Experiments on simulation models were conducted to determine appropriate risk mitigation strategies for a supply chain network under supplier risks. To model the supply chain network, an agent-based simulation model was designed and implemented using the Unified Modeling Language (UML) standard and Java Agent DEvelopment (JADE) platform respectively. In order to model the risks in the simulation models, unexpected events were randomly generated to mimic the risks that possibly occur in the supply chain. Particularly, four types of risks were considered, including rare and short, rare but long, frequent but short, and frequent and long risks. In addition, two risk mitigation strategies, Le, having a redundant supplier and reserving more inventory, were applied and compared. By considering average total operating cost and average product shortage, results are generated for appropriate mitigation strategy for each type of risk.