A Multi-Agent System for Production Networks Simulation: Toward a Pheromone-Based Model for Agents’ Coordination

Abstract

In this chapter, the authors present a multi-agent model aimed to investigate emergent organizational structures in production networks and their reification by means of pheromone-based algorithms. The model considers agents (firms) embedded in a production network, interacting among them through business-to-business relations. The evolution of the network structure is endogenous, as it takes into account the individual behavior of the firms and their interactions. The firms are adaptive agents taking investment decisions according to their business efficiency. They adapt their prices to be competitive and get a larger share of the market. Also, they adapt their business relations with their suppliers in order to reduce costs of inputs and get orders satisfied. The agent’s proactivity, with very simple decision mechanisms at the micro level, leads to the emergence of meta-stable business clusters and supply chains at the macro level. Pheromone-based algorithms reify dynamically these clusters as explicit graphs. The results of the authors’ simulations show the impact of the transportation cost and the geographical reach on the regionalization of production and on wealth patterns. Individual firms, with local B2B interactions and decisions, form stable production systems based on the supply/demand and market growth mechanisms leading to the maturation of the market.