Measurement Data Distribution Selection for Full-Range Fit

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing statistical modeling methods often fail to accurately describe measurement data over the entire value interval, leading to modeling errors when data points fall outside the distribution range, as they assign zero frequency to values outside the interval, making it difficult to predict such occurrences.

Innovation Solution

A method is proposed to evaluate a random sample of measurement data by checking if the initial statistical distribution is suitable for the entire value interval, determining the moments of skewness and kurtosis, and selecting a suitable statistical distribution from a defined set that can describe the frequency of measurement data values within the entire value interval.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If a statistical distribution is created exclusively from existing measurement data, then the distribution fits the data well over a large part of the value interval, but values outside the distribution range cannot be predicted and modeling errors occur

Engineering Contradiction:
Improvedistribution fit qualityVSAvoidpredictive capability
Core Design Contradiction:
Measurement precisionVSReliability

Solution Approach 1:

The patent applies preliminary action by performing a suitability test before finalizing the statistical distribution model. The method checks whether the distribution can describe measurement data over the entire value interval, including extreme values, before accepting it for predictive purposes. This preliminary validation prevents selecting distributions that would fail to predict out-of-range values later.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent implements feedback by using the suitability test results to iteratively refine the statistical distribution selection. The test provides feedback on whether the distribution adequately covers the entire value interval, allowing the selection process to adjust and improve the model until it meets the required predictive capability for the full range of possible measurements.

Inventive Principle:
Principle #23Feedback

2Ease of manufacture

If a statistical distribution assigns zero frequency to values outside its interval, then the distribution is mathematically well-defined, but it cannot predict or describe occurrences outside this interval

Engineering Contradiction:
Improvedistribution definition simplicityVSAvoidvalue range coverage
Core Design Contradiction:
Ease of manufactureVSAdaptability or versatility

Solution Approach 1:

The patent performs a preliminary suitability test to verify that the distribution's value interval adequately covers the entire range of possible measurement data before finalizing the model. This advance check ensures the distribution is both mathematically sound and sufficiently versatile for the application.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent introduces dynamics by making the distribution selection adaptive rather than static. The suitability test allows the method to evaluate and select distributions based on their actual performance in covering the measurement data range, rather than relying solely on theoretical mathematical properties. This dynamic approach balances mathematical definition with practical adaptability.

Inventive Principle:
Principle #15Dynamics

3Ease of operation

If a normal distribution is used to model measurement data, then the modeling process is simple and well-established, but it cannot accurately describe data with limited value intervals or asymmetric distributions

Engineering Contradiction:
Improvemodeling simplicityVSAvoiddistribution accuracy
Core Design Contradiction:
Ease of operationVSMeasurement precision

Solution Approach 1:

The patent applies parameter changes by moving beyond the fixed parameters of normal distribution to evaluate multiple distribution types with different parameters. The suitability test framework allows changing the distribution family (e.g., from normal to asymmetric or bounded distributions) based on the actual characteristics of the measurement data, thereby improving accuracy while maintaining systematic evaluation.

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The patent uses feedback from the suitability test to guide the selection of appropriate distribution parameters and types. The test results provide information about which distributions accurately represent the measurement data, allowing the method to select the most appropriate model rather than defaulting to normal distribution, thus improving precision while maintaining operational clarity.

Inventive Principle:
Principle #23Feedback

Data Source

PatentEP3916495A1Method and arrangement for evaluating or preparing an evaluation of a sample of measurement data from a measurement of a plurality of workpieces
Publication Date: 2021.12.01 CARL ZEISS INDUSTRIELLE MESSTECHNIKE GMBH
  • EP3916495A1 patent drawingFigure 1
  • EP3916495A1 patent drawingFigure 2
  • EP3916495A1 patent drawingFigure 3

AI summary

The invention relates to a method and an arrangement for evaluating a sample of measurement data from a measurement of a plurality of workpieces, wherein a system of statistical distributions exists or is established which is capable of describing a frequency of measurement data values ​​as a function of the measurement data values, wherein instances of the system of statistical distributions are distinguishable from one another by a moment value of two moments, namely a skewness and a kurtosis, of the respective statistical distribution, and wherein - for a value interval of the measurement data, which is a predetermined value interval or a value interval of the measurement data actually occurring in the sample, a set of all those statistical distributions is defined from the system of statistical distributions (step 5) which are capable of describing a frequency of measurement data values ​​in the entire value interval.- from the sample of measurement data, a moment value for skewness and kurtosis is determined according to a first statistical distribution (step 7), - based on the determined moment values, it is checked whether a statistical distribution exists in the defined quantity that exhibits the determined moment values ​​for skewness and kurtosis, and a corresponding test result is generated (step 9). The invention also relates to a method and an arrangement for preparing an evaluation of a sample of measurement data.