Cold Chain Logistics Hub Location Optimization Considering Direct Transportation and Transportation Network Disruptions

Cold chain logistics (CCL) is critical for preventing quality deterioration in temperature-sensitive products. The rise in direct transportation has transformed traditional transport modes into hybrid configurations, complicating location decisions under disruptions. This paper proposes a multi-objective optimization framework to identify resilient CCL hub locations by integrating relative demand-weight adjustment and stochastic disaster simulations. First, Interpretive Structural Modeling (ISM) adjusts demand weights to reflect hub-bypass flows in hybrid configurations. Second, Monte Carlo Simulation and Gaussian decay models generate 500 disrupted transportation network scenarios to evaluate system robustness under geological hazards. Finally, a centrality-based pre-screening technique identifies candidate nodes before solving the model using the Non-dominated Sorting Genetic Algorithm II (NSGA-II). Validation in the Beijing–Tianjin–Hebei region (BTHr) reveals that the Closest-to-Ideal Solution (CIS) distance exhibits a non-monotonic pattern as the number of hubs increases, with the six-hub configuration providing the best-balanced solution under most tested parameter settings. The resulting hub layout generally avoids disaster-concentrated areas while maintaining accessibility. Furthermore, the spatial analysis indicates differentiated roles: hub-based transportation is better suited to well-connected, stable corridors, whereas direct transportation can complement hub-based services where direct connections remain viable. This hybrid mode coordination provides a pragmatic decision-making framework for regional logistics planners to enhance efficiency and environmental adaptability of cold chain networks.