Product Inspection Risk Calculation for Unknown Distributions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing product inspection devices face challenges in accurately calculating consumer risk and producer risk, especially when the distribution of products deviates from known distributions, leading to incorrect classification of conforming and defective articles.
Innovation Solution
A product inspection device that measures characteristic values, calculates measurement standard deviation, and determines conformity based on a product standard, while also calculating and adjusting consumer and producer risks by accounting for measurement variations, allowing for accurate risk assessment independent of product distribution.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the inspection standard is set more rigorous than the product standard, then the consumer risk is reduced, but the conforming article rate decreases
Solution Approach 1:
The system dynamically changes the inspection standard parameters (upper and lower limit values) based on calculated risk values. By adjusting these parameters according to actual measurement variations and risk calculations, the system achieves both high consumer protection and maintained conforming article rates, resolving the contradiction between reliability and productivity.
Solution Approach 2:
The inspection standard is made dynamic rather than fixed. The system adjusts the inspection standard based on calculated consumer risk and producer risk, allowing the standard to adapt to actual measurement conditions. This dynamic approach enables the system to maintain high consumer risk reduction while preserving conforming article rates that would otherwise be lost with a rigid rigorous standard.
2Measurement precision
If the distribution of products is assumed to be a known distribution (normal, triangular, rectangular), then the conforming article distribution can be estimated, but accuracy is low when the actual distribution deviates from the assumed distribution
Solution Approach 1:
The patent extracts the dependency on assumed product distributions by directly calculating consumer risk and producer risk based on measurement variations without requiring assumptions about the underlying product distribution. This extraction of the distribution assumption eliminates the accuracy problems that arise when actual distributions deviate from assumed distributions.
Solution Approach 2:
The patent introduces measurement variations as an intermediary concept that mediates between the inspection standard and the actual product distribution. By calculating risks based on measurement variations rather than directly assuming product distributions, the system achieves accurate risk assessment without needing to know or assume the actual product distribution type.
Data Source
AI summary
A product inspection device and method for correctly calculating consumer and producer risks irrespective of the type of distribution of products. A characteristic value representing a predetermined product characteristic is measured for each product as a product measurement value, and a standard deviation of measurement variations in measurement results is calculated as a measurement value standard deviation. The products are determined to be conforming based on whether the measured product measurement value falls within a range of a product standard. Consumer and producer risks are calculated based on the measurement variations. The calculated consumer and producer risks are respectively and successively added up and it is determined whether the number of products having undergone the adding have reached a predetermined number. If so, the added up consumer risk and producer risk are divided by the number of products to calculate a final consumer risk and a final producer risk.