||The Container Ship Scheduling Problem: A Praqctical Approach
||Finnbogason, Ivar Freyr
||Larsen, Allan (Logistics & ITS, Department of Transport, Technical University of Denmark, DTU, DK-2800 Kgs. Lyngby, Denmark)
||Technical University of Denmark, DTU, DK-2800 Kgs. Lyngby, Denmark
||A transport system of a shipping company is composed of a set of different rotations that are served by its fleet of vessels. The transport market is volatile highly competitive forcing the companies to constantly adjust and improve their systems to better serve their markets.
This thesis aims to find a solution method that can be useful for the planners of shipping companies when planning complex systems of rotations.
In the first part several different solutions approaches are explored with out success.
The final approach is based on set-partitioning and is in line with the procedures currently used by the planners of Eimskip and Containerships. The solution approach is considerably more work intensive then what was hoped to achieve but at the same time is likely to be beneficial for experienced planners. Using this method enables them to explore the effectiveness of their ideas, to make adjustments of the system and to find new rotations that will improve the quality of the over all system.
The idea is based on designing a number of feasible rotations. Then using set-partitioning to select the best combination and time start dates for each rotation so that all constraints are met. The selected set of rotations is finally checked for capacities. At this point this method differs from most other solution approaches as the flow of containers is not directly considered except in the constraints of the set-partitioning model and then finally at the end of the process to verify its feasibility. This method requires considerable work to be done by the planer but at the same time is able to deliver solutions and gives feedback on possible improvements to the system.
When it was time to start thinking about a thesis project I was clear on only one thing. I wanted to tackle a real life project. With out the help of the people that work and live in the real world this would not be possible. I there for want to start by thanking the grate people at Eimskip PRO for their grate patients and understanding. Johann and Ingvar, thank you for answering my questions and giving my free access to your staff and your information systems. To the two Birgirs, Leo and Davíð thanks for all the inside information and data. I met the same good will with the staff of Containerships. There Andris Velninis supplied my with all the data asked for and answered a wealth of questions. Thanks for that!
The thesis is the final opportunity to learn while still supported by the professors and lectures of the university. Thanks for your help Allan Larsen and Michael Berliner Pedersen, and not the least for your patience with my some times hard to understand e-mails and formulations.
The Glaucus team, Allan Olsen and Anders N. Nielsen deserves my gratitude for running my data all too many times and informing me on the workings of their interesting software, Glaucus.
Last but not least I would like to thank Haraldur Guðmundsson for proofreading the text and giving great comments on the setup of this thesis.
||Technical University of Denmark (DTU) : Kgs. Lyngby, Denmark
Creation date: 2010-12-30
Update date: 2011-04-13