Resource management in transportation networks: addressing challenges and optimizing efficiency

Abstract

The American economy heavily relies on transportation, which contributed 8% to the GDP of the United States in 2020. The transportation sector, being the fourth-largest contributor to the overall GDP, encompasses various industries such as airlines, trucking, railroads, shipping, logistics firms, and transportation infrastructure providers. The efficiency of transport systems plays a vital role in facilitating better access to markets, employment opportunities, and additional investments. However, inefficiencies within the transportation sector, including poor resource management, pose challenges that need to be addressed for enhanced sector-wide efficiency and economic benefits. This dissertation endeavors to develop effective optimization techniques and algorithms to tackle diverse transportation challenges, aiming to optimize resource allocation and enhance resource utilization within the transportation sector. This abstract highlights the key research contributions and findings of the dissertation: Firstly, the dissertation focuses on addressing hub location problems in transportation networks. Existing hub covering models often lead to congestion and inefficiency, compromising network coverage. To overcome this, two hub covering location models are proposed, taking into account the busy probability of hub nodes. The first model assumes the known number of servers in each hub, while the second model considers the number of servers in each hub as a decision variable. Metaheuristics based on the Genetic algorithm and Tabu Search are developed to solve the models, demonstrating their efficiency through experimental results on American Airlines domestic flights in 2019. Next, the complexity of the multiple allocation hub maximal covering problem (MAHMCP) is explored. A branch and cut approach is developed to solve this problem, providing a stronger theoretical model compared to previous approaches. The effectiveness of the proposed model is demonstrated through an empirical study using the Australia Post (AP) dataset, showcasing optimal solutions and fast run times for instances up to 100 nodes. Furthermore, the dissertation analyzes the chassis inventory management problem in an intermodal transportation system. A chassis connects shipping containers to trucks. Having a chassis available at the right time and place is essential for efficient loading, unloading and subsequent transportation of goods. Among the most significant challenges that intermodal transportation companies face is chassis shortages at terminals. To address this, A multi-time period mathematical model with n terminals is developed to determine the daily moves of chassis and empty containers between terminals, ensuring efficient loading, unloading, and transportation of goods while minimizing costs, and meeting demand requirements. Lastly, the dissertation addresses one of the main challenges faced by the US truckload industry - high turnover rates among truckers due to demanding working conditions. To mitigate this challenge, a relay network approach is proposed, consisting of a depot and relay location problem as well as a routing problem. This comprehensive approach aims to reduce drivers' away-from-home times. In conclusion, this dissertation emphasizes the importance of addressing sector-specific challenges and optimizing resource allocation and utilization within the transportation sector. The developed optimization techniques and algorithms contribute to enhancing resource management and system efficiency.