Study On The Planning And Scheduling Of Resources For Loading Containers In Container Terminals

Container terminals play the role of hub in the network of seaborne transporta-tion. With the increasing of transportation volume, container terminals are becoming abottleneck of the network of seaborne transportation. As an important activity occur-ring in terminals, loading containers onto vessels a?ects the turnaround time of vesselsand the competition position of terminals. Therefore, improving the e?ciency ande?ectiveness of loading activity is an important issue in container terminals.Loading activity covers the process from receiving outbound containers for tem-poral storage in terminal yards to loading containers onto their designated vessel-slots.The activity is decomposed into several inter-dependent decisions, and the followingfour decisions are in depth studied in this paper: berth and quay crane allocation, yardlocation assignment, loading sequence, and block and yard crane allocation.Based on a mathematical model for the allocation of berths and quay cranes, thispaper takes into consideration the restriction of the coverage ranges of quay cranes andallows for limited adjustments of the allocated quay cranes during loading and dis-charging. A mixed integer programming model is constructed for the problem. Insteadof allocating the two resources sequentially, this paper allocates them simultaneouslyby using a sub-gradient-based Lagrangian relaxation method. Using actual data, theperformance of the algorithm is demonstrated.This paper corrects an error in a yard-location assignment model, and moreover,this paper relaxes an assumption imposed for the original model, i.e., no more thanone relocation movement for an arriving container, and proposes an improved model tofit a more realistic situation. The numerical experiments show that the total expectednumber of relocation movements has been underestimated by the original model andthat the solution quality has been improved by the the new model.In contrast to the literature treating the loading sequence problem as a static prob-lem, this paper considers the dynamic impact of container rehandling on the setup time of yard cranes for the retrieval of containers, and treats the problem as a dynamic prob-lem. A GRASP-based (Greedy randomized adaptive search procedure) algorithm isdesigned to solve this problem.This paper puts the four basic yard activities into a single model rather than inmultiple independent models as was generally done in literature, and studies their de-mand for the allocation of the binding resources: blocks and rail mounted gantry cranes(RMGC). A mixed integer programming model is constructed, and an iteration-baseddecomposition method is proposed for the problem. Based on the constructed modeland the proposed solution method, a decision support system is developed and imple-mented for a terminal in Tianjin seaport. The experiment results show that the decisionsupport system is a good platform for making plans, and could reduce the planningtime greatly and create economic benefits for the company.