AI-Integrated Digital Twin Framework for Quality and Performance Management in Sustainable Buildings

Operational quality in sustainable buildings is often compromised by fragmentation across Building Information Modeling (BIM), the Internet of Things (IoT), and Building Management Systems (BMS). This paper proposes an Artificial Intelligence (AI)-integrated Semantic Digital Twin (SDT) framework that redefines quality management as a traceable and governance-ready operational process. The framework combines ontology-driven semantics, hybrid physics-AI modeling, and a requirements-to-KPI traceability graph to connect lifecycle intent, data provenance, analytic outputs, and operator decisions. A seven-layer architecture is introduced together with an assurance loop that validates data, quantifies uncertainty, records explainability artifacts, and enforces human approval before controlled actuation. To address evaluation needs, the paper also defines a benchmark-ready Design Science Research blueprint using BOPTEST and governance metrics, including energy use intensity, comfort violation hours, CO2 exceedance hours, fault detection performance, uncertainty calibration, traceability completeness, and approval latency. The contribution is therefore not only an architectural model, but also a practical audit-ready basis for trustworthy and scalable building operations.