Optimizing carbon capture and storage (CCS) infrastructure development using a python tool for source-sink matching and cluster formation

The large-scale deployment of Carbon Capture and Storage (CCS) infrastructure is critical for Kazakhstan's decarbonization strategy, particularly in hard-to-abate industries. This study develops an optimization-driven framework for CCS infrastructure planning, focusing on source-sink matching and cost-effective clustering of emitters. A Python-based mixed-integer linear programming (MILP) model is applied to identify optimal CCS hub configurations by minimizing the total cost of capture, transportation, and storage while accounting for environmental impact. The results demonstrate that a 100 km clustering radius achieves the lowest cost scenario, with full redirection costing approximately $23.30 billion and partial redirection $24.02 billion. However, the CO 2 net abated is highest in this scenario, with emissions reaching 140.55 MtCO 2 and 142.73 MtCO 2 , respectively. A 240–300 km clustering radius offers a more balanced trade-off, stabilizing costs at approximately $34.57 billion while reducing emissions. The analysis also reveals that capture costs represent the greatest portion of total CCS expenses (50.47 % at 100 km), while transportation costs increase with clustering radius, rising from 45.61 % to 61.01 % at 180 km. Leveraging existing pipeline infrastructure significantly improves cost efficiency, particularly in shorter clustering radii. The findings highlight the importance of integrating cost-efficient CCS hub networks and policy interventions, including regulatory frameworks, financial incentives, and international cooperation, to accelerate Kazakhstan's transition to a low-carbon economy. Future research should refine dynamic clustering models and incorporate real-time storage capacity assessments for improved decision-making. To enhance model repeatability, we provide an open-source Python code for future use. • First optimized CCS hubs in Central Asia using MILP modeling. • Developed a new method with full and partial redirection for source-sink matching. • Identified 17 storage sites with 5784 Mt CO 2 capacity. • Found 100 km clustering minimizes cost, while 240–300 km balances emissions. • Created an open-source Python tool for flexible CCS planning.