| In the rapid development of globalization, China has become a world factory, exporting more and more goods to the world while importing large volume of raw materials. A large part of those goods are transported by container ships. Container terminals are thus of extreme importance in streamlining global trade. Many different factors affect the service efficiency in container terminals. This dissertation focuses on vehicle dispatching problems, which decides, for each internal vehicle in ports, where to find a new task after finishing the previous one. Our works include:1. A modified mathematical formulation of the problem is established.Based on extensive investigation of several ports and previous researches by others, we establish a modified mathematical formulation of the vehicle dispatching problem, which can reflect both crane operation delay and multi-ship scenarios.2. A vehicle-initiated multi-factor online dispatching strategy is proposed.To help making dispatching decisions, the online strategy employs an evaluation function. Proportions of each factors in the evaluation functions is determined by a method using discrete event simulation and evolutionary searching algorithm combined.3. Extensive simulation and theoretical analysis is carried out to evaluate the effect and stability of the strategy.A series of simulation experiments is done to evaluation the performance of the strategy. By comparing the result with that of dedicated vehicle dispatching method and with the lower-bound of the best result in theory, we prove our method delivers substantial efficiency improvement in terms of reducing ship serving time and vehicle empty traveling in most cases. This systematic way of evaluation can also be used as a common platform to verify other dispatching strategy.4. Simulation is also used in determining the number of working vehicles needed in a port, so that current tasks can be completed in time delay with justified cost. |
PDF Full Text Request