Do we always need to interview human beings in the flesh or can we get reliable information by observing thousands of I-individuals who are pre-programmed to follow the same utility function as their human counterparts? Where is the frontier? Agent-based modeling is a form of computational simulation where artificial agents are given a set of rules and then interact freely between each-others. Agents can have different roles; they can represent consumers, companies, governments and, generally speaking, will act according to their own preferences or own rules of actions. In this paper the authors demonstrate how agent-based modeling can be used to simulate the future demand for air travel on a global scale under a variety of scenarios.
Tipping points, viral marketing, the butterfly effect - Marketers today want to understand and take advantage of these 'emergent' phenomena. Outside of marketing research, innovative social and computer scientists have developed a new approach to understanding complex 'non-linear' phenomena. This approach is known as agent-based simulation.This presentation introduces agent-based models and explains the way they work, with examples of applications for marketing and marketing research.