Industrial Data Model Adaptation for OT Data Semantification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing data from machines and devices in a Smart Factory environment is not normalized and lacks semantic information, making it difficult to identify and utilize effectively, and current tools require laborious manual work and expert knowledge for normalization and semantification.
Innovation Solution
A method and apparatus that automate the normalization and semantification of OT data by using contextual and semantic information from engineering projects, electrical construction diagrams, and vendor-specific frameworks, assisted by generative AI technologies to align pre-structured models with target information models.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If manual normalization and semantification of OT data is performed, then data quality and semantic information are improved, but labor time and complexity increase significantly
Solution Approach 1:
The system enables self-service by allowing the data collection platform to automatically perform normalization and semantification of OT data without requiring manual expert intervention. The platform autonomously identifies data points, adds semantic information, and normalizes data formats using built-in capabilities and vendor-specific frameworks.
Solution Approach 2:
Vendor-specific information models act as intermediaries between raw OT data and the data collection platform. These models serve as a bridge that automatically translates diverse machine data formats into standardized semantic representations, eliminating the need for manual data processing while preserving data quality.
2Measurement precision
If expert tools and manual processes are used for data normalization, then data semantification accuracy is improved, but ease of operation deteriorates due to skill requirements
Solution Approach 1:
The data collection platform performs self-service normalization and semantification automatically, eliminating the need for operators to possess expert knowledge. The system independently identifies data points, determines their semantic meaning, and applies appropriate normalization rules without human intervention.
Solution Approach 2:
Vendor-specific information models serve as intermediaries that encapsulate domain expertise and translation rules. These models automatically map diverse machine data formats to standardized semantic representations, providing accurate semantification while simplifying the operator's task to merely selecting appropriate models.
3Productivity
If diverse machine data formats are collected without normalization, then data collection speed is improved, but interoperability and data usability deteriorate
Solution Approach 1:
The system performs preliminary normalization and semantification actions automatically during the data collection process. By pre-processing data formats and adding semantic information upfront, the system maintains high collection speed while ensuring interoperability and usability of the collected data for various applications.
Solution Approach 2:
The data collection platform implements universal normalization capabilities that handle diverse machine data formats through vendor-specific information models. This multi-functional approach enables the system to collect, normalize, and semantify data from different sources simultaneously, maintaining both speed and interoperability.
Data Source
Figure 1
Figure 2
AI summary
The described solution empowers OT engineers, system integrators etc to enhance and adapt existing data models e. g. in the Brownfield of an industrial plant, for newly introduced data, by reducing the skill gap to develop industrial information models, which accelerates IT-OT integration. Moreover, it also increases the adoption of relevant standards. Further on, the solution also empowers the IT engineers on the IT layer to efficiently use the OT data uniformly and provide valuable data insights over it.