City University of Hong Kong

CityU Institutional Repository >
3_CityU Electronic Theses and Dissertations >
ETD - Dept. of Management Sciences  >
MS - Doctor of Philosophy  >

Please use this identifier to cite or link to this item:

Title: Freight allocation problems in the transportation industry
Other Titles: Jiao tong yun shu ye zhong de huo wu fen pei wen ti
Authors: Qin, Hu ( 秦虎)
Department: Department of Management Sciences
Degree: Doctor of Philosophy
Issue Date: 2011
Publisher: City University of Hong Kong
Subjects: Freight and freightage -- Management.
Notes: CityU Call Number: HE199.A2 .Q25 2011
130 leaves : ill. 30 cm.
Thesis (Ph.D.)--City University of Hong Kong, 2011.
Includes bibliographical references (leaves 121-130)
Type: thesis
Abstract: This thesis studies freight allocation problems faced by the transportation practitioners at both strategic and operational levels. Shippers and carriers are the key players in the transportation market. The shippers are manufacturers, buying agents, third-party logistics providers (3PL) and any organization that needs to move freight. The carriers are the transportation service providers who can move the freight, such as shipping and trucking companies. The problems investigated in the thesis are all motivated by real projects and related to long-distance ocean shipping. In the transportation market, carriers often offer price discounts to encourage shippers to purchase more transportation service. Usually, the discount rate offered by each carrier is related to the total freight quantity gained from the shipper across all shipping lanes. The first problem in the thesis aims at helping a shipper allocate forecasted annual demand of each shipping lane to candidate carriers with consideration of discount and minimum quantity commitment while minimizing the total transportation cost. This problem is NP-hard in the strong sense, and is therefore unlikely to be solvable optimally in reasonable computation time for large instances. Hence, we propose a heuristic-based algorithm that combines a filter-and-fan search scheme and a tabu search mechanism for the problem. Experiments on a large number of randomly generated test instances show that as the problem size increases, our algorithm produces superior solutions in less computation time and requires less computer memory, compared to a leading mixed integer programming (MIP) solver. When conducting freight allocation, the shipper might need to consider some other aspects. Next, we study a freight allocation problem for a shipper who acts as the buying office of a large international retail distributor. The task of the shipper is to plan the distribution of goods from Asia to various sales divisions across Europe. The goods are transported along shipping lanes by carriers, many of which have collaborated to form strategic alliances. Each lane must be serviced by a minimum number of carriers that must belong to a minimum number of alliances and each carrier requires a minimum total freight quantity over all lanes. Moreover, the allocation must not assign an overly high proportion of freight to the more expensive shipping companies servicing any particular lane in order to ensure fairness for different sales divisions, which we call the lane cost balancing constraint. The problem is to allocate projected annual demand of each shipping lane to carriers taking all the above constraints into account, such that the total transportation cost is minimized. We formulate this problem into an MIP model, and show that not only is finding an optimal solution computationally intractable, but so is finding a feasible solution. Therefore, to produce high-quality solutions practically, we devised a meta-heuristic approach for the problem based on tabu search. Our approach has since been developed into an application that is currently employed by the shipper, and it has proved to be a powerful and effective support tool. The two problems already described both appear at the strategic level. In practice, many shippers need to optimize the distribution of goods at the operational level. The third problem in the thesis produces execution plans for a 3PL to send a batch of shipments to their retail stores distributed in the United States. Each shipment is firstly shipped to one of distribution hubs in the United States from one warehouse hub in China by container transportation and then sent to its retail store by some parcel express company. Different distribution hubs lead to different parcel delivery costs for each shipment and we assume each distribution hub corresponds to only one shipping route. In addition, one shipment may consists of one or multiple items and for simplifying store operations the items belonging to the same shipment are required to be transported along a unique shipping route. The problem makes decisions on how to allocate shipments to shipping routes and how to load items into containers with various sizes while minimizing the sum of container and parcel delivery costs. We formulate this problem as a variant of the variable size bin-packing problem (VSBPP), which is obviously NP-complete. To solve this problem practically, we propose a genetic algorithm that embeds some local search heuristics and heuristics for the VSBPP. The practical usefulness of the model as well as its solution approach is substantiated by its deployment with a multinational 3PL.
Online Catalog Link:
Appears in Collections:MS - Doctor of Philosophy

Files in This Item:

File Description SizeFormat
abstract.html132 BHTMLView/Open
fulltext.html132 BHTMLView/Open

Items in CityU IR are protected by copyright, with all rights reserved, unless otherwise indicated.


Valid XHTML 1.0!
DSpace Software © 2013 CityU Library - Send feedback to Library Systems
Privacy Policy · Copyright · Disclaimer