Building Alarm Classification Using Multi-Agent Generative AI
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Building management systems often generate false alarms, diverting operator attention and resources away from actual issues, and there is a need for automated methods to distinguish between false and true alarms.
Innovation Solution
Implementing a Generative Artificial Intelligence (AI) Large Language Model-based system with primary agents and subagents trained in specific domains to analyze alarms, generate plausible causes, and classify them as false or true based on domain knowledge and collaborative reasoning.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a Building Management System monitors and raises alarms for abnormalities, then the system can detect potential issues, but false alarms divert operator attention and waste time
Solution Approach 1:
The system employs autonomous AI agents that automatically analyze alarms, determine their validity, and classify them without requiring operator intervention. The primary agent receives alarms, generates scenarios, assigns subagents, collects results, and makes autonomous classification decisions, enabling the system to serve itself in the alarm analysis process.
Solution Approach 2:
The patent replaces manual operator analysis with an automated AI-based system. The primary agent and subagents use generative AI models to automatically process alarm data, generate plausible scenarios, and classify alarms as false or true, substituting the mechanical human decision-making process with an automated intelligent system.
2Productivity
If operators manually analyze each alarm, then they can determine true vs false alarms, but this increases operational complexity and reduces efficiency
Solution Approach 1:
The alarm analysis system is segmented into hierarchical levels: a primary agent that handles high-level coordination and scenario generation, and multiple subagents that handle specific analysis tasks. This segmentation allows complex alarm processing to be divided into manageable, specialized functions, improving overall efficiency while maintaining organizational structure.
Solution Approach 2:
The AI-based alarm analysis system performs multiple functions within a unified framework: receiving alarms, generating scenarios, assigning tasks to subagents, collecting results, and making final classifications. This multi-functionality consolidates what would otherwise require separate systems into a single integrated platform, improving productivity without proportionally increasing complexity.
3Ease of operation
If the system automatically classifies alarms, then operator workload is reduced, but the system requires sophisticated AI models and domain knowledge
Solution Approach 1:
The AI complexity is segmented and distributed across multiple agents with specialized domain knowledge. Rather than requiring one monolithic complex model, the system uses multiple smaller, domain-specific subagents (each handling specific scenario domains) coordinated by a primary agent, making the overall system more manageable and easier to operate.
Solution Approach 2:
The system automatically performs the complex AI analysis and classification without requiring operator involvement in the technical decision-making process. Operators simply receive the automated classifications and act accordingly, transferring the complexity burden from human operators to the autonomous AI system.
Data Source
AI summary
An alarm is raised. One of a plurality of Generative Artificial Intelligence (AI) Large Language Model-based autonomous primary agents is activated based at least in part on an alarm type of the alarm and performs an initial analysis of the alarm and creates plausible causes for the alarm. The corresponding Generative AI Large Language Model-based autonomous primary agent autonomously assigns each of the plausible causes to one or more Generative AI Large Language Model-based autonomous subagents that perform an analysis of the assigned plausible cause and returns a result back to the Generative AI Large Language Model-based autonomous primary agent, which classifies the alarm as a false alarm or a true alarm.