A multi-agent approach to supply chain management in the chemical industry

Supply chains in the chemical industry typically span across several corporations and involve several departments within each. The functioning and characteristics of the entities and the intra-enterprise and inter-enterprise interactions have to be explicitly considered in decision-making. In this chapter, we describe an agent-based model of a refinery's supply chain. Software agents emulate the entities such as procurement, sales, operations, storage and logistics departments of the refinery as well as the suppliers, logistics service providers, and oil-exchanges. These agents model the embedded business policies and thus mimic the different business processes of the enterprise. Uncertainties are captured by stochastic elements embedded in the agents. The dynamics of the supply chain is emulated by discrete event simulation of the agent-based model. The application of the supply chain model and simulation in decision-making is illustrated here. Different business processes and supply chain configurations are evaluated based on their effect on entity-specific as well as supply chain wide key performance indicators. This enables well-rounded decisions related to both the structure and parameters of the supply chain.