Industrial Asset Data Retrieval via Semantic Graph Mapping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Industrial assets' lifecycle data objects are stored in disparate data models, making it difficult for users to query and utilize them due to the need for familiarity with various proprietary data models and siloed information.
Innovation Solution
A system and method for retrieving data objects from industrial assets by mapping queries to types of data objects using a graph that represents the asset and semantic associations, enabling abstraction from data source intricacies and facilitating interoperability across different data models.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If data objects are stored in proprietary data models from multiple vendors, then each software package can maintain its own data structure and semantics, but users cannot easily query asset information without familiarity with each specific data model
Solution Approach 1:
The patent introduces an intermediary layer (semantic model with graph database) that sits between the proprietary data models and user queries. This intermediary translates diverse vendor-specific data models into a unified semantic representation, allowing users to query asset information without needing to understand individual proprietary formats. The semantic model acts as a mediator that reconciles different data structures while preserving their meanings.
Solution Approach 2:
The system segments the data architecture into distinct layers: proprietary data model layer, semantic model layer, and query interface layer. Each layer handles specific concerns independently, allowing the proprietary layers to maintain their formats while the semantic layer provides unified access. This segmentation isolates complexity within each layer while enabling simplicity at the user interface.
2Reliability
If data objects are stored in siloed proprietary formats, then each software package maintains data integrity and semantics, but the data cannot be easily integrated or shared across different systems
Solution Approach 1:
The semantic model serves as an intermediary that preserves the integrity of source data while enabling interoperability. It maps concepts from different proprietary data models to a unified semantic framework, maintaining the original data's meaning and structure in the source systems while providing standardized access across systems. This allows data to remain reliable in its native format while becoming adaptable for cross-system use.
Solution Approach 2:
The semantic model provides a universal framework that can represent data from multiple proprietary sources with different structures and semantics. It defines common asset concepts and relationships that work across diverse data models, enabling the same semantic model to handle XML, JSON, SQL, and other formats uniformly. This universality allows a single interface to serve multiple data sources with different characteristics.
3Ease of operation
If a unified data model is imposed across all lifecycle phases, then querying becomes simpler and more consistent, but the proprietary data structures and semantics of individual software packages are lost or distorted
Solution Approach 1:
The semantic model acts as a lossless intermediary that captures the semantics of proprietary data models without imposing a rigid unified structure on the source data. It uses flexible mapping mechanisms that preserve the meaning and relationships defined in each vendor's data model while presenting a consistent query interface. This allows simplified querying without distorting or losing the original data semantics.
Solution Approach 2:
The system applies local quality by allowing each proprietary data model to maintain its own structure and semantics locally, while the semantic model provides global consistency for querying. Each data source can have its own optimized structure for its specific purposes, while the semantic layer ensures these local variations can be accessed through a unified interface without losing their individual characteristics.
Data Source
AI summary
A computer-implemented method for retrieving data that relates to at least one industrial asset includes the steps of: receiving, from a requesting entity, a query for data objects from one or more lifecycle phases of the at least one industrial asset; mapping the query to one or more types of data objects that are available from one or more given data sources relating to the at least one industrial asset; obtaining, from the one or more data sources, one or more data objects of the one or more types; producing, from the one or more data objects, a response to the query; and transmitting the response to the requesting entity.