Agent-based simulation tool for evaluating pooled queue performance at marine container terminals

Abstract

Truck queuing at marine container terminal gates is one of the main sources of delay at terminals and is an area of concern because delays increase the logistical cost of transporting a container. Previous studies on terminal gates focused on the performance of strategies such as the appointment system and extended gate hours. However, no study has yet evaluated the performance of pooling trucks into a single queue at the gates. Previous studies on pooling offered mixed opinions on whether pooling was beneficial, but none of those studies attempted to model the movements of the entities in the queue. In a human system (no vehicles), the movements are not as important because the time to move up one space in the queue is negligible; however, because of the size and weight of the trucks at the gates, the time to move is significant and should be considered. A study was conducted with agent-based simulation to model the terminal gate system with two queuing strategies, a pooled queue and nonpooled queues, because analytical solutions could not capture vehicle movements within the queue. In the study, a car-following model was used to capture a realistic representation of how vehicles move within the queue. The developed simulation model was used to evaluate queuing strategies in various conditions. Results indicate that the use of a pooled queue yields significantly lower average queuing times and lower variability in queuing times.