Méthode de modélisation-simulation
Using AnyLogic and agent based approach to model consumer market Maxim Garifullin, Andrei Borshchev, Timofei Popkov. EUROSIM 2007, September 9-13, Ljubljana, Slovenia
During the last decade numerous developments have resulted in significant methodological progress in consumer market modeling. While traditional statistical and equation-based methods continue to be useful, new approaches, such as Agent Based Modeling (ABM) successfully address their well known limitations, e.g. the necessity of “perfect mix” assumption for aggregated categories of consumers . In the highly dynamic, competitive and complex market environments (telecom, insurance, leasing, health, etc) the consumer’s choice essentially depends on a number of individual characteristics, inherent dynamics of the consumer, network of contacts, external influences that may be best captured within the Agent Based modeling paradigm, especially taken the high availability of individual-centric data from the CRM (Customer Relationships Management) systems. The leading market analysts worldwide are using ABM to gain deeper insight in the market dynamics and to elaborate optimal strategies for their companies. Although there are no universal straightforward instructions for building Agent Based models, there are certain common steps and patterns. The goal of this paper is to introduce the patterns in consumer market modeling that are most frequently met in our consulting practice. In the final section we also discuss the limitations of ABM.