Logistics Hub Capacity Deployment in Hyperconnected Transportation Network Under Uncertainty

Abstract

Modern logistics systems worldwide are facing unprecedented challenges due to the explosive growth of e-commerce, driving the need for resilient systems to tackle problems such as vulnerable supplies, volatile demands, and fragile transportation networks. Motivated by the innovative concept of the Physical Internet, this paper focuses on resilient capacity deployment of open-access logistics hubs in hyperconnected transportation under demand uncertainty and geographical disruptions. We propose a two-stage stochastic optimization model, aiming to smartly deploy the hub capacity to achieve delivery timeliness, high consolidation and network resilience while minimizing hub set-up budget and truck fleet cost. Four optimal hub network configurations are derived by applying scenarios at four stress testing levels into the optimization model, including deterministic demands without hub disruptions, deterministic demands with hub disruptions, stochastic demands without hub disruptions as well as stochastic demands with hub disruptions. To test the performances of different optimal networks, a simulation-based study is then performed over an automotive delivery-to-dealer network and dataset in the Southeast US region. Our results demonstrate the key impacts of various uncertainties on hub capacity deployment in terms of capacity configuration distribution, network resilience, delivery timeliness, and cost-effectiveness. Overall, this study provides a reliable network capacity deployment approach with persistent and sustainable economic and social performances in hyperconnected networks, and the results validate the relationship between capacity deployment and network resilience under different types of uncertainties.