<![CDATA[Optimizing the utilization of idle industrial assets is a crucial issue in asset-based management strategies, especially in the high-risk and capital-intensive petrochemical sector. PT Pupuk Iskandar Muda (PT PIM) as part of the development of the Arun Lhokseumawe Special Economic Zone (SEZ), is faced with the urgency to determine the most rational and value-added operational scheme in reactivating the ex-PT AAF Hydrogen Peroxide (H?O?) Factory. There are three alternative schemes analysed in this context: lease, operating cooperation (KSO) and self-management, each of which has different financial characteristics and risk profiles. The complexity of such decision-making requires a data-driven approach that is able to integrate economic and risk aspects holistically. This study aims to determine the most feasible operational scheme based on economic indicators such as Net Present Value (NPV), Internal Rate of Return (IRR) and Payback Period (PP), as well as two main risk dimensions, namely operational risk and market risk. This study also aims to build a decision tree-based decision-making model using the C4.5 algorithm to classify the feasibility of the scheme in a systematic and transparent manner. The research method uses a descriptive-analytical quantitative approach. The data used are secondary data from the project feasibility study and PT PIM's internal documents. Modeling was carried out by processing six main variables in categorical form to form a decision tree structure through the C4.5 algorithm. The process starts from calculating the entropy value, gain, to determining the root node and branch of the decision. The classification results show that market risk is the most decisive attribute in the decision-making process, followed by NPV as the main separating indicator between the "Feasible" and "Very Feasible" feasibility levels. The main findings of this study are that the self-management scheme is the most feasible option with the best financial performance: NPV of ±Rp 38.77 billion, IRR of 14.48%, and Payback Period of 5.58 years. The structure of the established decision tree is able to explicitly describe how the combination of risk and financial indicators can lead to different final decisions. This model serves not only as a classification tool, but also as an analytical instrument in understanding the relationships between variables and their implications for investment feasibility. This research makes a methodological contribution to the application of the C4.5 algorithm for data-driven strategic decision-making. This approach has been proven to improve accuracy, objectivity, and transparency in assessing the operational feasibility of industrial assets, and can be replicated for similar cases in other sectors.]]>
