Building Sustainability Management Using LLM Recommendations
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current building management systems lack the ability to autonomously generate and implement sustainable recommendations for improving building sustainability performance without manual intervention, particularly in handling unstructured data and dynamic user contexts.
Innovation Solution
A method utilizing a generative large language model (LLM) to autonomously generate recommendations for improving building sustainability, including natural language summaries, based on unstructured data from building information models, specifications, and operational data, without requiring manual input, and implementing these recommendations to enhance sustainability performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If a generative LLM is used to autonomously generate sustainability recommendations, then the extent of automation is improved, but the device complexity increases
Solution Approach 1:
The patent introduces a building management system as an intermediary layer between raw building data and sustainability recommendations. This intermediary processes unstructured data from multiple sources (building information models, operational data, specifications) and presents structured recommendations to users, thereby managing system complexity while maintaining high automation levels
Solution Approach 2:
The system segments the complex task of sustainability recommendation generation into distinct functional modules: data collection from multiple sources, unstructured data processing, recommendation generation, and user presentation. This segmentation allows each module to be optimized independently, reducing overall system complexity while preserving automation capabilities
2Adaptability or versatility
If unstructured data from multiple sources is processed, then the adaptability is improved, but the difficulty of detecting and measuring increases
Solution Approach 1:
The building management system is designed with multi-functional capabilities to handle diverse data types from various sources including building information models, operational data, and specifications. This universal processing approach enables the system to adapt to different data formats and sources without requiring separate processing pipelines for each type
Solution Approach 2:
The system creates standardized data representations and models that replicate the essential characteristics of diverse unstructured data sources. By working with these standardized copies rather than raw heterogeneous data, the system reduces processing difficulty while maintaining adaptability to various data formats
3Ease of operation
If natural language summaries are dynamically generated, then the ease of operation is improved, but the loss of time in processing increases
Solution Approach 1:
The system performs preliminary processing and structuring of sustainability recommendations before presenting them to users. By pre-organizing data and generating key insights in advance, the system reduces the time required for final presentation and improves ease of operation without significant additional processing time
Solution Approach 2:
The patent replaces manual analysis and interpretation of sustainability data with automated natural language generation. This substitution of mechanical human processes with computational processes improves ease of operation by providing clear, accessible summaries while the automation actually reduces overall processing time compared to manual analysis
Data Source
AI summary
Systems and methods are disclosed relating to building management systems with sustainability improvement for a building. For example, a system can include at least one machine learning model configured using training data that includes at least one of unstructured data or structured data regarding sustainability of buildings. The system can provide inputs, such as prompts, to the at least one machine learning model regarding a sustainability performance of the building, and generate, according to the inputs, responses regarding the sustainability performance of the building, such as responses for detecting factors and/or sources contributing to the sustainability performance of the building.