Optimizing CAPEX and OPEX through Predictive AI Strategies Aligned with ISO 55010 for Improved Maintenance Cost Accuracy

Abstract

Ensuring accurate maintenance budgeting is a critical requirement for sustaining operational reliability in coal-fired power plants, particularly in facilities operating under strict fiscal constraints. At the Palabuhanratu Coal-Fired Power Plant (CFPP), the main challenge addressed in this study is the persistent maintenance budget deviation of up to 30%, driven by reliance on reactive planning and the absence of predictive cost estimation tools. This misalignment between financial planning and asset performance has resulted in inefficient resource allocation, cost overruns, and unplanned downtime. To overcome this, the study aims to develop and implement a predictive artificial intelligence (AI)-driven cost management system, aligned with ISO 55010 principles, optimizing capital expenditure (CAPEX) and operational expenditure (OPEX) allocation while improving maintenance cost accuracy. The methodology integrates multivariate regression modelling and reliability analysis using Python, trained on three years of historical operational data from Enterprise Resource Planning (ERP) and Computerized Maintenance Management System (CMMS) platforms. The system is deployed via an interactive Looker Studio dashboard that enables real-time monitoring and cross-functional alignment between finance and engineering. Implementation results show a reduction in maintenance costs from USD 144,000 to USD 41,867, a decrease in MWh loss from 72.7 GWh to 43.1 GWh, and total annual savings of USD 2.19 million. The findings demonstrate that predictive AI, when integrated with ISO 55010-based asset management, can significantly enhance cost efficiency, operational reliability, and budgetary precision. This approach not only addresses critical challenges in maintenance cost management but also provides a replicable framework for other power plants, supporting broader adoption of data-driven asset optimization practices in the energy industry.