Key Pages


|
OUTLOOK |

Biology and Biotechnology
|
Chemistry |
Complexity |
Computer Science |
Ecology and Earth Sciences |
Economics |
Energy |
Materials |
Mathematics |
Nanotechnology |
Social Sciences |
Space Sciences |
Physics |
Psychology |
Geography of Science |
Structure of Science |

|
THE PROJECT |

Project Description
|
Project Team |

|
INSTITUTIONS |

Horizon Scanning Centre
|
Institute for the Future |

|


Category:
Science and Technology
Domain:
Keywords:
agents, computer science, automation, logistics, globalisation
Outlook:
The application of artificial intelligence to commerce may make trade and logistics more efficient.
Summary Analysis:
Automated trading is already commonplace in financial markets. Within 20 years, a significant proportion of day-to-day transactions may involve negotiation between artificially intelligent software trading agents. Over this period, more and more of the minute-to-minute/second-to-second decisions that enable liquidity in financial markets and mobility in logistics systems will be delegated to AI software agents. As global markets and logistics systems grow more complex, AI may be applied extensively, first by producers and later by consumers, to automate parts of the decision making process that currently require humans.

Initially, these systems will be developed and introduced by firms that seek to gain a competitive advantage over human opponents. For example, recent research has shown that well-designed agents can consistently beat humans in many kinds of markets. Human traders may disappear altogether from many global markets.

Artificial intelligence is already widely used in the largest and most complex global logistics systems – such as those operated by Tankers International and UPS. Over the next two decades, these technologies will increasingly be available to small and medium-sized enterprises. Conceivably, as these two systems – artificially intelligent trade and logistics – develop in tandem, they could fuse into a single global system of trade and commerce run by software and machines, watched over by human supervisors.

The trend towards automation of transactions will progress most rapidly in sectors that already exhibit a high degree of standardization. For instance, transportation of all types (freight, passengers, data) will rapidly employ AI to maximize network efficiency. However, the application of AI will not be appropriate in all markets, especially those that require extensive communication between the sales force and the customer.

Implications:

  • Improved efficiency and productivity in the realm of global trade and commerce

Early Indicators:

  • Organisation in 2002 of Trading Agent Competition, a robust international scientific competition in trading agent software
  • Growing use of AI in financial trading, retail e-commerce, and global logistics and supply chain management
  • DayJet's plans to use agent-based models to provide demand-sensitive pricing and flight and crew scheduling for air charter service

What to Watch:

  • Affluent members of technologically advanced societies begin to use personal shopping agents.

Parallels/Precedents:

  • Rise of just-in-time manufacturing in the 1980s

Enablers/drivers:

  • Further development of artificial intelligence
  • Expansion of grid computing
  • Further development of complex economic modeling
  • Globalisation
  • Deregulation of financial markets

Leaders:
Regions:

  • US, Asia (Japan and Korea), EU

Institutions:

  • Swedish Institute of Computer Science
  • CMU
  • University of Michigan
  • University of Southampton (work by Nick Jennings)
  • HP Bristol (work by David Cliff)
  • Cardiff University, Innovative Manufacturing Research Centre [link]
  • Cranfield University, Supply Chain Research Centre [link]
  • University of Southampton, Intelligence, Agents, Multimedia Group [link]
  • Imperial College, London, Complex System Modelling Group [link]
  • AgentLink III (European Commission action) [link]
  • Chartered Institute of Logistics and Transport [link]

Figures:
Sources:

  • Captain Eric Wolf, et al. 2003. "Using agents to model logistics." [link]
  • Ian Foster, et al. 2004. "Brains and brawn: Why grid and agents need each other." Keynote presented at Third International Joint conference on Autonomous Agents and Multi-Agent Systems: New York.
  • Lee, Hau. 2004. "Simple Theories For Complex Logistics." Optimization Magazine, July. [link]
  • Norman M. Sadeh, David W. Hildum, and Dag Kjenstad. 2003. "Agent-based e-Supply Chain Decision Support." Journal of Organizational Computing and Electronic Commerce 13, no. 3.
  • Trautvetter, Chad. "DayJet reveals first plans for air-limo operations." Aviation International News. 3 May 2005. [link]
  • "Robo-traders" Economist. 28 Nov 2002. [link]
  • "Trading Agent Competition" Trading Agent Competition [link]
  • Kowalczyk, Ryszard, Mihaela Ulieru, and Rainer Unland. 2002. "Integrating Mobile and Intelligent Agents in Advanced E-Commerce: A Survey." [link]
  • "Trading agents-the new world of artificial intelligence." The Dolphin (November 2001). [link]
  • "Automated Discussions on Internet Agent-Based Markets: Discussion" [link]
  • "Agent-Based Market Space" Agent-Based Market Space [link]
  • "Trading Agent Competition" Trading Agent Competition [link]
  • "AAMAS" AAMAS [link]
  • Pigaty, Leo and James C. Workman. 2004. "Testing the Survivability of Logistics Information Systems." Army Logistician [link]
  • "Trading Agent Competition" Trading Agent Competition - Research [link]
  • Petrie, Charles J. 1996 "Agent-Based Engineering, the Web, and Intelligence." IEEE Expert. [link]
  • Holguín-Veras, José. 2001. "Towards a 21 st Century agenda of freight transportation" [link]
  • Michael Luck, 50 Facts About Agent-Based Computing [link]
  • Anna Nagurney, ed, Innovations in Financial and Economic Networks, Edward Elgar, 2003, ISBN 1-84376-415-6
  • US Department of Defense Office of Force Transformation, Operational Sense and Respond Logistics: Coevolution of an adaptive enterprise capability, 2004 [link]
  • Paul Davidsson et al, Agent-Based Approaches to Transport Logistics [link]
  • Proshun Sinha-Ray et al, Container World: Global Agent-Based Modelling of the Container Transport Business [link]


At A Glance:
When:
11–20 years
Where:
Global
How Fast:
Years
Likelihood:
High
Impact:
Medium
Controversy:
Low


Related Outlooks:

About this outlook: An outlook is an internally consistent, plausible view of the future based on the best expertise available. It is not a prediction of the future. The AT-A-GLANCE ratings suggest the scope, scale, and uncertainty associated with this outlook. Each outlook is also a working document, with contributors adding comments and edits to improve the forecast over time. Please see the revision history for earlier versions.


Posted at Dec 22/2006 02:38AM:
Simon Thompson

Hi - just worth pointing out that Dave Cliffe has moved to Southampton.



Forum Home  -  Site Home  -  Find Pages: