CONTAINER TRUCK ROUTING AND SCHEDULING PROBLEMS UNDER A SHARED RESOURCE ENVIRONMENT

*PhD Defense*
Time
May 24 2017 20:30–22:30
Location
4080 AIR Building
Kyungsoo Jeong
Kyungsoo Jeong
TSE PhD Candidate
Abstract

More frequent vehicle movements are required for moving containers in a local area due to low unit volume that a single vehicle can handle compared with vessels and rails involved in the container supply chain. For this reason, truck operations for moving containers significantly affect not only transportation cost itself but also product price. They have inherent operational inefficiencies associated with empty container movements and container processes at facilities such as warehouses, distribution centers and intermodal terminals. One critical issue facing the trucking industry is the pressing need for truck routing plans that reduce such inefficiencies. Hence, this dissertation proposes to apply the concept of sharing resources, which is an emerging economic model, to container truck operations in order to resolve this issue. Two shareable resources – vehicles and containers – are considered.

This study extends the literature on routing and scheduling problems that arise from container movements, and examines the possible benefits of sharing resources across customers. A series of truck container routing and scheduling problems were developed by assuming different levels of resource sharing among; (1) customers of one trucking operator, (2) customers across collaborations of multiple operators, and (3) customers over multi-day operations. To enable a trucking company to operate its fleet under a shared resource environment, two operational strategies – street turning and decoupling operations – together with temporal precedence constraints – in addition to the time constraints that are typically included in the vehicle routing problem with time windows (VRPTW) – were adopted to address the proposed problems.
Two meta-heuristic algorithms based on a variable neighborhood search (VNS) scheme were developed to solve the proposed problems, including temporal precedence constraints – which are computationally more expensive – for real-world applications. To address flexible time windows resulting from temporal precedence constraints, a novel feasibility check algorithm was developed.

Results from a series of numerical experiments confirm that the proposed approach leverages the advantages of resource sharing, and the meta-heuristic algorithms are efficient solution approaches for each problem with the targeted resource sharing. Consequently, this dissertation offers a platform for the development of a decision-support tool for drayage companies by applying three different levels of resource sharing into their operations.