Behavior of an evolving business network in a social network

Abstract

We investigated the recruitment dynamics of multilevel marketing (MLM) as it evolves under a real social network. Agent-based modeling was used to simulate the business network under three general types of social networks: ordered, small-world, and random. The recruiter (σ₊₁), the disillusioned (σ₋₁), and the recruitable (σ₀) are the three possible states of each individual. The randomness p of the social network characterizes the randomness in the linkages, which represents 'friendship' between agents. The tolerance field (FT) characterizes the agent's 'patience' with respect to the status quo. It was shown that as p approaches 1, the maximum number of agents, N_Rmax, involved in the system is greater and achieved at a lesser time, ∆t. FT was also determined to be a good indicator of the general behavior of the entire business system. For the tolerance field of the recruiter greater than both the recruitable and the disillusioned, the resulting business network has a higher probability to stabilize.