Operation Monitoring via Camera-Based Action Verification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for monitoring operator actions in manufacturing processes fail to accurately detect inappropriate or inefficient actions, as they rely solely on inspection results of the workpiece, making it difficult to determine quality without specifying the operation content, especially when multiple actions are performed.
Innovation Solution
An operation monitoring method that uses cameras to capture images and moving images of both the operator and workpiece, determining the appropriateness of the operator's actions by comparing them to predetermined procedures, and providing real-time feedback to improve accuracy and efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If quality determination is performed based on inspection results of workpiece appearance, then quality control is simplified, but inappropriate actions (inefficient or dangerous actions) of the operator are overlooked
Solution Approach 1:
The monitoring system segments the evaluation into two independent parts: (1) workpiece appearance inspection and (2) operator action monitoring. By separating these evaluation dimensions, the system can detect both quality issues and inappropriate operator actions independently, resolving the contradiction between simplified quality control and comprehensive operation monitoring.
Solution Approach 2:
The system introduces an intermediary component - the operation monitoring device - that captures images of operator actions and uses AI models to evaluate appropriateness. This intermediary enables comprehensive monitoring without complicating the core quality control process, as it operates in parallel to the appearance inspection.
2Device complexity
If quality determination is performed without specifying operation content, then the process is simplified, but it becomes difficult to determine quality when the operator performs multiple actions
Solution Approach 1:
The system performs preliminary identification of the operation type before conducting quality determination. By first recognizing what operation the operator is performing (e.g., assembly, inspection, adjustment) and then applying the appropriate quality criteria for that specific operation, the system accurately evaluates quality even when multiple different actions are performed, without significantly increasing overall process complexity.
3Measurement precision
If comprehensive monitoring of operator actions is implemented, then operation appropriateness is accurately determined, but system complexity increases
Solution Approach 1:
The system implements feedback by providing operation appropriateness evaluation results back to operators in real-time. This feedback mechanism enables accurate monitoring of operator actions while maintaining relatively simple system architecture, as the same AI model and image capture infrastructure serve both monitoring and guidance functions.
Solution Approach 2:
The operation monitoring device and AI evaluation model serve multiple functions: (1) identifying operation types, (2) evaluating action appropriateness, (3) providing feedback to operators, and (4) recording data for later analysis. This multi-functionality reduces overall system complexity by consolidating multiple monitoring and evaluation capabilities into a single integrated system.
Data Source
AI summary
The operation monitoring method includes the steps of determining whether or not the operator is performing a predetermined action on a workpiece, and performing quality determination of the workpiece when it is determined that the operator is performing the predetermined action.