[CS.AI] Brick-DICL: Dynamic In-Context Learning for Building Management Schema

Abstract

Building Management Systems (BMS) are essential for optimizing energy efficiency and operational performance in modern buildings. However, the lack of standardization across BMS points from different manufacturers creates significant barriers to integration and data utilization. While the Brick schema offers a standardized ontology for building systems, mapping BMS points to appropriate Brick classes presents three critical challenges:

1. Extensive Number of Brick Classes: 936 classes in the latest version.
2. Limited Domain-Specific Knowledge: Large language models (LLMs) lack sufficient domain knowledge.
3. Substantial Manual Verification Effort: A significant amount of manual effort is required for verification.

To address these challenges, we propose Brick-DICL, a two-stage dynamic in-context learning framework for automated Brick schema classification. Brick-DICL consists of two primary components:

• metadata-RAG: retrieves relevant examples to enhance LLMs' domain knowledge.
• class-RAG: narrows down potential Brick classes to address the large classification space.

Additionally, we implement a multi-LLM filtering mechanism that compares predictions across multiple models, flagging low-confidence classifications for human review.

Results

1. General Applicability: Brick-DICL is applicable to any building management system regardless of manufacturer or metadata format.
2. Novel and Powerful: As the first dynamic in-context learning approach for Brick schema classification, Brick-DICL achieves significant classification accuracy improvements on building datasets, outperforming existing methods.
3. Efficient: Our multi-LLM filtering strategy reduces manual verification effort, enabling rapid digital building onboarding.

Extensive experiments demonstrate Brick-DICL's effectiveness across diverse building datasets, accelerating the path toward standardized, interoperable building management systems.

Blogger's Review: Brick-DICL effectively addresses the standardization issues in building management system classification through dynamic in-context learning, showcasing its broad applicability and efficiency in real-world applications. The multi-LLM filtering mechanism offers a novel approach for future building data integration, warranting further exploration.