InsightBuild: LLM-Powered Causal Reasoning in Smart Building Systems

TL;DR

InsightBuild combines causal inference and large language models to generate human-readable explanations of energy anomalies in smart buildings, aiding facility managers in diagnostics and efficiency improvements.

Contribution

This work introduces a novel two-stage framework integrating causal discovery with LLMs for explainable energy analysis in smart buildings.

Findings

Effective causal explanations generated for real-world datasets.

Improved diagnosis of energy anomalies with human-readable insights.

Demonstrated applicability on Google and Berkeley building data.

Abstract

Smart buildings generate vast streams of sensor and control data, but facility managers often lack clear explanations for anomalous energy usage. We propose InsightBuild, a two-stage framework that integrates causality analysis with a fine-tuned large language model (LLM) to provide human-readable, causal explanations of energy consumption patterns. First, a lightweight causal inference module applies Granger causality tests and structural causal discovery on building telemetry (e.g., temperature, HVAC settings, occupancy) drawn from Google Smart Buildings and Berkeley Office datasets. Next, an LLM, fine-tuned on aligned pairs of sensor-level causes and textual explanations, receives as input the detected causal relations and generates concise, actionable explanations. We evaluate InsightBuild on two real-world datasets (Google: 2017-2022; Berkeley: 2018-2020), using expert-annotated ground-truth causes for a held-out set of anomalies. Our results demonstrate that combining explicit causal discovery with LLM-based natural language generation yields clear, precise explanations that assist facility managers in diagnosing and mitigating energy inefficiencies.