| Summary |
Jen-Chi Yeh, 2006.06
Graduate Institute of Transportation Science, Tamkang University
Most of the import and export freights are transported by sea transportation in Taiwan. Berth Allocation Problem (BAP) is essential problem for the operation of the port authority. The time and location assignment of the berths to vessels is cricial to the efficiency of the port operation. In the current practice in Taiwan, this decision is made by port authority and shipowner in the daily berth allocation meeting, which may not be the best policy to be adopted. A series of academic researchs have devoted on the optimal BAP model for this problem. However, there is no model address this problem from the point of views of both port authority and shipowners, which is the focus of the model proposed in this study.
In this research, we attempt to explore the application of the VRP (vehicle routing problem) concept on the BAP. With the focus of the interests of the port authority and shipowner, a multi-objective programming model is formulated in this study. There are two objectives in this model, i.e., the minimization of the total time in port of the shipowners, and the minimization of the working cost of the port authority.
With the NP-Hard nautrre of the porposed model, it is harder to get the optimal solution when the size of problem increasing. Therefore, the artificial immune algorithm (IA) and Tabu Search algorithm (TA) are proposed as the main solution procedure of the study for its diversity search ability. In order to reduce subject judgements in the multiple- objective programming, a series of fuzzy functions for the objectives are developed for this study. Finally, to evaluate the propsed model and the algorithms developed in the study, a series of case studies include one data from Keelung harbor are tested in this study. The results of these numerical tests indicate that the IA performs better in large size problem. The solution of the proposed procedure provides better result than the conclusion of the daily berth meeting approach. |
