Dynamic Space and Time Partitioning for Yard Crane Workload Management in Container Terminals

We propose a new hierarchical scheme for yard crane (YC) workload management in container terminals. We also propose time partitioning algorithm and space partition algorithm for deploying YCs to handle the changing job arrival patterns in a row of yard blocks. The main differences between our approach and most of the methods in literature are: (1) Average vehicle job waiting time instead of the number of jobs is used to balance YC workload and to evaluate the quality of a partition; (2) YC working zone assignment is not in units of yard blocks and our space partition algorithm generates more flexible divisions of the workload from all blocks; (3) YC deployment frequency is not fixed but is decided by our time partition algorithm with the objective of minimizing average vehicle waiting times. The scheme combines simulation and optimization to achieve our objective for a row of yard blocks. Experimental results show that the proposed binary partition algorithm TP2 makes substantial improvements in job waiting times over the basic partition scheme and another existing algorithm (Ng, 2005) in all tested job arrival scenarios.