Industrial Knowledge Graph for Event Detection and Remedies
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Industrial automation systems face challenges in effectively contextualizing large amounts of diverse data and building intelligence or knowledge to enable analysis, prediction, and decision-making, as they often receive limited types of data, hindering the identification of events and determination of remedies within these systems.
Innovation Solution
A unified data system that applies an asset model to industrial data from various sources, including machine, human, and enterprise data, to contextualize and generate a knowledge graph, using machine learning and generative AI to identify events, causes, and provide remedies, thereby enabling the visualization of relationships and connections within the industrial automation system.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If large amounts of diverse industrial data are collected from multiple sources, then the quantity and variety of data increases, but the difficulty of contextualizing and building intelligence on the data increases
Solution Approach 1:
The system segments the complex contextualization task into distinct processing stages: data collection from multiple sources, data contextualization using asset models, event determination, and knowledge graph generation. Each stage handles specific aspects of the data, making the overall complex process more manageable and systematic.
Solution Approach 2:
The patent introduces intermediate structures including asset models that serve as mediators between raw industrial data and meaningful insights, event data as an intermediary representation of system states, and a knowledge graph as an intermediary that connects contextualized data to actionable intelligence. These intermediaries bridge the gap between diverse data sources and decision-making processes.
2Device complexity
If limited types of data (e.g., only machine data) are received, then the data processing system is simpler, but the ability to identify events and determine remedies is reduced
Solution Approach 1:
The system is designed to universally process multiple types of industrial data including machine data, human data, and enterprise data through a unified architecture. The asset models and knowledge graph structure are generic enough to handle diverse data types while providing specific analytical capabilities for each data category, enabling comprehensive event identification and remedy determination.
3Measurement precision
If contextualized data is used to build intelligence, then the quality of analysis and decision-making improves, but the time and computational resources required increase
Solution Approach 1:
The system performs preliminary contextualization of industrial data using asset models before full analysis is required. By pre-processing and structuring data into contextualized formats and building the knowledge graph in advance, the system reduces the computational burden and time required for real-time event identification and remedy determination.
Data Source
AI summary
A system includes processing circuitry and a memory, accessible by the processing circuitry, the memory storing instructions that, when executed by the processing circuitry, cause the processing circuitry to perform operations including receiving industrial data collected from an industrial automation system during performance of an industrial automation process and applying an asset model to the industrial data to contextualize the industrial data. When executed, the instructions also cause the processing circuitry to perform operations including determining event data based on the industrial data, generating a knowledge graph based on the industrial data and the event data, identifying an event based on the event data, and providing for display via a graphical user interface (GUI), one or more remedies for the event based on the knowledge graph.