Resilient batch capture and storage siting for carbon networks under supply uncertainty

We study an integrated planning problem for truck-based carbon capture, utilization, and storage (CCUS) networks that co-optimizes strategic storage siting and capacity decisions with operational multi-pickup batch collection under stochastic CO2 supply and demand. Unlike conventional single-site pickup policies, we adopt a multi-pickup batch strategy in which a tanker truck visits multiple emitters in a single tour, consolidating loads before delivering to an intermediate storage hub. To quantify the delay cost inherent in batch formation, we embed a state-dependent M/G/1 queueing model at each capture site and derive the optimal batch threshold that balances dispatch frequency against latency-induced leakage penalties. The resulting two-stage stochastic optimization model is nonlinear due to the queueing-based latency term; to improve tractability, we employ a second-order cone programming (SOCP) convex reformulation together with an outer-approximation (OA) algorithm for solving realistically sized instances.Computational case studies based on a dispersed provincial network in Shandong, China, and a clustered industrial hub on the Texas Gulf Coast, USA, show that batch pickup consistently reduces total system cost by 4.5–6.9% relative to single-site pickup benchmarks, with savings driven primarily by a 15% reduction in capture-to-storage transport costs and a 67% reduction in queueing-related latency costs. Sensitivity analyses reveal that batch savings increase with supply and demand variability and are most pronounced when the system is capacity-constrained, while the proposed OA algorithm achieves near-optimal solution quality with improved scalability over direct mixed-integer SOCP solving. These results provide actionable guidance for when and where multi-pickup consolidation should be adopted in early-stage CCUS logistics networks.