Date of Award

2019

Degree Type

Open Access Dissertation

Degree Name

Engineering and Industrial Applied Mathematics Joint PhD with California State University Long Beach, PhD

Program

Institute of Mathematical Sciences

Advisor/Supervisor/Committee Chair

Anastasios Chassiakos

Dissertation or Thesis Committee Member

Hossein Jula

Dissertation or Thesis Committee Member

Allon Percus

Dissertation or Thesis Committee Member

Henry Schellhorn

Terms of Use & License Information

Terms of Use for work posted in Scholarship@Claremont.

Rights Information

© 2019 Timothy M VanderBeek

Keywords

Chassis Exchange, Chassis Processing Facility, Container Movements, Port of Long Beach, Port of Los Angeles, Transportation Optimization

Subject Categories

Applied Mathematics | Transportation Engineering

Abstract

This work studies the concept of “Centralized Processing of Chassis,” and its potential impact on port drayage efficiency. The concept revolves around an off-dock terminal (or several off-dock terminals), referred to as Chassis Processing Facilities (CPFs). A CPF is located close to the port, where trucks will go to exchange chassis, thereby reducing traffic at the marine terminals and resulting in reduced travel times and reduced congestion. This work is divided into two major studies: one at the strategic planning level, and one at the operational level for individual trucking companies.

In the first study, an analytical framework for modeling and optimization of chassis movements in transportation networks with CPFs is developed, and a case study in the Long Beach/Los Angeles (LB/LA) port area is performed. Comparisons between current practices at ports, in which chassis exchanges occur at marine terminals, and proposed practices, in which the exchanges happen at CPFs, are performed. The results of this study indicate that a reduction of total travel time by up to 20% can be achieved when using the CPFs. The study also shows that, in the LB/LA port area, the return on investment for establishing additional CPF locations decreases sharply for any more than three CPFs. Overall, the findings indicate that travel time can be significantly reduced through implementation of CPFs which has important implications in reducing negative environmental impacts of the port as well as operational costs for trucking companies.

In the second study, scheduling of chassis and container movements is optimized at the operational level for individual trucking companies, when CPFs are available for use within a major metropolitan area. A multi-objective optimization problem is formulated in which the weighted combination of the total travel time for the schedules of all vehicles in the company fleet and the maximum work span across all vehicle drivers during the day is minimized. Time-varying dynamic models for the movements of chassis and containers are developed and used in the optimization process. The optimal solution is obtained through a genetic algorithm, and the effectiveness of the developed methodology is evaluated through a case study which once again focuses on the LB/LA port area. The case study uses a trucking company located in the Los Angeles region, which can utilize three candidate CPFs for exchange of chassis. The company assigns container movement tasks to its fleet of trucks, with warehouse locations spread across the region. In the simulation scenarios developed for the case study, the use of CPFs at the trucking company level, can provide improvements up to 30% (depending upon the specific scenario) over the cases not using any CPFs. It is found in this work that for typical cases where the number of jobs is much larger than the number of vehicles in the company fleet, the greatest benefit from CPF use would be in the cases where there are some significant job-to-job differences with respect to chassis usage and type.

Lastly, in addition to the formulation and optimization for initially planning daily activities, the study further models the problem in a dynamic environment, in which traffic network parameters can change drastically from initial daily predictions. In order to perform the optimization in a dynamic formulation with varying noise levels, a method by which noise could be injected into the initial daily predictions is developed to support the model inputs for the case study and an incremental optimization approach is implemented. Results indicate that a modest potential benefit of approximately 2% may be expected if dynamic re-routing is performed. However, in practice it will be important to weigh the cost of the additional real-time queries required to enable the dynamic re-routing against the potential benefits for the specific company and job set in question prior to implementation.

ISBN

9781085796767

Download

Share

COinS