Quality Management Apparatus for Production Facility Defect Identification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods for defect detection in production facilities, such as surface mounting lines, fail to accurately identify the specific facility member causing defects, leading to inefficient maintenance and quality management, as they cannot distinguish between defects caused by individual components or processes like solder printing and reflow.
Innovation Solution
A quality management apparatus that monitors and records observation values during product processing, using a causal factor estimator to compare defective and non-defective observation values, thereby identifying operation abnormalities that may cause defects by analyzing significant changes in these values, and providing information for precise maintenance actions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If conventional defect detection methods are used in production facilities, then defect detection is automated, but the specific facility member causing defects cannot be identified
Solution Approach 1:
The invention segments the production facility into multiple facility members (e.g., nozzles, cassettes, conveyors) and assigns unique identifiers to each. By dividing the overall defect detection problem into facility-member-specific sub-problems, the system can trace defects back to individual components rather than treating the entire production line as a single unit.
Solution Approach 2:
The invention implements a feedback mechanism where defect information from inspection apparatus is fed back to the facility member management system. This feedback loop enables the system to correlate specific defects with specific facility members by comparing facility member IDs recorded during processing with defect locations, thereby identifying the actual cause of defects.
2Loss of information
If facility member information is collected and analyzed, then defect location clues are provided, but conclusive evidence of the defective component is not obtained
Solution Approach 1:
The invention performs preliminary actions by recording facility member IDs and processing conditions before defects occur. By pre-collecting and storing the correspondence between facility members and processed products, the system is prepared to immediately identify the defective facility member when a defect is detected, eliminating the need for trial-and-error investigation.
Solution Approach 2:
The invention introduces an intermediary data structure (facility member ID recording mechanism) that bridges the gap between facility operations and defect detection. This intermediary system records which facility member processed which product, enabling precise tracing from defect back to the specific facility member without requiring direct observation or expert analysis.
3Reliability
If trial-and-error methods are used to identify facility abnormalities, then expert knowledge is required, but maintenance time and complexity increase
Solution Approach 1:
The invention enables the production facility system to self-diagnose defects by automatically correlating defect information with recorded facility member data. Instead of requiring expert intervention and trial-and-error methods, the system autonomously identifies the defective facility member by comparing inspection results with pre-recorded processing information, significantly reducing maintenance time and eliminating the need for expert knowledge.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A production facility including a function of monitoring an observation value for plural observation items in order to detect an operation abnormality of an own quality management apparatus and an inspection apparatus that inspects a product processed with the production facility are provided in a production line. The quality management apparatus includes: an observation data acquisition unit configured to acquire the observation data in which an observation value of each observation item observed with the production facility during the processing of each product; and a causal factor estimator configured to compare the observation value during the processing of a defective to the observation value during the processing of a nondefective to determine whether the operation abnormality of the production facility occurs using the observation data, the operation abnormality being able to become a causal factor, when the inspection apparatus detects a defect in a certain product.