Industrial Knowledge Graph Contextualization for Multi-Asset Insights
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional data analytics and digital transformation of asset data in industrial environments are inefficient due to the difficulty in identifying and fixing issues across multiple assets, the challenge of determining interrelationships between data from various systems and applications, and the limited time spent on modeling data for insights, leading to suboptimal use of computing resources.
Innovation Solution
A knowledge graph system is employed to construct relationships between industrial assets, using configuration rules and data sources to provide insights and optimize integrity operating windows, with a cognitive advisor that offers real-time recommendations and notifications, and a dashboard visualization for easy troubleshooting.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If traditional data analytics methods are used for asset management, then human interaction and manual analysis are required, but this leads to inefficiency and difficulty in identifying issues across multiple assets
Solution Approach 1:
The patent introduces an intermediary system comprising a knowledge graph and contextualization engine that automatically processes and interrelates data from multiple asset management systems. This intermediary layer transforms raw data into contextualized insights, enabling automated issue identification across assets without requiring manual human analysis, thus resolving the contradiction between maintaining comprehensive analysis and improving efficiency.
Solution Approach 2:
The patent replaces the mechanical system of manual human analysis with an automated computational system. The knowledge graph infrastructure and contextualization algorithms automatically perform data aggregation, relationship determination, and issue identification tasks that traditionally required human specialists, thereby dramatically improving productivity while reducing time loss.
2Loss of information
If data from multiple systems and applications is integrated to analyze asset relationships, then comprehensive insights can be obtained, but determining interrelationships becomes difficult and time-consuming
Solution Approach 1:
The patent implements preliminary action by pre-establishing a knowledge graph infrastructure that defines schemas, relationships, and contextualization rules before data integration occurs. This pre-configured framework enables automated mapping and relationship determination when data from multiple systems is ingested, eliminating the time-consuming manual process of determining interrelationships while maintaining complete information from all data sources.
3Productivity
If limited time is spent on modeling data for insights, then computing resources are saved, but the quality and usefulness of data analytics deteriorates
Solution Approach 1:
The patent implements self-service through automated contextualization algorithms that autonomously process raw data and generate actionable insights without requiring extensive manual modeling efforts. The system automatically applies contextualization rules, performs data enrichment, and generates recommendations, thereby maintaining high-quality analytics while minimizing the time and computing resources spent on manual data modeling activities.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Various embodiments described herein relate to providing and/or employing a knowledge graph related to one or more assets. In this regard, a request to generate knowledge graph data related to one or more assets is received. The request includes an asset descriptor describing the one or more assets. In response to the request, aggregated operational technology data is obtained based on the asset descriptor and from one or more data sources associated with the one or more assets. Furthermore, the aggregated operational technology data is contextualized, based on configuration data for the one or assets and a set of contextualization rules for the one or more data sources, to generate the knowledge graph data. The knowledge graph data is also allocated within a knowledge graph data structure configured for the one or more assets.