problems are
usually considered individually at each SC entity.
However, such decisions often influence the other
members in the chain and thus an integrated approach
should be considered. By modelling system-wide SC
networks, different SC problems, like production
planning, coordination, order distribution, among
others, can be integrated and solved simultaneously so
that the solution is beneficial to all entities in a long-
term base. In an attempt to make progress in this area,
researchers use various methods for modelling the
dynamics of SCs. In the literature review, due to their
distinctive characteristics, multi-agent-based systems
have emerged as one of the most adequate modelling
tools for tackling various aspects of SC problems. In
this work, a multi-agent supply chain system (MASCS)
model that integrates different SC processes is
presented. The proposed model allows modelling
different SCs with multi-products and different
operational policies considering information asymmetry
and distributed/decentralized mode of control.
In this article the details of the MASCS model
development and implementation are presented.
Furthermore, the applicability of the proposed MASCS
is briefly demonstrated through the solution of a SC
example. The obtained results are discussed and
research extensions are outlined.
INTRODUCTION
A supply chain (SC) is a network of trading partners
linked through upstream and downstream connections
where the main aim is to produce and deliver
products/services to the ultimate consumers so as to
provide global SC profit. Traditionally, managers have
been focusing on the management of their internal
operations to improve profitability and thus an internal
concern has been the main objective. However, supply
chain management (SCM) calls for the integration of
the SC operational activities in order to
organize/manage the flows between entities as if they
form a single organization. Moreover, with the
businesses globalization, inter-organizational
coordination is becoming strategically important for
companies to augment responsiveness while
maintaining SC efficiency.
Numerous studies have demonstrated that substantial
benefits can be obtained from an integrated SCM. Such
integration provides tremendous challenges to managers
(Arshinder 2008). Although a completely integrated
solution may exist with an optimal system performance,
such solution is not always in the best interest of every
individual member. As a result, each SC member
attempts to optimize a part of the system without giving
full consideration to the impact of their myopic
decisions on the total system performance. Optimizing
the portions of the system yields sub-optimal
performance, resulting in an inefficient allocation of
scarce resources, higher system costs, compromised
customer service, and a weakened strategic position. A
key issue in SCM is then how to coordinate the
independent players to work together as a whole so as
to pursue the common goal of chain profitability.
In an attempt to make progress in this area, researchers
use various methods for modelling the dynamics of
SCs. The multi-agent system (MAS) approach has
appeared as one of the most adequate modelling tool for
tackling various aspects of SC coordination problems
(Santa-Eulalia et al. 2011). Indeed, considering the fact
that most of the SCs involve enterprises with
independent ownerships (requiring the ability to model
information asymmetry and distributed/decentralized
mode of controls), applicability of traditional modelling
approaches is limited and unrealistic (Govindu and
Chinnam 2010). Additionally, a holistic model is
required to explore different coordination mechanisms
and their value in SC where different set of
combinations of coordination mechanisms can be tried
with the help of simulation. Moreover, most of the
models describing coordination mechanisms are dealt in
two-level SC, which needs to be extended to multi-level
SCs (Arshinder 2008).
Proceedings 26th European Conference on Modelling and
Simulation ©ECMS Klaus G. Troitzsch, Michael Möhring,
Ulf Lotzmann (Editors)
ISBN: 978-0-9564944-4-3 / ISBN: 978-0-9564944-5-0 (CD)