Quality Control Defect Grouping by Spatial Proximity
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing quality control systems in industries like the aircraft sector face challenges in achieving high detection sensitivity while minimizing false positive detections, leading to increased costs and inspector workload due to the trade-off between sensitivity and false alarm rates, especially in human-machine teamed processes.
Innovation Solution
A method that subdivides potential defects into at least two subgroups based on spatial proximity, allowing for user-adjustable settings to assign these subgroups to individual quality routines, facilitating efficient validation by human inspectors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If detection sensitivity is increased to reduce missed defects, then detection reliability is improved, but the number of false positive detections increases
Solution Approach 1:
The patent segments the quality control process into multiple independent routines with different sensitivity thresholds. The detection system divides potential defects into different groups based on their characteristics, and assigns them to different verification routines. This segmentation allows high sensitivity detection without overwhelming false positives reaching the final decision stage, as each segment is handled with appropriate verification intensity.
Solution Approach 2:
The patent changes the sensitivity parameter dynamically based on the type and characteristics of detected potential defects. Different detection routines operate with different sensitivity thresholds, and the system adjusts which routine processes which defect based on spatial proximity and other characteristics. This parameter change strategy maintains high overall detection reliability while reducing false positives through selective application of sensitivity levels.
2Measurement precision
If manual verification of all detected defects is performed to reduce false positives, then detection precision is improved, but inspection time and costs increase
Solution Approach 1:
The patent applies partial verification action by not manually verifying all detected defects with the same level of scrutiny. Instead, it performs full manual verification only on defects that meet specific criteria (such as spatial proximity thresholds), while other defects are processed through automated routines or require less intensive verification. This partial action approach maintains necessary precision for critical defects while reducing overall inspection time.
Solution Approach 2:
The patent applies different verification quality levels to different defects based on their local characteristics. Defects with certain properties (e.g., specific spatial relationships, sizes, or locations) receive more rigorous verification, while others receive streamlined processing. This local quality approach ensures high verification accuracy where needed without uniformly increasing inspection time across all defects.
3Measurement precision
If algorithm complexity is increased to reduce false positives, then detection precision is improved, but development time and costs increase
Solution Approach 1:
The patent segments the detection algorithm into multiple simpler routines rather than using one complex algorithm. Each routine is designed to be relatively simple and focused on specific defect characteristics, but collectively they achieve high detection precision through coordinated operation. This segmentation reduces individual algorithm complexity and development effort while maintaining overall precision.
Solution Approach 2:
The patent creates a universal detection framework that handles multiple defect types and scenarios through a common architecture. The system uses a standardized set of detection routines that can be applied universally across different inspection contexts, reducing the need for highly specialized complex algorithms for each specific case. This multi-functionality approach simplifies development while maintaining precision across diverse inspection requirements.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A method for a quality control routine of components comprises the following steps: receiving of a data record based on a quality control carried out on a component; detecting of potential defects in the component using the received data record; subdividing the potential defects in the component into at least two subgroups, wherein a user-adjustable setting of a spatial proximity of the potential defects in the component to one another is used for the subdividing into the respective subgroup; assigning the respective subgroups to a respective individual further quality routine so that an efficient quality control routine can be carried out based on the individual assignment.