As a main international transportation hub in the center of the area with the world's fastest growing economy, Hong Kong has the world's largest container terminals and the world's largest air-cargo terminals. The throughputs of containers and international air-cargoes are both ranked top in the world in 1999. Relative to the freight processing capacities of container terminals and air-cargo terminals, the land transportation of containers and air-cargoes has been the bottleneck. Characteristics of freight transportation in Hong Kong complicate the transportation activities and make routing vehicle fleets difficult. This thesis studies two types of routing problems: the vehicle routing problem in air-cargo transportation and the driver-task assignment problem in short-haul in-land conta...[ Read more ]

As a main international transportation hub in the center of the area with the world's fastest growing economy, Hong Kong has the world's largest container terminals and the world's largest air-cargo terminals. The throughputs of containers and international air-cargoes are both ranked top in the world in 1999. Relative to the freight processing capacities of container terminals and air-cargo terminals, the land transportation of containers and air-cargoes has been the bottleneck. Characteristics of freight transportation in Hong Kong complicate the transportation activities and make routing vehicle fleets difficult. This thesis studies two types of routing problems: the vehicle routing problem in air-cargo transportation and the driver-task assignment problem in short-haul in-land container transportation. The main goal of the thesis is to model the problems and develop solution procedures.

First, we provide the background and the motivation of our research. Second, we study the vehicle routing problem with backhauls and time windows. Our problem extends that in the existing literature by involving complex operation issues. We formulate the problem into a mixed integer programming model, and then we convert the constraints and operation issues into the matching conditions of the multi-attribute labels of vehicles, customer demands and routes. Two labelling algorithms are developed to solve the problem. Numerical experiments are conducted on benchmarks, randomly generated problems and real problems from the industry. Third, we study dynamic driver-task assignment problem. Different from the problems in the existing literature, our problem involves both start time windows and random service times explicitly. We formulate the problem in a stochastic optimization framework with the objective of minimizing the costs of current drive-task assignment and the expected future costs. A time-window sliding solution procedure is developed to estimate the expected future costs by solving minimum cost flow problems iteratively. The procedure is tested on randomly generated problems for assessing the efficiency of the algorithm and the benefits of considering uncertainty in the model. Forth, we study the driver-task assignment problem with random service times where the start time windows of tasks can be violated by paying penalties. The problem is formulated into a multistage stochastic optimization model. A label-based solution procedure is developed to build routes into future stages. The expected costs of these routes are used to estimate the costs of future assignments. Experiments are conducted on randomly generated problem sets. At last, it is concluded that the procedures developed in our research can provide quality solutions quickly.