Title: Hyperconnected Logistic Hub Networks: Dynamic Capacity Planning and Management
Date: November 19, 2025
Time: 8:00 am – 10:00 am (EST)
Meeting Link: Microsoft Teams Meeting
Xiaoyue Liu
Ph.D. Candidate in Operations Research
School of Industrial and Systems Engineering
Georgia Institute of Technology
Thesis Committee:
Dr. Benoit Montreuil (advisor), School of Industrial and Systems Engineering, Georgia Institute of Technology
Dr. Mathieu Dahan, School of Industrial and Systems Engineering, Georgia Institute of Technology
Dr. Walid Klibi, Supply Chain Center of Excellence, KEDGE Business School
Dr. Valerie Thomas, School of Industrial and Systems Engineering, Georgia Institute of Technology
Dr. Lóri Tavasszy, Freight Transport & Logistics, Delft University of Technology
Abstract:
Over the past decade, the logistics and transportation industry has faced unprecedented challenges driven by surging demand variability, frequent supply disruptions, and growing sustainability pressures. These trends have accelerated the transformation of supply chains from traditional, static, and fragmented structures toward more agile, distributed, and dynamic network designs. The Physical Internet paradigm facilitates this transformation by promoting an open, modular, and hyperconnected logistics ecosystem where assets and capacities are flexibly shared and efficiently coordinated. Within this context, this dissertation thesis focuses on advancing dynamic capacity planning and management in hyperconnected logistics networks, aiming to determine optimal capacity allocations that balance operational efficiency, service reliability, cost effectiveness, and environmental sustainability.
In Chapter 2, we study the stochastic hub capacity-routing problem in hyperconnected relay transportation networks, where long-haul shipments are decomposed into multiple short-haul segments operated by local drivers between hubs. The problem focuses on determining optimal hub capacity reservation and routing decisions for relay truck carriers to ensure efficient long-haul freight movements. We formulate this decision problem as a two-stage stochastic optimization model that minimizes hub and subsequent transportation costs under demand and travel time uncertainties. To address the model’s computational complexity, we develop a combinatorial Benders decomposition algorithm enhanced with a tailored branch-and-cut framework and heuristic initial cut pool generation strategies. The proposed algorithm is validated through a case study in the automotive delivery sector, demonstrating its effectiveness. The findings further offer managerial guidance for relay truck carriers in developing hub contracting strategies.
In Chapter 3, we study the stochastic modular and mobile capacity planning problem, which captures the emerging need for adaptive resource allocation and relocation in hyperconnected supply chains. We formulate the problem as a multi-stage stochastic programming model that incorporates both modular facility short-term leasing and mobile resource relocation decisions under both demand and supply uncertainty. To solve this problem efficiently, we develop an enhanced stochastic dual dynamic integer programming algorithm that integrates strengthened cut generation, an alternating cut strategy, and a parallelization framework to accelerate convergence. We apply our methodology to a real-world case study in the modular construction sector, where the company dynamically adjusts modular assembly centers and mobile production units to meet evolving project demands. The results provide insights into how modularity and mobility can be strategically leveraged to enhance responsiveness in large-scale supply networks under uncertainty.
In Chapter 4, we study the two-echelon vehicle sharing and repositioning problem within hyperconnected urban logistics systems. The problem considers containerized delivery operations across a multi-tier hub network, where vehicles of varying sizes are shared and repositioned among logistic hubs to serve fluctuating urban demands. We formulate the problem as a multi-period integer programming model that integrates route-based vehicle planning with flow-based container routing to dynamically allocate and manage service capacity. To efficiently solve large-scale instances, we propose a decomposition-based heuristic algorithm that partitions the problem into echelon-structured subproblems, achieving high-quality solutions with substantial computational efficiency. We conduct a case study using real-world freight flow data from the Atlanta metropolitan area. The results show that integrating hyperconnected urban logistics networks with dynamic vehicle repositioning enables more efficient, cost-effective, and sustainable operations.