Date of Award
Open Access Dissertation
Civil and Environmental Engineering
Nathan N. Huynh
The growth of e-commerce is changing consumer expectations for delivery services. Competition is making “faster and cheaper” a nonnegotiable part of success; hence, retailers are seeking carriers that can deliver with ever-shorter lead times and at the lowest possible cost. In this environment, large carriers have a competitive advantage due to their market power. For small-to-medium-sized less-than-truckload (LTL) carriers to stay in business, they must improve their efficiency and lower cost. A potential strategy to achieve these goals is to collaborate with other carriers. In collaborating, carriers would exchange and perform certain jobs for each other to lower their transportation costs. To this end, this dissertation presents three studies, as explained in the subsequent paragraphs.
The first study investigates the potential benefits of collaboration among LTL carriers in fulfilling pickup and delivery jobs. It adopts the commonly used centralized collaborative planning scheme where a central authority pools all jobs and allocates them to the carriers with the objective of minimizing the total transportation cost. However, in this study, LTL carriers are allowed to retain some of the jobs. This extension to the centralized planning scheme is necessary to make collaboration practical. To this end, a mathematical model and a solution method based on large neighborhood search (LNS) are proposed. The experimental results from hypothetical networks provide several important insights regarding cost savings for various collaboration scenarios.
The second study examines three methods of collaboration and profit-sharing for LTL carriers. The collaboration methods are evaluated under a centralized collaboration scheme proposed in the first study. Collaboration Method 1 is a two-step approach where the first step involves the central authority determining the job allocation for the shared jobs and the vehicle routes for each carrier that includes their retained and allocated jobs to maximize total profit. The second step involves dividing the total profit among the carriers using a contribution-based profit-sharing model. Collaboration Method 2 is a one-step approach where the central authority determines the job allocation and vehicle routes, and at the same time, allocates profit to the carriers with fairness constraints included in the model. Collaboration Method 3 is also a two-step approach similar to Method 1, except that in the first step, the central authority determines the job allocation and vehicle routes for only the shared jobs (not including retained jobs). Mathematical models and solution algorithms based on LNS are proposed for all three methods. Numerical experiments are conducted, and the results indicate that Method 1 outperforms Methods 2 and 3.
The last study addresses the LTL pickup and delivery problem where carriers collaborate in a decentralized and dynamic manner. It is assumed that carriers do not have any collaboration agreement with one another; however, they collaborate indirectly due to the need to outsource jobs and/or the opportunity to bid for jobs to increase profit. In the context of this study, trucks operate independently and make decisions on outsourcing and acquiring jobs on their own. To address this problem, an approach based on a multi-round combinatorial auction (CA) is proposed. The novelty of the proposed approach is that in each round of the auction, trucks can bid for multiple bundles in contrast to the commonly used approach of a single bundle. Mathematical models and solution algorithms based on LNS are proposed. Numerical experiments are conducted and the results indicate that the proposed auction method provides a higher profit and a higher number of jobs served in less time.
Padmanabhan, B.(2023). Assessment and Development of Collaboration Framework for Less-Than-Truckload Carriers. (Doctoral dissertation). Retrieved from https://scholarcommons.sc.edu/etd/7395
Available for download on Sunday, August 31, 2025