A Stochastic Model for Interterminal Container Transportation

We propose a novel semiopen queuing network (SOQN) model for the interterminal transportation (ITT) problem where multiple container terminals use a common fleet of vehicles (automated lift vehicles, automated guided vehicles, multitrailers, and barges) to transport containers between terminals. To solve the overall queuing network, our solution approach uses a network decomposition method where the original SOQN is decomposed to a closed and an open queuing network (with bulk-service capacity). To our knowledge, this is the first work that considers bulk service in SOQNs. We develop theoretical upper and lower bounds on the throughput time estimates of our model, and provide an extension for the case when service times at the terminal handling stations depend on the number of containers being loaded/unloaded. We numerically validate our model using simulated data where we find that our model results in errors of less than 5% for vehicle utilization. We also show that our model results in better estimates for the ITT problem when compared to existing approaches in the literature. Finally, we apply our model to real-world data from the Port of Rotterdam and show that it can be used to analyze throughput time trade-offs with alternate dwell point policies, different vehicle types, and variable vehicle capacities.