Today's supply chains span across continents, involve numerous entities with different dynamics, and contend with various uncertainties. This paper presents an agent-based model for decision support in a multi-site lube additive manufacturing enterprise. The supply chain comprises raw material suppliers, the lube additive enterprise, and customers. The enterprise consists of a central sales department and a number of production sites at different locations. Each production site has its own commercial, scheduling, procurement, storage, operations, and packaging departments. Supply chain operation involves all these entities in three cycles of activities: enterprise-level coordination, plant operation, and inventory management. Each entity is modeled as an agent, with its own goals and tasks, implemented in Jadex following the belief-desire-intention (BDI) formalism. The model allows the user to simulate and analyze different supply chain policies, configurations, parameters, and scenarios. Its capability for decision support is illustrated through two case studies. © 2009 Springer Berlin Heidelberg.