Information Technology Reference
In-Depth Information
Socrates: A Production-Driven SCM Agent
Carlos R. Jaimez Gonzalez and Maria Fasli
University of Essex, Department of Computer Science
Wivenhoe Park, Colchester CO4 3SQ, UK
{ crjaim, mfasli } @essex.ac.uk
Abstract. The Trading Agent Competition (TAC) is an open-invitation forum
designed to encourage research into electronic markets and trading agents. In this
paper we present the Socrates trading agent and the strategies that were developed
for and used in the TAC Supply Chain Management game as part of the 2004
competition. The resulting behaviour and performance in the TAC competition as
well as in a series of controlled experiments are discussed.
1
Introduction
In today's highly interconnected and networked world more and more businesses and
organizations choose to do business online. This is a dynamic environment where man-
ufacturers may negotiate with suppliers on the one hand, while at the same time compete
for customer orders and have to arrange their production schedule and delivery so that
orders are delivered on time. The ability to respond to changes as they happen and adapt
to variations in customer demand and the restrictions as imposed by procurement, is of
paramount importance. This is the kind of environment that agent technology is best
suited for: dynamic, constrained and real-time. However, to be able to build agents that
offer solutions to problems such as supply chain management, we need to have a very
good understanding of the domain itself and the problems that arise in it. In particu-
lar, firstly we need to gain a better understanding of the problems that arise in supply
chain negotiation situations. Secondly, we need to explore strategies for coping in dy-
namic and competitive environments, and finally develop agent-based systems for an
automated supply chain process. Trial and error in a real environment carries very high
risks. To demonstrate the potential of applying agent technology in complex domains
like supply chain management, realistic testbeds are required that allow researchers
and practitioners to test out and evaluate ideas and techniques. The TAC Supply Chain
Management (SCM) game was designed to capture many of the dynamics of such an
environment and provides an ideal forum for researchers to test, evaluate and learn.
This paper presents the Socrates trading agent and the strategies that were used in
the TAC Supply Chain Management game in the 2004 competition. As a good strategy
for obtaining components was essential for a good overall performance in the game, we
decided to concentrate on this aspect of the agent. Socrates is a production-driven agent
that attempts to keep factory utilisation to the maximum for as long as possible during
the game. We present the problems that arise in dealing with the suppliers and the strate-
gies that were developed to tackle them. The rest of the paper is organized as follows.
The following section presents the TAC SCM game and a section describing related
Search WWH ::




Custom Search