Description
Supply chain is a worldwide network of suppliers, factories, warehouses, distribution centers, and retailers through which raw materials are acquired, transformed, and delivered to customers.
IJCSI International Journal of Computer Science Issues, Vol. 7, Issue 5, September 2010
ISSN (Online): 1694-0814
www.IJCSI.org
198
A Review of Supply Chain Management using Multi-Agent System
Abstract
Supply chain consist of various components/ identities like
supplier, manufacturer, factories, warehouses, distributions
agents etc. These identities are involved for supplying raw
materials, components which reassembles in factory to produce a
finished product. With the increasing importance of computer-
based communication technologies, communication networks are
becoming crucial in supply chain management. Given the
objectives of the supply chain: to have the right products in the
right quantities, at the right place, at the right moment and at
minimal cost, supply chain management is situated at the
intersection of different professional sectors. This is particularly
the case in construction, since building needs for its fabrication
the incorporation of a number of industrial products. This paper
focuses on an ongoing development and research activities of
MAS (Multi Agent System) for supply chain management and
provides a review of the main approaches to supply chain
communications as used mainly in manufacturing industries.
KEYWORDS: Information exchanges, Multi Agent System,
knowledge sharing and supply chain management
1. Introduction
Supply chain is a worldwide network of suppliers, factories,
warehouses, distribution centers, and retailers through which raw
materials are acquired, transformed, and delivered to customers.
In recent years, new software architecture for managing the
supply chain at the tactical and operational levels has emerged. It
views the supply chain as composed of a set of intelligent
software agents, each responsible for one or more activities in
the supply chain and each interacting with other agents in the
planning and execution of their responsibilities. Supply Chain
Management is the most effective approach to optimize working
capital levels, streamline accounts receivable processes, and
eliminate excess costs linked to payments.
2. Literature Survey
Analysts estimate that such efforts can improve working Capital
levels, streamline accounts receivable processes, and eliminate
excess costs linked to payments. Analysts estimate that such
efforts can improve working capital levels by 25% [2]. Today,
the best companies in a broad range of industries are
implementing supply chain management solutions to improve
business performance and free cash resources for growth and
innovation. Supply Chain Management is about managing the
physical flow of product and related flows of information from
purchasing through production, distribution and Delivery of the
finished product to the customer. This requires thinking beyond
the established boundaries, strengthening the linkages between
the supply chain functions and finding ways to pull them
together. The result is an organization that provides a better
service at a lower cost. MihaelaUlieru et al. give a approach
based on the holonic enterprise model [10] with the Foundation
for Intelligent Physical Agents (FIPA) Contract Net protocols
applied within different levels of the supply chain. The
negotiation on prices is made possible by the implementation of
an XML rule-based system that is also flexible in terms of
configuration. According to Pericles A., the system is viewed as
an organization or collection of roles that relate to each other and
form an interaction model. Roles in the system are descriptions
of business entities, whose functions are modeled at an abstract
level. Whole system is divided in Business Description, Product
Description, and Order Management Holarchy, Manufacturing
Holarchy.ole Modeling. Author has given the following System
Fig. 1 Class Architecture for customer Agents
JADE Content Classes
Setup(addBehaviour()
)
Request_Behavio
Concept : Concept
Predicates : Predicate
Action : AgentAction
Action()
Content :
ContentManager
Content_Language:
Codec
Customer Agent
Concept
Ontology Codec Content
JADE
Predicate
Agent Action
Vivek Kumar
1
and Dr. S. Srinivasan
2
1
Research Scholar, Department of computer Science & Engineering, Suresh Gyan Vihar, University
Jaipur, Rajasthan 302 004, India
2
Professor & Head, Computer Science Department
PDM Engineering College,
Bhadurgarh, Haryana 124 507, India
IJCSI International Journal of Computer Science Issues, Vol. 7, Issue 5, September 2010
ISSN (Online): 1694-0814
www.IJCSI.org
199
Yevgeniya Kovalchuk presented a way to manage the supply
chain activities & try to automate their business processes [18].
In practice, all the activities are highly connected and
interdependent. The project is mainly focused on the demand
part of the supply chain. In particular, different methods for
predicting customer offer prices that could result in customer
orders are explored and compared in the system. RuiCarvalho et
al. presented multi-agent technology as a sound alternative to
classical optimization techniques that can contribute to solve
hard problems. To prove this point , the MAS with the following
functionalities designed : simulation of an almost infinite number
of agents, heuristics for decision making, possibility to choose
among alternative decision strategies and tactics, different
evaluation criteria and evaluation functions, different message
sequences, and stochastic or deterministic behavior.
Fig. 2 Compartmentalization of operational information coordination
José Alberto R. P. Sardinha1 et al. presented a flexible
architecture based on a distributed multi-agent architecture [15]
of a dynamic supply chain. Intelligent agents tackle sub problems
of a dynamic SCM. Authors present an implementation of this
architecture by using international test bed for SCM solutions.
The mail purpose of this work is to present a multi-agent
architecture for a generic supply chain that uses a direct sales
model, which links customers directly to a manufacturer through
the Internet. Robert de Souza et al. addressed two main issues
[6]: Can we chart the complex logistical dynamics of disk drive
manufacturing? What are the critical success factors that impact
the economics of hard disk drive manufacturing? The backdrop
for this study is the (global) supply chain much in evidence in
disk drive manufacture. Fu-ren Lin et al. analyzed the impact of
various levels of information sharing including order, inventory,
and demand information, which is based on transaction costs
[14]. This study further examines the effects on supply chain
performance in electronic commerce. Specifically, the multi
agent simulation system Swarm is employed to simulate and
analyze the buyer–seller correlation in sharing information
among business partners in supply chains. Information sharing
between firms helps to lower the total cost and increase the order
fulfillment rate. In other words, information sharing may reduce
the demand uncertainty that firms normally encounter. Onn
Shehory et al. discussed suitability of agent modeling techniques
[4] to agent-based systems development. In evaluating existing
modeling techniques, addressed criteria from software
engineering as well as characteristics of agent-based systems.
Evaluation shows that some aspects of modeling techniques for
agent-based systems may benefit from further enhancements.
This technique tries to answer the following questions: (1) which
agent-based system characteristics and software engineering
principles are addressed within AOSE modeling techniques, and
to what extent? (2) What should be the properties of the future
agent-oriented modeling techniques? Rasoul Karimi et al.
developed a new multi attributes procurement auction [11]. It is
new because it has been defined for a special model of supply
chain, in that customer has a new scoring rule, and producers
have new strategies for biding. Multi Attribute Procurement
Auction is a good solution for Supply Chain problem which fits
its requirements. The implementation of the Swarm simulation
system incorporates multiple agents with the decision model to,
in turn; determine the relationship with their trading partners.
Fig. 3 demonstrates a purchasing agent’s decision model to
determine which supplier should issue the purchase order.
Fig. 3 Swarm implementation for modeling supply chains.
A trading partner contains several agents, including order
management, inventory management, policy management,
production, production planning, and purchasing. Among them,
the purchasing agent proposes the decision model to determine
from which supplier products should be purchased. The
purchasing agent buys goods from suppliers that offer the lowest
price. The price issued by a supplier is the outcome of weighing
production cost and coordination cost. The final price is the one
issued by a supplier. Yang Hang et al. proposed a CSET
framework [13] for whole supply chain in collaborative manner
by incorporating it with the Just-in-Time (JIT) principle, known
as CSET. The core of the CSET model is based on intelligent
agent technology. This paper defines such double-agent
mechanisms in details, as well as demonstrating its merits via
simulation study. The MAS proposed here was implemented
using LISP and had as first source of inspiration the agent
creation language named RUBA [Ventura97]. The system has as
main blocks i) an environment agent, in charge of the meaningful
functioning of the system and event execution simulation
[Michael90], ii) client agents, with needs to be satisfied and iii)
firm agents, that have also needs but are capable of product
manufacturing. The system also includes a blackboard, where
agents can post their messages, and a set of communication rules
(a communication protocol inspired on the Contract Net protocol
Supplie Manufacturer Distributor Retailer
Supplie Manufacturer Distributor Retailer
Monolayer information sharing
Monolayer information sharing
Inventory
Management
Cost Management
Agent
Work Flow
Agent
Order
Management
Production Agent Production Planning
Agent
OFP Model Swarm
Statistics Swarm
OFP Observer
Swarm
IJCSI International Journal of Computer Science Issues, Vol. 7, Issue 5, September 2010
ISSN (Online): 1694-0814
www.IJCSI.org
200
[Smith80; Smith81]), common to all agents and that makes
possible message exchange. KQML [Finin94], a standard
message format, and KIF [Genesereth94], a message content
specification, served as the basis for the communication protocol
of our MAS [Huns99].The main elements of the system, agents,
blackboard and a communication protocol, are essential for
functioning. These agents are intelligent, because they are able to
present successful behavior [Albus91]. Figure 4 shows the
system behaviour. Fu-ren Lin et al. used multiagent simulation
system, swarm, for simulating trust mechanism and analyzing the
supply chain performance in four different market environments
[19].Supply chain performance is evaluated by comparing the
order fulfillment process of a mold industry both with and
without trust mechanisms. From the experimental result ,
Fig 4. Environment Agent and Agent Behavior
they found that the trust mechanism reduced the average cycle
time rate and raised the in-time order fulfillment rate as the
premium paying for better quality and shorter cycle time.
Charles M. Macal et al. gave a new approach [5] to modeling
systems comprised of interacting autonomous agents. &
described the foundations of ABMS, identifies ABMS toolkits
and development methods illustrated through a supply chain
example, and provides thoughts on the appropriate contexts for
ABMS versus conventional modeling techniques. William E.
Walsh et al. highlighted some issue that must be understood to
make progress in modeling supply chain formation [3].
Described some difficulties that arise from resource contention.
They suggested that market-based approaches can be effective in
solving them. Mario Verdicchio et al. considered commitment as
a concept [17] that underlies the whole multi-agent environment,
that is, an inter-agent state, react a business relation between two
companies that make themselves represented by software agents.
Michael N. Huhns et al. found after this research that supply
chain problems cost companies [8] between 9 to 20 percent of
their value over a six month period. The problems range from
part shortages to poorly utilized plant capacity. Qing Zhang et al.
provide a review of coordination of operational information in
supply chain [12] . Then the essentials for information
coordination are indicated.Vivek Kumar et al. gave a solution for
the construction, architecture, coordination and designing of
agents. This paper integrates bilateral negotiation, Order
monitoring system and Production Planning and Scheduling
multiagent system. Ali Fuat- Guneri et al gave the concept of
supply chain management process[16], in which the firm select
best supplier , takes the competitive advantage to other
companies. As supplier selection is an important issue and with
the multiple criteria decision making approach, the supplier
selection problem includes both tangible and intangible factors.
The aim of this paper is to present an integrated fuzzy and linear
programming approach to the problem. Firstly, linguistic values
expressed in trapezoidal fuzzy numbers are applied to assess
weights and ratings of supplier selection criteria. Then a
hierarchy multiple model based on fuzzy set theory is expressed
and fuzzy positive and negative ideal solutions are used to find
each supplier’s closeness coefficient. Finally, a linear
programming model based on the coefficients of suppliers,
buyer’s budgeting, suppliers’ quality and capacity constraints is
developed and order quantities assigned to each supplier
according to the linear programming model. Amor et al.
presented Malaca [9], an agent architecture that combines the use
of Component- based Software Engineering and Aspect-Oriented
Software Development.
Fig. 5 Conceptualization of the aspect model in Malaca
Malaca supports the separate (re)use of the domain-specific
functionality of an agent from other communication concerns,
providing explicit support for the design and configuration of
agent architectures and allows the development of agent-based
software so that it is easy to understand, maintain and reuse. Ka-
Chi Lam et al. investigated a selection model based on Fuzzy
Principal Component Analysis (PCA) [7] for solving the material
supplier selection problem from the perspective of property
developers. First, the Triangular Fuzzy Numbers is used to
quantify the decision makers' subjective judgments. Second,
PCA is employed to compress the data of the selection criteria
Warehouse
Client
Agent
Creates
Information
Blackboard
Factory
Agent
-----
------
Problem
Aspect
-role : String
-role Instance : String
+handleMessage( message : Message ) :
Message
+handleInputMessage( message : Message )
+handleOutputMessage( message : Message ) :
Message
<<enumeration>>
Aspect Scope
AGENT SCOPE
PROTOCOL SCOPE
CONVERSATION_SCOPE
Message
Return
Handles
Coordination
Aspect
Distribution
Aspect
Representation Aspect
Component
Message Transport Service
Role
Coordin
ate Role
Encoding Format
Interation
Protocol
Component
Acti
on
IJCSI International Journal of Computer Science Issues, Vol. 7, Issue 5, September 2010
ISSN (Online): 1694-0814
www.IJCSI.org
201
and eliminating the multi-collinearity among them. Third, the
linear combined score of PCA (SCOREPCA) is used to rank the
Suppliers.
Four material purchases are used to validate the proposed
selection model.The results show that the proposed model can be
adopted in construction material supplier selection by the
property developers.
Table 1: Summary
S.No. Title Name &
Authors
Explanation & Conclusion
1. “supply Chain
Models in Hard
Disk Drive
Manufacturing”
Robert de Souza
and Heng Poh
Khong
This paper seeks to address two
main issues: Can we chart the
complex logistical dynamics of
disk drive manufacturing? What
are the critical success factors that
impact the economics of hard disk
drive manufacturing?
The pressures in the disk drive
industry are classic supply chain
economics; value, timing, supply,
demand and technology
development that all play a part
into price erosion patterns. To
address such issues the authors'
postulate that the five chains
interact to give rise to
complexities, static models cannot
easily handle.
2. Modeling supply
chain Formation
in Multiagent
System
William E.
In this paper the authors highlight
some issues that must be
understood to make progress in
molding supply chain formation.
Supply chain formation is an
Walsh and
Michael P.
Wellman
important problem in the
commercial world and can be
improved by greater automated
support. The problem is salient to
the MAS community and
deserving of continued research.
3. Evaluation of
Modeling
Techniques for
Agent-Based
Systems
Onn Shehory
and Arnon
Sturm
Author discusses suitability of
agent modeling techniques to
agent-based systems development.
In evaluating existing modeling
techniques, and address criteria
from software engineering as well
as characteristics of agent-based
systems.
Based on these findings, we
intend in future research, to
address the needs of agent-based
system developers. This should be
done in order to find the required
modeling techniques and
components for building agent-
based systems.
4. Effects of
Information
Sharing on
Supply Chain
Performance in
Electronic
Commerce
Fu-ren Lin,
Sheng-hsiu
Huang, and
Sheng-cheng Lin
Findings indicate that the more
detailed information shared
between firms, the lower the total
cost, the higher and the order
fulfillment rate. And the shorter
the order cycle time. In other
words, information sharing may
reduce the demand uncertainty
that firms normally encounter.
Firms that share information
between trading partners tend to
transact with a reduced suppliers.
This work investigated the buyer–
seller relationship in electronic
commerce with an Extranet as the
platform for sharing information.
Using the Swarm simulation
system, based on transaction
costs, we have identified effects of
sharing various levels of
information between supply chain
partners.
0 0.25
M VG/V
H
G/
H
VB/VL P/L
DMs’ importance wrights
Supplies’ ratings
Uncertain
Unceatin Values
(in 5 point sale
Fuzzy refraction
Fig. 6. Membership functions of DMs' importance weights and
suppliers'ratings (modified from)
IJCSI International Journal of Computer Science Issues, Vol. 7, Issue 5, September 2010
ISSN (Online): 1694-0814
www.IJCSI.org
202
5. Commitments
for Agent-Based
Supply Chain
Management
Mario
Verdicchio and
Marco
Colombetti
As there are several analogies
between a company in a business
network and an agent, the Multi-
Agent System paradigm can be a
valid approach for modeling
supply chain networks. We
consider commitment as a concept
that underlies the whole multi-
agent environment, that is, an
inter-agent state, reacting a
6. Building
Holonic Supply
Chain
Management
Systems: An e-
Logistics
Application for
the Telephone
Manufacturing
Industry
MihaelaUlieru
and
MirceaCobzaru
Approach is based on the holonic
enterprise model with the
Foundation for Intelligent
Physical Agents (FIPA) Contract
Net protocols applied within
different levels of the supply
chain holarchy. To accommodate
differentiation of interests and
provide an allocation of resources
throughout the supply chain
holarchy, we use nested protocols
as interaction mechanisms among
agents. Agents are interacting
through a price system embedded
into specific protocols. The
negotiation on prices is made
possible by the implementation of
an XML rule-based system that is
also flexible in terms of
configuration and can provide
portable data across networks.
As the effectiveness of centralized
command and control in SCM
starts to be questioned, there is a
critical need to organize
supply chain systems in a
decentralized and outsourced
manner. Agent-based models can
easily be distributed across a
network due to their modular
nature. Therefore, the distribution
of decision-making and execution
capabilities to achieve system
decentralization is possible
through models of operation with
communication among them. The
ontology structure of the JADE
framework is, in our opinion, one
of the best designed to address the
issues of accessing and sharing
information pertinent to a specific
application.
7. A Multiagent
Systems
Approach for
Managing
Supply-Chain
Problems: new
It was modelled and implemented
a MAS with the following
functionalities: simulation of an
almost infinite number of agents,
heuristics for decision making,
possibility to choose among
tools and results
Rui Carvalho,
Luís Custódio
alternative decision strategies and
tactics, different evaluation
criteria and evaluation functions,
different message sequences, and
stochastic or deterministic
behavior.
When we applied our MAS to a
problem of SC management at
HP, we obtained results with
stock outs for every product of the
bill of materials. On the contrary,
some authors using mathematical
tools only simulated the stock out
of only one product of the bill of
materials.
8. A Multi-Agent
Architecture for
a Dynamic
Supply Chain
Management
José Alberto R.
P. Sardinha1,
Marco S.
Molinaro2,
Patrick M.
Paranhos2,
Pedro M.
Cunha2,
Ruy L. Milidiú2,
Carlos J. P. de
Lucena2
This paper presents a flexible
architecture for dealing with the
next generation of SCM problems,
based on a distributed multi-agent
architecture of a dynamic supply
chain. We define intelligent agent
roles that tackle sub problems of a
dynamic SCM.
We also present an
implementation of this
architecture used in the
international test bed for SCM
solutions, the Trading Agent
SCM competition, as well as some
experimental results.
A multi-agent design is used in
the architecture, because we
believe it facilitates the
development of modular entities
that are distributed and reusable.
The design was also used to
implement an agent entry for the
Trading Agent Competition. This
system competed against 32
entries, and was able to classify to
the quarter-finals of the 2005
competition.
9. How to Model
With Agents
Proceedings of
the 2006 Winter
Simulation
Conference
Charles M.
Macal and
Michael J. North
Agent-based modeling and
simulation (ABMS) is a new
approach to modeling systems
comprised of interacting
autonomous agents. ABMS
promises to have far-reaching
effects on the way that businesses
use computers to support
decision-making.
Computational advances make
possible a growing number of
agent-based applications across
IJCSI International Journal of Computer Science Issues, Vol. 7, Issue 5, September 2010
ISSN (Online): 1694-0814
www.IJCSI.org
203
many fields. Applications range
from modeling agent behavior in
the stock market and supply
chains
10. New Multi
Attributes
Procurement
Auction for
Agent- Based
Supply Chain
Formation
Rasoul Karimi,
Caro Lucas and
Behzad Moshiri
In this article, this constraint has
been relaxed and a new
procurement auction is defined. In
this auction, seller agents can take
different strategies based on their
risk attribute. These strategies is
analyzed and compared
mathematically.
Authors define a new MAPA
which is usable under the new
model of supply chain. In this
MAPA, the producer could have
two different strategies based on
its risk attribute. These two
strategies are compared
mathematically and also in a
simulation.
11. Multi-Agent
Decision
Support System
for Supply
Chain
Management
Yevgeniya
Kovalchuk
The research approach followed is
presented. The results achieved so
far along with the plans for future
work are given next.
Various techniques for predicting
bidding prices in the context of
dynamic competitive
environments are explored. Apart
from the SCM, the solutions can
be used in forecasting financial
markets and participating in on-
line auctions.
12. Double-agent
Architecture for
Collaborative
Supply Chain
Formation
Yang Hang and
Simon Fong
The model is supported by
double-agent architecture with
each type of agents who makes
provisional plans of order
distribution by Pareto optimality
and JIT coordination respectively
As a result, pipelining
manufacturing flow is achieved.
This is significant to dynamic
supply chain formation as it can
help to optimize constraints and
costs across production,
distribution, inventory, and
transportation.
13, Essentials for
Information
Coordination in
Supply Chain
Provide a review of coordination
of operational information in
supply chain which is classified
into information types, their
Systems
Qing Zhang and
Wuhan
impact on supply chain
performance, and the policy of
information sharing
Multi-agent computational
environments are suitable for
studying classes of coordination
issues involving multiple
autonomous or semi-autonomous
optimizing agents where
knowledge is distributed and
agents communicate through
messages.
14. Effects of Trust
Mechanisms on
Supply Chain
Performance
Using Multi-
agent Simulation
and Analysis
Fu-ren Lin ,Yu-
wei Song and
Yi-peng Lo
The multiagent simulation system
Swarm is employed to simulate
and analyze the buyer–seller
correlation in sharing information
among business partners in supply
chains
The deeper the information
sharing level, the higher in-time
order fulfillment rate and the
shorter order cycle time, as
information sharing may reduce
the demand uncertainty that firms
normally encounter. Finally, firms
that share information between
trading partners tend to transact
with a reduced set of suppliers.
15. A Multiagent
Conceptualizatio
n For Supply-
Chain
Management
Vivek kumar ,
Amit Kumar
Goel , Prof.
S.Srinivisan
Paper present solution for the
construction, architecture,
coordination and designing of
agents. This paper integrates
bilateral negotiation, Order
monitoring system and Production
Planning and Scheduling
multiagent System.
The wide adoption of the Internet
as an open environment and the
increasing popularity of machine
independent programming
languages, such as Java, make the
widespread adoption of multi-
agent technology a feasible goal
16. An integrated
fuzzy-lp
approach for a
supplier
selection
problem in
supply chain
management
Ali Fuat Guneri,
A hierarchy multiple model based
on fuzzy set theory is expressed
and fuzzy positive and negative
ideal solutions are used to find
each supplier’s closeness
coefficient. Finally, a linear
programming model based on the
coefficients of suppliers, buyer’s
budgeting, suppliers’ quality and
capacity constraints is developed
and order quantities assigned to
IJCSI International Journal of Computer Science Issues, Vol. 7, Issue 5, September 2010
ISSN (Online): 1694-0814
www.IJCSI.org
204
Atakan Yucel ,
Gokhan
Ayyildiz
each supplier according to the
linear programming model.
Fuzzy set theory approach helps
to convert decision-makers’
experience to meaningful results
by applying linguistic values to
assess each criterion and
alternative suppliers.
17. Malaca: A
component and
aspect-oriented
agent
architecture”
Information and
Software
Technology
Mercedes Amor
*, Lidia Fuentes
An agent architecture that
combines the use of Component-
based Software Engineering and
Aspect-Oriented Software
Development
Provided explicit support for the
design and configuration of agent
architectures and allows the
development of agent-based
software
18. A material
supplier
selection model
for property
developers using
Fuzzy Principal
Component
Analysis”
Automation in
Construction
Ka-Chi Lam ,
Ran Tao, Mike
Chun-Kit Lam
Tthe Triangular Fuzzy Numbers is
used to quantify the decision
makers' subjective judgments.
Second, PCA is employed to
compress the data of the selection
criteria and eliminating the
multicollinearity among them.
The model can efficiently
eliminate the multicollinearity
among the supplier's attributes
and help to reduce the trade-offs
and repeatability errors in the
selection process.and the proposed
selection model can also reduce
the subjective errors on the sense
that the weight assigned for each ?
is generated automatically.
3. Conclusion
Multi-agent system is a loosely coupled network of
software agents that interact to solve problems that are
beyond the individual capacities or knowledge of each
problem solver. The general goal of MAS is to create
systems that interconnect separately developed agentsThus
enabling the ensemble to function beyond the capabilities
of any singular agent in the set-up in agent model. This
research can demonstrate that agent technology is suitable
to solve communication concerns for a distributed
environment. Multi-agent systems try to solve the entire
problem by collaboration with each other and result in
preferable answer for complex problems. For further
works, it is recommended for developing this model to
have multi retailer and even multi distributor and apply the
auction mechanism between them.
References
[1] Ali Fuat Guneri, Atakan Yucel , Gokhan Ayyildiz “An
integrated fuzzy-lp approach for a supplier selection problem in
supply chain management” Expert Systems with Applications 36
(2009) 9223–9228
[2] C. Iglesias, M. Garijo, J. Centeno-Gonzalez, and V. J. R.,
"Analysis and Design of Multiagent Systems using MAS-
CommonKADS," presented at Agent Theories, Architectures,
and Languages, 1998.
[3] Charles M. Macal and Michael J. North Tutorial on Agent-
Based Modeling And Simulation Part 2: How to Model With
Agents Proceedings of the 2006 Winter Simulation Conference
[4] Fu-ren Lin, Sheng-hsiu Huang, and Sheng-cheng Lin ,
“Effects of Information Sharing on Supply Chain Performance in
Electronic Commerce“ ,IEEE Transactions On Engineering
Management, Vol. 49, No. 3, August 2002.
[5] Fu-ren Lin ,Yu-wei Song and Yi-peng Lo ,”Effects of Trust
Mechanisms on Supply Chain Performance Using Multi-agent
Simulation and Analysis” , Proccding of the First Workshop on
Knowledge Economy and Electronic Commerce.
[6] José Alberto R. P. Sardinha1, Marco S. Molinaro2, Patrick
M. Paranhos2, Pedro M. Cunha2,Ruy L. Milidiú2, Carlos J. P. de
Lucena2 , “A Multi-Agent Architecture for a Dynamic Supply
Chain Management” , American Association for Artificial
Intelligence ,2006.
[7] Ka-Chi Lam , Ran Tao, Mike Chun-Kit Lam “A material
supplier selection model for property developers using Fuzzy
Principal Component Analysis” Automation in Construction 19
(2010) 608–618
[8] Mario Verdicchio and Marco Colombetti ,” Commitments for
Agent-Based Supply Chain Management” , ACM SIGecom
Exchanges, Vol. 3, No. 1, 2002.
[9] Mercedes Amor *, Lidia Fuentes “Malaca: A component and
aspect-oriented agent architecture” Information and Software
Technology 51 (2009) 1052–1065
[10] MihaelaUlieru, Senior Member, IEEE, and MirceaCobzaru ,
“BuildingHolonic Supply Chain Management Systems: An e-
Logistics Application for the Telephone Manufacturing Industry”
IEEE transactions on industrial informatics, vol. 1, no. 1,
February 2005.
[11] Onn Shehory and Arnon Sturm , “Evaluation of Modeling
Techniques for Agent-Based Systems” , AGENTS’01, February
11-13, 2001, Montréal, Quebec, Canada.
[12] Qing Zhang and Wuhan, “Essentials for Information
Coordination in Supply Chain Systems”, Asian Social Science
Vol. 4, No. 10 ,Oct 2008
IJCSI International Journal of Computer Science Issues, Vol. 7, Issue 5, September 2010
ISSN (Online): 1694-0814
www.IJCSI.org
205
[13] Rasoul Karimi, Caro Lucas and Behzad Moshiri ,” New
Multi Attributes Procurement Auction for Agent- Based Supply
Chain Formation” ,IJCSNS International Journal of Computer
Science and Network Security, VOL.7 No.4, April 2007.
[14] Robert de Souza and Heng Poh Khong , “supply Chain
Models in Hard Disk Drive Manufacturing” , lEEE ON
Magnetics. VOL 35. No 1. March 1999
[15] Rui Carvalho, Luís Custódio , “A Multiagent Systems
Approach for Managing Supply-Chain Problems: new tools and
results “ , Inteligencia Artificial V. 9, No 25, 2005.
[16] Vivek kumar , Amit Kumar Goel , Prof. S.Srinivisan, “A
Multiagent Conceptulization For Supply-Chain Management”,
Ubiquitous Computing and Communication Journal, Vol 4, No. 5
, 2009
[17] William E. Walsh and Michael P. Wellman,” Modling
supply chain Formation in Multiagent System” , Artificial
Intelligence, vol 1788: Agent Mediated Electronic Commerce
II,Springer-Verlag, 2000
[18] Yevgeniya Kovalchuk , “Multi-Agent Decision Support
System for Supply Chain Management” 10th Int. Conf. on
Electronic Commerce (ICEC) ’08 Innsbruck, Austria.
[19] Yang Hang and Simon Fong , “Double-agent Architecture
for Collaborative Supply Chain Formation” , Proceedings of
iiWAS2008.
Mr. Vivek Kumar has completed his M.Phil (Computer
Science) in 2009. Apart from this, he did M.Tech. (Computer
Science, 2005) & MIT in 2001. He has 10 years of teaching
experience in various engineering Colleges. Presently he is
working as faculty in Gurgaon Institute of Technology and
Management, Gurgaon, Haryana, India. Under the guidance of
Dr. Srinivasan, he is pursuing Ph.D. from Department of
Computer Science and Engineering, S. Gyan Vihar University,
Jaipur, India
He has published one international & two national (Conference
Proceeding) papers on Supply Chain Management through
Multi-Agent System.
Dr S Srinivasan obtained his M.Sc (1971), M.Phil(1973) and
Ph.D. (1979) from Madurai University . He served as Lecturer
for 7 years in National Institute of Tehnology in the Computer
Applications Department . Later he joined Industry as IT Head
for 18 years . Again he started his teaching career serving as
Professor and Head of the Department of Computer Science,
PDM College of Engineering , Haryana, India. He has published
several papers in Multi-Agent Technology Systems and its
applications . He is member of Computer Society of India.
Attended various national and international seminars and
conferences and presented papers on Artificial Intelligence and
Multi-Agent Technology.
doc_430233897.pdf
Supply chain is a worldwide network of suppliers, factories, warehouses, distribution centers, and retailers through which raw materials are acquired, transformed, and delivered to customers.
IJCSI International Journal of Computer Science Issues, Vol. 7, Issue 5, September 2010
ISSN (Online): 1694-0814
www.IJCSI.org
198
A Review of Supply Chain Management using Multi-Agent System
Abstract
Supply chain consist of various components/ identities like
supplier, manufacturer, factories, warehouses, distributions
agents etc. These identities are involved for supplying raw
materials, components which reassembles in factory to produce a
finished product. With the increasing importance of computer-
based communication technologies, communication networks are
becoming crucial in supply chain management. Given the
objectives of the supply chain: to have the right products in the
right quantities, at the right place, at the right moment and at
minimal cost, supply chain management is situated at the
intersection of different professional sectors. This is particularly
the case in construction, since building needs for its fabrication
the incorporation of a number of industrial products. This paper
focuses on an ongoing development and research activities of
MAS (Multi Agent System) for supply chain management and
provides a review of the main approaches to supply chain
communications as used mainly in manufacturing industries.
KEYWORDS: Information exchanges, Multi Agent System,
knowledge sharing and supply chain management
1. Introduction
Supply chain is a worldwide network of suppliers, factories,
warehouses, distribution centers, and retailers through which raw
materials are acquired, transformed, and delivered to customers.
In recent years, new software architecture for managing the
supply chain at the tactical and operational levels has emerged. It
views the supply chain as composed of a set of intelligent
software agents, each responsible for one or more activities in
the supply chain and each interacting with other agents in the
planning and execution of their responsibilities. Supply Chain
Management is the most effective approach to optimize working
capital levels, streamline accounts receivable processes, and
eliminate excess costs linked to payments.
2. Literature Survey
Analysts estimate that such efforts can improve working Capital
levels, streamline accounts receivable processes, and eliminate
excess costs linked to payments. Analysts estimate that such
efforts can improve working capital levels by 25% [2]. Today,
the best companies in a broad range of industries are
implementing supply chain management solutions to improve
business performance and free cash resources for growth and
innovation. Supply Chain Management is about managing the
physical flow of product and related flows of information from
purchasing through production, distribution and Delivery of the
finished product to the customer. This requires thinking beyond
the established boundaries, strengthening the linkages between
the supply chain functions and finding ways to pull them
together. The result is an organization that provides a better
service at a lower cost. MihaelaUlieru et al. give a approach
based on the holonic enterprise model [10] with the Foundation
for Intelligent Physical Agents (FIPA) Contract Net protocols
applied within different levels of the supply chain. The
negotiation on prices is made possible by the implementation of
an XML rule-based system that is also flexible in terms of
configuration. According to Pericles A., the system is viewed as
an organization or collection of roles that relate to each other and
form an interaction model. Roles in the system are descriptions
of business entities, whose functions are modeled at an abstract
level. Whole system is divided in Business Description, Product
Description, and Order Management Holarchy, Manufacturing
Holarchy.ole Modeling. Author has given the following System
Fig. 1 Class Architecture for customer Agents
JADE Content Classes
Setup(addBehaviour()
)
Request_Behavio
Concept : Concept
Predicates : Predicate
Action : AgentAction
Action()
Content :
ContentManager
Content_Language:
Codec
Customer Agent
Concept
Ontology Codec Content
JADE
Predicate
Agent Action
Vivek Kumar
1
and Dr. S. Srinivasan
2
1
Research Scholar, Department of computer Science & Engineering, Suresh Gyan Vihar, University
Jaipur, Rajasthan 302 004, India
2
Professor & Head, Computer Science Department
PDM Engineering College,
Bhadurgarh, Haryana 124 507, India
IJCSI International Journal of Computer Science Issues, Vol. 7, Issue 5, September 2010
ISSN (Online): 1694-0814
www.IJCSI.org
199
Yevgeniya Kovalchuk presented a way to manage the supply
chain activities & try to automate their business processes [18].
In practice, all the activities are highly connected and
interdependent. The project is mainly focused on the demand
part of the supply chain. In particular, different methods for
predicting customer offer prices that could result in customer
orders are explored and compared in the system. RuiCarvalho et
al. presented multi-agent technology as a sound alternative to
classical optimization techniques that can contribute to solve
hard problems. To prove this point , the MAS with the following
functionalities designed : simulation of an almost infinite number
of agents, heuristics for decision making, possibility to choose
among alternative decision strategies and tactics, different
evaluation criteria and evaluation functions, different message
sequences, and stochastic or deterministic behavior.
Fig. 2 Compartmentalization of operational information coordination
José Alberto R. P. Sardinha1 et al. presented a flexible
architecture based on a distributed multi-agent architecture [15]
of a dynamic supply chain. Intelligent agents tackle sub problems
of a dynamic SCM. Authors present an implementation of this
architecture by using international test bed for SCM solutions.
The mail purpose of this work is to present a multi-agent
architecture for a generic supply chain that uses a direct sales
model, which links customers directly to a manufacturer through
the Internet. Robert de Souza et al. addressed two main issues
[6]: Can we chart the complex logistical dynamics of disk drive
manufacturing? What are the critical success factors that impact
the economics of hard disk drive manufacturing? The backdrop
for this study is the (global) supply chain much in evidence in
disk drive manufacture. Fu-ren Lin et al. analyzed the impact of
various levels of information sharing including order, inventory,
and demand information, which is based on transaction costs
[14]. This study further examines the effects on supply chain
performance in electronic commerce. Specifically, the multi
agent simulation system Swarm is employed to simulate and
analyze the buyer–seller correlation in sharing information
among business partners in supply chains. Information sharing
between firms helps to lower the total cost and increase the order
fulfillment rate. In other words, information sharing may reduce
the demand uncertainty that firms normally encounter. Onn
Shehory et al. discussed suitability of agent modeling techniques
[4] to agent-based systems development. In evaluating existing
modeling techniques, addressed criteria from software
engineering as well as characteristics of agent-based systems.
Evaluation shows that some aspects of modeling techniques for
agent-based systems may benefit from further enhancements.
This technique tries to answer the following questions: (1) which
agent-based system characteristics and software engineering
principles are addressed within AOSE modeling techniques, and
to what extent? (2) What should be the properties of the future
agent-oriented modeling techniques? Rasoul Karimi et al.
developed a new multi attributes procurement auction [11]. It is
new because it has been defined for a special model of supply
chain, in that customer has a new scoring rule, and producers
have new strategies for biding. Multi Attribute Procurement
Auction is a good solution for Supply Chain problem which fits
its requirements. The implementation of the Swarm simulation
system incorporates multiple agents with the decision model to,
in turn; determine the relationship with their trading partners.
Fig. 3 demonstrates a purchasing agent’s decision model to
determine which supplier should issue the purchase order.
Fig. 3 Swarm implementation for modeling supply chains.
A trading partner contains several agents, including order
management, inventory management, policy management,
production, production planning, and purchasing. Among them,
the purchasing agent proposes the decision model to determine
from which supplier products should be purchased. The
purchasing agent buys goods from suppliers that offer the lowest
price. The price issued by a supplier is the outcome of weighing
production cost and coordination cost. The final price is the one
issued by a supplier. Yang Hang et al. proposed a CSET
framework [13] for whole supply chain in collaborative manner
by incorporating it with the Just-in-Time (JIT) principle, known
as CSET. The core of the CSET model is based on intelligent
agent technology. This paper defines such double-agent
mechanisms in details, as well as demonstrating its merits via
simulation study. The MAS proposed here was implemented
using LISP and had as first source of inspiration the agent
creation language named RUBA [Ventura97]. The system has as
main blocks i) an environment agent, in charge of the meaningful
functioning of the system and event execution simulation
[Michael90], ii) client agents, with needs to be satisfied and iii)
firm agents, that have also needs but are capable of product
manufacturing. The system also includes a blackboard, where
agents can post their messages, and a set of communication rules
(a communication protocol inspired on the Contract Net protocol
Supplie Manufacturer Distributor Retailer
Supplie Manufacturer Distributor Retailer
Monolayer information sharing
Monolayer information sharing
Inventory
Management
Cost Management
Agent
Work Flow
Agent
Order
Management
Production Agent Production Planning
Agent
OFP Model Swarm
Statistics Swarm
OFP Observer
Swarm
IJCSI International Journal of Computer Science Issues, Vol. 7, Issue 5, September 2010
ISSN (Online): 1694-0814
www.IJCSI.org
200
[Smith80; Smith81]), common to all agents and that makes
possible message exchange. KQML [Finin94], a standard
message format, and KIF [Genesereth94], a message content
specification, served as the basis for the communication protocol
of our MAS [Huns99].The main elements of the system, agents,
blackboard and a communication protocol, are essential for
functioning. These agents are intelligent, because they are able to
present successful behavior [Albus91]. Figure 4 shows the
system behaviour. Fu-ren Lin et al. used multiagent simulation
system, swarm, for simulating trust mechanism and analyzing the
supply chain performance in four different market environments
[19].Supply chain performance is evaluated by comparing the
order fulfillment process of a mold industry both with and
without trust mechanisms. From the experimental result ,
Fig 4. Environment Agent and Agent Behavior
they found that the trust mechanism reduced the average cycle
time rate and raised the in-time order fulfillment rate as the
premium paying for better quality and shorter cycle time.
Charles M. Macal et al. gave a new approach [5] to modeling
systems comprised of interacting autonomous agents. &
described the foundations of ABMS, identifies ABMS toolkits
and development methods illustrated through a supply chain
example, and provides thoughts on the appropriate contexts for
ABMS versus conventional modeling techniques. William E.
Walsh et al. highlighted some issue that must be understood to
make progress in modeling supply chain formation [3].
Described some difficulties that arise from resource contention.
They suggested that market-based approaches can be effective in
solving them. Mario Verdicchio et al. considered commitment as
a concept [17] that underlies the whole multi-agent environment,
that is, an inter-agent state, react a business relation between two
companies that make themselves represented by software agents.
Michael N. Huhns et al. found after this research that supply
chain problems cost companies [8] between 9 to 20 percent of
their value over a six month period. The problems range from
part shortages to poorly utilized plant capacity. Qing Zhang et al.
provide a review of coordination of operational information in
supply chain [12] . Then the essentials for information
coordination are indicated.Vivek Kumar et al. gave a solution for
the construction, architecture, coordination and designing of
agents. This paper integrates bilateral negotiation, Order
monitoring system and Production Planning and Scheduling
multiagent system. Ali Fuat- Guneri et al gave the concept of
supply chain management process[16], in which the firm select
best supplier , takes the competitive advantage to other
companies. As supplier selection is an important issue and with
the multiple criteria decision making approach, the supplier
selection problem includes both tangible and intangible factors.
The aim of this paper is to present an integrated fuzzy and linear
programming approach to the problem. Firstly, linguistic values
expressed in trapezoidal fuzzy numbers are applied to assess
weights and ratings of supplier selection criteria. Then a
hierarchy multiple model based on fuzzy set theory is expressed
and fuzzy positive and negative ideal solutions are used to find
each supplier’s closeness coefficient. Finally, a linear
programming model based on the coefficients of suppliers,
buyer’s budgeting, suppliers’ quality and capacity constraints is
developed and order quantities assigned to each supplier
according to the linear programming model. Amor et al.
presented Malaca [9], an agent architecture that combines the use
of Component- based Software Engineering and Aspect-Oriented
Software Development.
Fig. 5 Conceptualization of the aspect model in Malaca
Malaca supports the separate (re)use of the domain-specific
functionality of an agent from other communication concerns,
providing explicit support for the design and configuration of
agent architectures and allows the development of agent-based
software so that it is easy to understand, maintain and reuse. Ka-
Chi Lam et al. investigated a selection model based on Fuzzy
Principal Component Analysis (PCA) [7] for solving the material
supplier selection problem from the perspective of property
developers. First, the Triangular Fuzzy Numbers is used to
quantify the decision makers' subjective judgments. Second,
PCA is employed to compress the data of the selection criteria
Warehouse
Client
Agent
Creates
Information
Blackboard
Factory
Agent
-----
------
Problem
Aspect
-role : String
-role Instance : String
+handleMessage( message : Message ) :
Message
+handleInputMessage( message : Message )
+handleOutputMessage( message : Message ) :
Message
<<enumeration>>
Aspect Scope
AGENT SCOPE
PROTOCOL SCOPE
CONVERSATION_SCOPE
Message
Return
Handles
Coordination
Aspect
Distribution
Aspect
Representation Aspect
Component
Message Transport Service
Role
Coordin
ate Role
Encoding Format
Interation
Protocol
Component
Acti
on
IJCSI International Journal of Computer Science Issues, Vol. 7, Issue 5, September 2010
ISSN (Online): 1694-0814
www.IJCSI.org
201
and eliminating the multi-collinearity among them. Third, the
linear combined score of PCA (SCOREPCA) is used to rank the
Suppliers.
Four material purchases are used to validate the proposed
selection model.The results show that the proposed model can be
adopted in construction material supplier selection by the
property developers.
Table 1: Summary
S.No. Title Name &
Authors
Explanation & Conclusion
1. “supply Chain
Models in Hard
Disk Drive
Manufacturing”
Robert de Souza
and Heng Poh
Khong
This paper seeks to address two
main issues: Can we chart the
complex logistical dynamics of
disk drive manufacturing? What
are the critical success factors that
impact the economics of hard disk
drive manufacturing?
The pressures in the disk drive
industry are classic supply chain
economics; value, timing, supply,
demand and technology
development that all play a part
into price erosion patterns. To
address such issues the authors'
postulate that the five chains
interact to give rise to
complexities, static models cannot
easily handle.
2. Modeling supply
chain Formation
in Multiagent
System
William E.
In this paper the authors highlight
some issues that must be
understood to make progress in
molding supply chain formation.
Supply chain formation is an
Walsh and
Michael P.
Wellman
important problem in the
commercial world and can be
improved by greater automated
support. The problem is salient to
the MAS community and
deserving of continued research.
3. Evaluation of
Modeling
Techniques for
Agent-Based
Systems
Onn Shehory
and Arnon
Sturm
Author discusses suitability of
agent modeling techniques to
agent-based systems development.
In evaluating existing modeling
techniques, and address criteria
from software engineering as well
as characteristics of agent-based
systems.
Based on these findings, we
intend in future research, to
address the needs of agent-based
system developers. This should be
done in order to find the required
modeling techniques and
components for building agent-
based systems.
4. Effects of
Information
Sharing on
Supply Chain
Performance in
Electronic
Commerce
Fu-ren Lin,
Sheng-hsiu
Huang, and
Sheng-cheng Lin
Findings indicate that the more
detailed information shared
between firms, the lower the total
cost, the higher and the order
fulfillment rate. And the shorter
the order cycle time. In other
words, information sharing may
reduce the demand uncertainty
that firms normally encounter.
Firms that share information
between trading partners tend to
transact with a reduced suppliers.
This work investigated the buyer–
seller relationship in electronic
commerce with an Extranet as the
platform for sharing information.
Using the Swarm simulation
system, based on transaction
costs, we have identified effects of
sharing various levels of
information between supply chain
partners.
0 0.25
M VG/V
H
G/
H
VB/VL P/L
DMs’ importance wrights
Supplies’ ratings
Uncertain
Unceatin Values
(in 5 point sale
Fuzzy refraction
Fig. 6. Membership functions of DMs' importance weights and
suppliers'ratings (modified from)
IJCSI International Journal of Computer Science Issues, Vol. 7, Issue 5, September 2010
ISSN (Online): 1694-0814
www.IJCSI.org
202
5. Commitments
for Agent-Based
Supply Chain
Management
Mario
Verdicchio and
Marco
Colombetti
As there are several analogies
between a company in a business
network and an agent, the Multi-
Agent System paradigm can be a
valid approach for modeling
supply chain networks. We
consider commitment as a concept
that underlies the whole multi-
agent environment, that is, an
inter-agent state, reacting a
6. Building
Holonic Supply
Chain
Management
Systems: An e-
Logistics
Application for
the Telephone
Manufacturing
Industry
MihaelaUlieru
and
MirceaCobzaru
Approach is based on the holonic
enterprise model with the
Foundation for Intelligent
Physical Agents (FIPA) Contract
Net protocols applied within
different levels of the supply
chain holarchy. To accommodate
differentiation of interests and
provide an allocation of resources
throughout the supply chain
holarchy, we use nested protocols
as interaction mechanisms among
agents. Agents are interacting
through a price system embedded
into specific protocols. The
negotiation on prices is made
possible by the implementation of
an XML rule-based system that is
also flexible in terms of
configuration and can provide
portable data across networks.
As the effectiveness of centralized
command and control in SCM
starts to be questioned, there is a
critical need to organize
supply chain systems in a
decentralized and outsourced
manner. Agent-based models can
easily be distributed across a
network due to their modular
nature. Therefore, the distribution
of decision-making and execution
capabilities to achieve system
decentralization is possible
through models of operation with
communication among them. The
ontology structure of the JADE
framework is, in our opinion, one
of the best designed to address the
issues of accessing and sharing
information pertinent to a specific
application.
7. A Multiagent
Systems
Approach for
Managing
Supply-Chain
Problems: new
It was modelled and implemented
a MAS with the following
functionalities: simulation of an
almost infinite number of agents,
heuristics for decision making,
possibility to choose among
tools and results
Rui Carvalho,
Luís Custódio
alternative decision strategies and
tactics, different evaluation
criteria and evaluation functions,
different message sequences, and
stochastic or deterministic
behavior.
When we applied our MAS to a
problem of SC management at
HP, we obtained results with
stock outs for every product of the
bill of materials. On the contrary,
some authors using mathematical
tools only simulated the stock out
of only one product of the bill of
materials.
8. A Multi-Agent
Architecture for
a Dynamic
Supply Chain
Management
José Alberto R.
P. Sardinha1,
Marco S.
Molinaro2,
Patrick M.
Paranhos2,
Pedro M.
Cunha2,
Ruy L. Milidiú2,
Carlos J. P. de
Lucena2
This paper presents a flexible
architecture for dealing with the
next generation of SCM problems,
based on a distributed multi-agent
architecture of a dynamic supply
chain. We define intelligent agent
roles that tackle sub problems of a
dynamic SCM.
We also present an
implementation of this
architecture used in the
international test bed for SCM
solutions, the Trading Agent
SCM competition, as well as some
experimental results.
A multi-agent design is used in
the architecture, because we
believe it facilitates the
development of modular entities
that are distributed and reusable.
The design was also used to
implement an agent entry for the
Trading Agent Competition. This
system competed against 32
entries, and was able to classify to
the quarter-finals of the 2005
competition.
9. How to Model
With Agents
Proceedings of
the 2006 Winter
Simulation
Conference
Charles M.
Macal and
Michael J. North
Agent-based modeling and
simulation (ABMS) is a new
approach to modeling systems
comprised of interacting
autonomous agents. ABMS
promises to have far-reaching
effects on the way that businesses
use computers to support
decision-making.
Computational advances make
possible a growing number of
agent-based applications across
IJCSI International Journal of Computer Science Issues, Vol. 7, Issue 5, September 2010
ISSN (Online): 1694-0814
www.IJCSI.org
203
many fields. Applications range
from modeling agent behavior in
the stock market and supply
chains
10. New Multi
Attributes
Procurement
Auction for
Agent- Based
Supply Chain
Formation
Rasoul Karimi,
Caro Lucas and
Behzad Moshiri
In this article, this constraint has
been relaxed and a new
procurement auction is defined. In
this auction, seller agents can take
different strategies based on their
risk attribute. These strategies is
analyzed and compared
mathematically.
Authors define a new MAPA
which is usable under the new
model of supply chain. In this
MAPA, the producer could have
two different strategies based on
its risk attribute. These two
strategies are compared
mathematically and also in a
simulation.
11. Multi-Agent
Decision
Support System
for Supply
Chain
Management
Yevgeniya
Kovalchuk
The research approach followed is
presented. The results achieved so
far along with the plans for future
work are given next.
Various techniques for predicting
bidding prices in the context of
dynamic competitive
environments are explored. Apart
from the SCM, the solutions can
be used in forecasting financial
markets and participating in on-
line auctions.
12. Double-agent
Architecture for
Collaborative
Supply Chain
Formation
Yang Hang and
Simon Fong
The model is supported by
double-agent architecture with
each type of agents who makes
provisional plans of order
distribution by Pareto optimality
and JIT coordination respectively
As a result, pipelining
manufacturing flow is achieved.
This is significant to dynamic
supply chain formation as it can
help to optimize constraints and
costs across production,
distribution, inventory, and
transportation.
13, Essentials for
Information
Coordination in
Supply Chain
Provide a review of coordination
of operational information in
supply chain which is classified
into information types, their
Systems
Qing Zhang and
Wuhan
impact on supply chain
performance, and the policy of
information sharing
Multi-agent computational
environments are suitable for
studying classes of coordination
issues involving multiple
autonomous or semi-autonomous
optimizing agents where
knowledge is distributed and
agents communicate through
messages.
14. Effects of Trust
Mechanisms on
Supply Chain
Performance
Using Multi-
agent Simulation
and Analysis
Fu-ren Lin ,Yu-
wei Song and
Yi-peng Lo
The multiagent simulation system
Swarm is employed to simulate
and analyze the buyer–seller
correlation in sharing information
among business partners in supply
chains
The deeper the information
sharing level, the higher in-time
order fulfillment rate and the
shorter order cycle time, as
information sharing may reduce
the demand uncertainty that firms
normally encounter. Finally, firms
that share information between
trading partners tend to transact
with a reduced set of suppliers.
15. A Multiagent
Conceptualizatio
n For Supply-
Chain
Management
Vivek kumar ,
Amit Kumar
Goel , Prof.
S.Srinivisan
Paper present solution for the
construction, architecture,
coordination and designing of
agents. This paper integrates
bilateral negotiation, Order
monitoring system and Production
Planning and Scheduling
multiagent System.
The wide adoption of the Internet
as an open environment and the
increasing popularity of machine
independent programming
languages, such as Java, make the
widespread adoption of multi-
agent technology a feasible goal
16. An integrated
fuzzy-lp
approach for a
supplier
selection
problem in
supply chain
management
Ali Fuat Guneri,
A hierarchy multiple model based
on fuzzy set theory is expressed
and fuzzy positive and negative
ideal solutions are used to find
each supplier’s closeness
coefficient. Finally, a linear
programming model based on the
coefficients of suppliers, buyer’s
budgeting, suppliers’ quality and
capacity constraints is developed
and order quantities assigned to
IJCSI International Journal of Computer Science Issues, Vol. 7, Issue 5, September 2010
ISSN (Online): 1694-0814
www.IJCSI.org
204
Atakan Yucel ,
Gokhan
Ayyildiz
each supplier according to the
linear programming model.
Fuzzy set theory approach helps
to convert decision-makers’
experience to meaningful results
by applying linguistic values to
assess each criterion and
alternative suppliers.
17. Malaca: A
component and
aspect-oriented
agent
architecture”
Information and
Software
Technology
Mercedes Amor
*, Lidia Fuentes
An agent architecture that
combines the use of Component-
based Software Engineering and
Aspect-Oriented Software
Development
Provided explicit support for the
design and configuration of agent
architectures and allows the
development of agent-based
software
18. A material
supplier
selection model
for property
developers using
Fuzzy Principal
Component
Analysis”
Automation in
Construction
Ka-Chi Lam ,
Ran Tao, Mike
Chun-Kit Lam
Tthe Triangular Fuzzy Numbers is
used to quantify the decision
makers' subjective judgments.
Second, PCA is employed to
compress the data of the selection
criteria and eliminating the
multicollinearity among them.
The model can efficiently
eliminate the multicollinearity
among the supplier's attributes
and help to reduce the trade-offs
and repeatability errors in the
selection process.and the proposed
selection model can also reduce
the subjective errors on the sense
that the weight assigned for each ?
is generated automatically.
3. Conclusion
Multi-agent system is a loosely coupled network of
software agents that interact to solve problems that are
beyond the individual capacities or knowledge of each
problem solver. The general goal of MAS is to create
systems that interconnect separately developed agentsThus
enabling the ensemble to function beyond the capabilities
of any singular agent in the set-up in agent model. This
research can demonstrate that agent technology is suitable
to solve communication concerns for a distributed
environment. Multi-agent systems try to solve the entire
problem by collaboration with each other and result in
preferable answer for complex problems. For further
works, it is recommended for developing this model to
have multi retailer and even multi distributor and apply the
auction mechanism between them.
References
[1] Ali Fuat Guneri, Atakan Yucel , Gokhan Ayyildiz “An
integrated fuzzy-lp approach for a supplier selection problem in
supply chain management” Expert Systems with Applications 36
(2009) 9223–9228
[2] C. Iglesias, M. Garijo, J. Centeno-Gonzalez, and V. J. R.,
"Analysis and Design of Multiagent Systems using MAS-
CommonKADS," presented at Agent Theories, Architectures,
and Languages, 1998.
[3] Charles M. Macal and Michael J. North Tutorial on Agent-
Based Modeling And Simulation Part 2: How to Model With
Agents Proceedings of the 2006 Winter Simulation Conference
[4] Fu-ren Lin, Sheng-hsiu Huang, and Sheng-cheng Lin ,
“Effects of Information Sharing on Supply Chain Performance in
Electronic Commerce“ ,IEEE Transactions On Engineering
Management, Vol. 49, No. 3, August 2002.
[5] Fu-ren Lin ,Yu-wei Song and Yi-peng Lo ,”Effects of Trust
Mechanisms on Supply Chain Performance Using Multi-agent
Simulation and Analysis” , Proccding of the First Workshop on
Knowledge Economy and Electronic Commerce.
[6] José Alberto R. P. Sardinha1, Marco S. Molinaro2, Patrick
M. Paranhos2, Pedro M. Cunha2,Ruy L. Milidiú2, Carlos J. P. de
Lucena2 , “A Multi-Agent Architecture for a Dynamic Supply
Chain Management” , American Association for Artificial
Intelligence ,2006.
[7] Ka-Chi Lam , Ran Tao, Mike Chun-Kit Lam “A material
supplier selection model for property developers using Fuzzy
Principal Component Analysis” Automation in Construction 19
(2010) 608–618
[8] Mario Verdicchio and Marco Colombetti ,” Commitments for
Agent-Based Supply Chain Management” , ACM SIGecom
Exchanges, Vol. 3, No. 1, 2002.
[9] Mercedes Amor *, Lidia Fuentes “Malaca: A component and
aspect-oriented agent architecture” Information and Software
Technology 51 (2009) 1052–1065
[10] MihaelaUlieru, Senior Member, IEEE, and MirceaCobzaru ,
“BuildingHolonic Supply Chain Management Systems: An e-
Logistics Application for the Telephone Manufacturing Industry”
IEEE transactions on industrial informatics, vol. 1, no. 1,
February 2005.
[11] Onn Shehory and Arnon Sturm , “Evaluation of Modeling
Techniques for Agent-Based Systems” , AGENTS’01, February
11-13, 2001, Montréal, Quebec, Canada.
[12] Qing Zhang and Wuhan, “Essentials for Information
Coordination in Supply Chain Systems”, Asian Social Science
Vol. 4, No. 10 ,Oct 2008
IJCSI International Journal of Computer Science Issues, Vol. 7, Issue 5, September 2010
ISSN (Online): 1694-0814
www.IJCSI.org
205
[13] Rasoul Karimi, Caro Lucas and Behzad Moshiri ,” New
Multi Attributes Procurement Auction for Agent- Based Supply
Chain Formation” ,IJCSNS International Journal of Computer
Science and Network Security, VOL.7 No.4, April 2007.
[14] Robert de Souza and Heng Poh Khong , “supply Chain
Models in Hard Disk Drive Manufacturing” , lEEE ON
Magnetics. VOL 35. No 1. March 1999
[15] Rui Carvalho, Luís Custódio , “A Multiagent Systems
Approach for Managing Supply-Chain Problems: new tools and
results “ , Inteligencia Artificial V. 9, No 25, 2005.
[16] Vivek kumar , Amit Kumar Goel , Prof. S.Srinivisan, “A
Multiagent Conceptulization For Supply-Chain Management”,
Ubiquitous Computing and Communication Journal, Vol 4, No. 5
, 2009
[17] William E. Walsh and Michael P. Wellman,” Modling
supply chain Formation in Multiagent System” , Artificial
Intelligence, vol 1788: Agent Mediated Electronic Commerce
II,Springer-Verlag, 2000
[18] Yevgeniya Kovalchuk , “Multi-Agent Decision Support
System for Supply Chain Management” 10th Int. Conf. on
Electronic Commerce (ICEC) ’08 Innsbruck, Austria.
[19] Yang Hang and Simon Fong , “Double-agent Architecture
for Collaborative Supply Chain Formation” , Proceedings of
iiWAS2008.
Mr. Vivek Kumar has completed his M.Phil (Computer
Science) in 2009. Apart from this, he did M.Tech. (Computer
Science, 2005) & MIT in 2001. He has 10 years of teaching
experience in various engineering Colleges. Presently he is
working as faculty in Gurgaon Institute of Technology and
Management, Gurgaon, Haryana, India. Under the guidance of
Dr. Srinivasan, he is pursuing Ph.D. from Department of
Computer Science and Engineering, S. Gyan Vihar University,
Jaipur, India
He has published one international & two national (Conference
Proceeding) papers on Supply Chain Management through
Multi-Agent System.
Dr S Srinivasan obtained his M.Sc (1971), M.Phil(1973) and
Ph.D. (1979) from Madurai University . He served as Lecturer
for 7 years in National Institute of Tehnology in the Computer
Applications Department . Later he joined Industry as IT Head
for 18 years . Again he started his teaching career serving as
Professor and Head of the Department of Computer Science,
PDM College of Engineering , Haryana, India. He has published
several papers in Multi-Agent Technology Systems and its
applications . He is member of Computer Society of India.
Attended various national and international seminars and
conferences and presented papers on Artificial Intelligence and
Multi-Agent Technology.
doc_430233897.pdf