Abnormal Characteristic Value Detection Using Local Quality and Dynamics
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for detecting abnormal characteristic values in products manufactured on the same line fail to accurately identify successively-alternate increase/decrease tendencies, leading to unnecessary alarm signals and delayed countermeasure operations, even when characteristic values are within control regions but outside normal regions.
Innovation Solution
A method and apparatus that determine whether a successively-alternate increase/decrease tendency exists in characteristic values and check if at least one value is within a control region but outside a normal region, generating an alarm signal only when both conditions are met, thereby preventing unnecessary countermeasure operations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If alarm signals are generated for all successively-alternate increase/decrease tendencies within the control region, then detection sensitivity is improved, but unnecessary alarm signals increase
Solution Approach 1:
The patent applies local quality by defining different regions (normal region vs. control region) within the overall control limits. The alarm is triggered only when characteristic values are in the control region but outside the normal region, creating a localized detection zone that distinguishes between acceptable variations and actual abnormalities.
Solution Approach 2:
The patent introduces dynamic adjustment of detection thresholds based on the presence of successively-alternate increase/decrease tendencies. When such tendencies are detected, the system dynamically adjusts what constitutes an alarm condition, only alarming when values are outside the normal region even though they remain within the broader control region.
2Loss of time
If alarm signals are generated for all successively-alternate increase/decrease tendencies, then countermeasure operations are performed earlier, but unnecessary countermeasure operations increase
Solution Approach 1:
The patent applies local quality by defining different regions (normal region vs. control region) within the overall control limits. The alarm is triggered only when characteristic values are in the control region but outside the normal region, creating a localized detection zone that distinguishes between acceptable variations and actual abnormalities.
Solution Approach 2:
The patent uses partial action by not triggering alarms for all deviations within the control region, but only for those that fall outside the narrower normal region. This partial monitoring approach reduces unnecessary countermeasure operations while still detecting genuine abnormalities in time.
3Measurement precision
If the normal region is made narrower to improve detection accuracy, then detection precision is improved, but false detection of normal values increases
Solution Approach 1:
The patent introduces dynamic adjustment of detection thresholds based on the presence of successively-alternate increase/decrease tendencies. When such tendencies are detected, the system dynamically adjusts what constitutes an alarm condition, only alarming when values are outside the normal region even though they remain within the broader control region.
Solution Approach 2:
The patent applies preliminary action by first detecting the presence of successively-alternate increase/decrease tendencies before determining whether to generate an alarm. This preliminary detection of the tendency pattern allows the system to distinguish between normal fluctuations and genuine abnormalities, reducing false detections.
Data Source
AI summary
In a method for detecting abnormal characteristic values of at least three products sequentially manufactured in the same manufacturing line, it is determined whether or not a successively-alternate increase/decrease tendency has occurred in a plurality of sequentially-obtained characteristic values of the products. Also, it is determined whether or not at least one of the characteristic values is located within a control region narrower than an allowable region and outside a normal region narrower than the control region. Further, when the successively-alternate increase/decrease tendency has occurred and the at least one characteristic value is located within the control region outside the normal region, an alarm state is detected.


