Production Order Status Monitoring Using IoT Model Matching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In factory operating systems, manually labeling and structuring data for AI processing is labor-intensive and inefficient, particularly for legacy devices with limited data interfaces, which can lead to errors in monitoring and optimizing production processes.
Innovation Solution
A method and apparatus that generate production and product IoT models from scheduling and design documents, respectively, and use data from automation control systems to automatically learn and match processing steps, reducing the need for manual data marking and enhancing production status monitoring.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual data marking and structuring is performed to train AI algorithms, then the system can recognize specific scenarios, but it requires an enormous amount of work and time
Solution Approach 1:
The system performs self-service by automatically generating structured data from unstructured production documents using NLP technology. The algorithm automatically extracts entities, relationships, and attributes without requiring manual marking, thereby resolving the contradiction between recognition accuracy and time consumption
Solution Approach 2:
The patent replaces the mechanical manual marking process with an automated NLP-based system. The algorithm automatically structures unstructured text data into machine-readable formats, substituting human labor with intelligent automation to achieve both accuracy and efficiency
2Reliability
If engineers manually investigate and observe to mark and structure data, then accurate production data can be obtained, but it takes weeks to communicate with operators and managers
Solution Approach 1:
The patent replaces the mechanical process of manual investigation and communication with an automated NLP system that directly processes production documents. The algorithm extracts and structures data automatically, eliminating the need for engineers to spend weeks communicating with operators and managers while maintaining data accuracy
Solution Approach 2:
The NLP algorithm acts as an intermediary between unstructured production documents and the automated control system. It automatically translates document content into structured data formats, serving as a mediator that eliminates the need for manual intervention while ensuring accurate data transmission
3Adaptability or versatility
If legacy devices with limited data interfaces are used, then existing factory equipment can be utilized, but they cannot provide abundant information for product process monitoring and optimization
Solution Approach 1:
The patent introduces an intermediary layer (the NLP-based data extraction system) that bridges legacy devices and the monitoring system. By processing production documents associated with legacy devices, the system extracts additional information that compensates for the limited interfaces of legacy equipment
Solution Approach 2:
The system creates a digital copy or representation of production data from documents associated with legacy devices. This virtual copy contains structured information that supplements the limited data available from legacy device interfaces, enabling comprehensive monitoring without modifying the original devices
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Provided are a method, apparatus, electronic device, medium, and program product for monitoring status of production order, relating to the technical field of Internet of Things (IoT). The method comprises: generating at least one production IoT model on the basis of a production scheduling system document, the production IoT model at least comprising process attributes of product processing; generating at least one product IoT model on the basis of a product design specification document, the product IoT model also comprises at least process attributes of product processing; for a production IoT model, finding a product IoT model having the same process attributes and performing association; learning the data of the production device collected by the data collection automation control system in the factory to obtain a data model representing the processing steps of the product; matching the processing steps with the process attributes in the product IoT model, and determining the production order status of the factory on the basis of the match result.