Multi-agent simulation of virtual consumer populations in a competitive market

Theoretical concepts dealing with consumer behaviors are issued from works led in various research areas: marketing, psychology, sociology and economics. This paper presents a multi-agent simulation of virtual consumer populations. Existing consumer behavioral models are not well suited for the realization of such market simulations including a large consumer population. In this work, a consumer behavioral model based on the new concept of behavioral attitude solving this problem is introduced. It proposes to integrate and organize most of the fundamental notions elaborated within the aforementioned research areas. This model uses a set of basic behavioral primitives such as imitation, conditioning and innovativeness. This paper shows the correlation and the difference of our modeling approach with the actual marketing studies and describes the realized CUstomer BEhavior Simulator (CUBES) software based on this model.