MACHINE LEARNING-BASED HIERARCHICAL CLUSTERING FOR PRIORITY CCUS ZONES IN INDONESIA

Indonesia's upstream oil and gas industry is facing significant challenges due to climate change, alongside the global shift toward clean energy transition to reduce CO₂ emissions and achieve the Net Zero Emission target by 2060. One of the key strategies in this effort is the implementation of Carbon Capture, Utilization, and Storage (CCUS).The adoption of the CCUS program is an integral part of SKK Migas' strategic plan to reach a production target of 1 million barrels of oil per day (BOPD) and 12 billion standard cubic feet of natural gas per day (BSCFD) by 2030. This study applies machine learning-based hierarchical clustering to analyze and classify CCUS project zones in Indonesia, utilizing data from the IEA CCUS Projects Database 2024. The methodology includes data collection, pre-processing, and clustering using a hierarchical algorithm to group projects with similar characteristics in CCUS implementation. The clustering process, interpreted through a dendrogram, considers key factors such as Announced Capacity (Mt CO₂/yr) and Estimated Capacity by IEA (Mt CO₂/yr). The Silhouette Coefficient after applying hierarchical clustering is 0.746, indicating well-defined cluster separation. The findings of this study provide valuable insights into the relationships among CCUS projects in Indonesia, categorizing them into priority zones. Additionally, this research supports strategic decision-making regarding CCUS project development, contributing to the achievement of the Net Zero Emission target and long-term energy security.