Carnegie Mellon University's CMieux team, led by Norman Sadeh, professor, Institute for Software Research, and graduate student Michael Benisch, won this year's Procurement Challenge, a tournament in which teams develop autonomous trading agents to procure PC components. The finals were July 14 at the AAAI 2008 conference in Chicago.
The challenge is part of the Trading Agent Competition, an annual event supported by SAP and CapGemini that promotes high quality research into trading agent technologies and features several trading scenarios. A Supply Chain Trading scenario was introduced in 2003 and has attracted a total of nearly 150 entries from 60 different teams.
Since 2007, the Supply Chain Trading competition has included the Procurement Challenge, which requires trading agents to manage procurement risk through the negotiation of long-term and one-off procurement contracts for computer components. Agents compete against one another in multiple component markets where component prices and availability fluctuate over a period of 100 days, with each day being simulated in 10 seconds.
"With product life cycles becoming shorter and supply and demand conditions creating market fluctuations that are increasingly difficult to predict, the supply chain trading competition addresses scenarios that are particularly critical in today's global economy," Sadeh said.
The CMieux team consisted of Perry Chu, a sophomore in the Computer Science Department and Lowell Labaro, a master's student in the Information Networking Institute. The CMieux team also finished 3rd in the Supply Chain Trading competition baseline game behind the University of Michigan and the University of Texas at Austin.
For more information
The Supply Chain Trading Competition:
The Procurement Challenge:
CMieux supply chain trading agent: