An Artificial Intelligence–Enabled Assessment Framework for Energy Conservation in Retrofitted and Operational Buildings

Accelerating energy efficiency improvements in the built environment is central to ASEAN’s decarbonization pathway and long-term climate commitments. However, the diverse operational profiles, climatic conditions, and retrofit strategies across the region pose significant challenges for consistent and credible assessment of energy conservation performance. To address this gap, this paper presents a standardized, data-driven evaluation framework that harnesses artificial intelligence (AI) and machine learning (ML) to analyze and verify energy performance in both retrofitted and actively operating buildings. Designed to be compatible with regional policies—including the objectives of the ASEAN Plan of Action for Energy Cooperation (APAEC), national green building rating tools, and emerging carbon management regulations—the framework integrates core indicators such as Energy Use Intensity (EUI), temporal operation patterns, equipment load profiles, and localized meteorological datasets. These inputs are synthesized to establish adaptive baseline models capable of capturing real-time performance variations and quantifying energy savings with higher accuracy than conventional static or rule-based methods. The proposed methodology is structured for application across commercial, institutional, and mixed-use developments, enabling facility owners, policymakers, and financial institutions to systematically track conservation outcomes and evaluate the effectiveness of retrofit interventions. Beyond performance verification, the framework is positioned to support regional green financing ecosystems, including sustainability-linked incentives, performance-based procurement, and digital MRV (measurement, reporting, and verification) systems increasingly adopted across ASEAN.