AI-Powered Digital Twin for Sustainable Campus Facility Management

Abstract

University campus buildings consume significant energy through chilled water (CHW) systems, yet facility and energy managers lack tools that connect sensor data to actionable, spatially grounded decisions without requiring programming expertise. We present BearcatAI, an end-to-end analytics platform developed at the University of Cincinnati that unifies machine learning demand prediction, a large language model (LLM) conversational interface, an interactive 3D GIS-derived campus visualization, and a ROS-based robotic inspection system into a single web-accessible workflow. The platform is informed by ten years of monthly CHW records (July 2015–July 2025) and validated against 271 days of daily meter data across 70 UC buildings (June 2025–February 2026). An ensemble of XGBoost and Random Forest models achieves R2=0.987 and MAPE =21.8% on DAAP, the priority building identified by the UC utility engineering team. Beyond building-specific models, we introduce use-type models that pool buildings by functional category (Academic, Research, Administrative, Housing, Recreation, Medical) and learn a transferable CHW intensity pattern without any building-specific identity features. Leave-One-Building-Out (LOBO) cross-validation shows that the Research type model generalizes to unseen research buildings with median R2=0.911, and the Academic model achieves R2=0.918 on DAAP without having seen it during training, validating the hypothesis that a functional type model can serve as a generalizable benchmark for campus CHW consumption. A natural language interface powered by BearcatGPT, UC’s hosted instance of GPT-4.1, free for students and faculty, allows utility managers to query anomalies, request forecasts, and generate building charts through conversation. Anomalous buildings are automatically highlighted in a 3D campus model spanning UC Main Campus, Medical Campus, and Digital Futures. A ROS-based inspection robot, trained in a Digital Futures digital twin via reinforcement learning, is under ongoing development and is described as a future integration target. The platform is designed for open deployment on JetStream2 academic cloud infrastructure.

The authors' abstract, as published at the source. Architecture, 2026 · DOI ↗