Proceedings of the 2009 Winter Simulation Conference M. D. Rossetti, R. R. Hill, B. Johansson, A. Dunkin and R. G. Ingalls, eds. A SIMULATION MODEL OF PORT OPERATIONS DURING CRISIS CONDITIONS Tugce G. Martagan Burak Eksioglu Sandra D. Eksioglu Allen G. Greenwood Department of Industrial and Systems Engineering 260 McCain Engineering Building, Box 9542 Mississippi State University Mississippi State, MS 39762, USA ABSTRACT We consider the supply chain for containerized items that arrive at a port in the U.S. whose final destination is also in the U.S. Ports are important entities in global supply chains. As such, when a port cannot operate because of a crisis, such as a natural or man-made disaster, it is critical that freight flow is not disrupted. We develop a simulation model that can be used to make effective re-routing decisions so that the time for freight to reach its final destination is not significantly increased in a crisis. The simulation model will evaluate and report the performance of the supply chain under different re-routing strate- gies. The output can be analyzed to find the best re-routing strategy that minimizes congestion and delays during crisis condi- tions. The model can also be used by various decision makers such as port managers, ocean carriers, or transportation com- panies for strategic decision making. 1 INTRODUCTION A wide variety of industries rely on efficient port operations to receive the raw materials for their businesses as well as to ship their products to their customers. Natural catastrophes (e.g. earthquakes, hurricanes) and man-made disasters (e.g. terror- ist attacks, fires) negatively impact these industries due to delays in the flow of materials through the affected port in a supply chain. Impacts of crisis conditions, such as congestions and increase in lead times, should be assessed to mitigate their nega- tive effects on the performance of supply chains. The fact that most of the supply chains are now tightly connected networks all around the world, intensifies the global impacts of threats from natural disasters, terrorist attacks, wars, worker strikes, etc. However, recent studies show that the majority of supply chains are incapable of dealing with crisis conditions, and have low level of disaster preparedness (Lee 2004, Hale and Moberg 2005). Supply chain risk management is still an issue which is in infancy (Juttner 2005) as most of the U.S. companies ignore the importance of drawing up emergency plans for crisis condi- tions (Lee 2004). Accordingly, the gap between theory and practice in disaster planning for supply chains is highlighted by Tang (2006). In this study, we describe a simulation model that can be used to effectively control freight transportation in order to mi- nimize supply chain disruptions during crisis conditions. The simulation model can be used to evaluate the performance of supply chains that include ports as part of the chain. In our supply chain setting, freight going to a crisis stricken port is re- routed to other ports. The objective is to minimize congestion and the increase in lead times during crisis conditions. The main performance measure is the lead time, which is defined as the total time the freight spends in the system from its origin to its final destination. To demonstrate the use of the model we simulated and evaluated the performance of several ports in the U.S. based on the following cases: (i) under normal conditions without any disruptions and (ii) under crisis conditions where one or more ports are affected by a disaster. The simulation model also enables the decision maker to perform what-if analyses by specify- ing different re-routing scenarios. Although one would expect a significant increase in lead time when there is a crisis at a port, it is not clear if there are significant differences among various re-routing strategies. Statistical analyses are conducted in order to evaluate whether or not there are significant differences in the lead time under normal and crisis conditions, and also among various re-routing strategies. The difference in lead time under several scenarios is estimated in Section 3. The simulation model is flexible and user-friendly, and is developed using ProModel. The model is intended for use by ocean carriers, logistic companies, port operators, and federal emergency management agencies. Using the model requires no 2832 978-1-4244-5771-7/09/$26.00 ©2009 IEEE