2002

Conférence

AAMAS '02: Proceedings of the first international joint conference on Autonomous agents and multiagent systems: part 1 Pages 184 - 190

Our goal is to create a virtual consumer population that can be used for simulating the effects of marketing strategies in a competing market context. That requires having a consumers' behavioral model allowing the representation of observed individual behaviors and the simulation of a large population of consumers. That also requires finding the parameters' values characterizing the virtual population that reproduces real market evolutions. This paper proposes a consumer behavioral model based on a set of behavioral primitives such as imitation, conditioning and innovativeness, which are founded on the new concept of behavioral attitude. It shows that this model provides an interpretation of the main concepts and cognitive features, issued from marketing research and psycho-sociology works on consumption. The paper presents also the CUstomer BEhavior Simulator (CUBES), which has been realized for implementing the customer model and leading multi-agents simulations. It shows how genetic algorithms (GA), in addition to multi-agent systems, are used to fit the characteristics of the virtual consumers' population into a global realistic market behavior.

@inproceedings{said2002agent,
  title={Agent-based interaction analysis of consumer behavior},
  author={Said, Lamjed Ben and Bouron, Thierry and Drogoul, Alexis},
  booktitle={Proceedings of the first international joint conference on Autonomous agents and multiagent systems: part 1},
  pages={184--190},
  year={2002}
}