Recently, there have been JMPer What preparations should we make before consulting and implementing experimental design? The extension of this topic is actually very large, and everyone’s self-questioning and self-answering tend to point to measurement system analysis. (Measure System Analysis, MSA) . After thinking about it, I discovered that what everyone wants to emphasize is not so much the analysis of the measurement system, as it is the guarantee of data quality. In general, the data is the result of the measurement. Therefore, the validity of the experimental data naturally needs to pass the measurement system. Effectiveness is guaranteed. But there is actually no inevitable causal relationship between the two. To put it another way, measurement system analysis is a basic and routine task, regardless of whether experimental design is carried out, but when you want to use data to drive quantitative decision-making, shouldn't the effectiveness of the measurement system be guaranteed by evaluation, verification and confirmation? ?
As a result, another more basic question emerged, namely: how to pass JMP Software to perform measurement system analysis? In this article, we will provide you with some guidelines for getting started.
First of all, some users use JMP get on MSA One of the reasons for the slight confusion may be JMP for MSA With its multi-platform support, sometimes it provides more choices, but it is more difficult to choose. We are right here to clarify.
In the current JMP There are mainly two supports MSA Functional platform ( Figure -1) ,they are, respectively:
- Analysis> Quality and process > Measurement system analysis
- Analysis> Quality and process > Variability / Counting and drawing
Figure-1 JMP for MSA Multi-platform support
For the "measurement system analysis" platform, it is first based on EMP (Evaluating the Measurement Process) Method to perform measurement system analysis. The concept was not proposed for a long time, it was developed by the United States SPC expert Donald J. Wheeler PhD in 1984 Years in his works Evaluating the Measurement Process First proposed in JMP The method cited is derived from its 2006 Published in EMP Ⅲ Using Imperfect Data (2006) In a book ( Figure -2) Explained content and steps.
Since this method is not yet popular in China, and few people know it, therefore, in the JMP that we have contacted Among the user groups, the usage rate is low, so this article will not go into details for now. In addition, the platform actually provides traditional " Gage R&R "The entrance to the analysis method, but this / The effect of entering the platform of "Counting Quantity Chart" is the same, so please refer to the following directly.
Figure-2 EMP Ⅲ Using Imperfect Data (2006)
The current mainstream measurement system analysis ideas in China are mainly based on the Automotive Industry Action Group (AIAG) The "Measurement System Analysis Reference Manual" is continuously updated.
Under the guidance of this policy, 6 There are two conventional dimensions to evaluate the measurement system. They are stability (stability) Bias (bias) , Linear (linearity) , Resolution (discrimination) , Repeatability (repeatability) And reproducibility (reproducibility) .
Among them, as shown in Figure-3 As shown, bias is the accuracy of the measurement system (accuracy) The most direct measurement of the measurement result reflects the difference between the average value of the measurement result and the true value; the repeatability and reproducibility are the accuracy of the measurement system (precision) The metrics reflect the fluctuation of the measurement results. They are the most critical indicators for evaluating the pros and cons of the measurement system. So, for "variability / For the "Counting Quantity Map" platform, most of its function items are related to this.
image 3 Bias and accuracy diagram
For the majority of Chinese users, in the above 6 Among these characteristics, the most familiar ones are "repeatability" and "reproducibility", the so-called Gage R&R , Almost all teaching materials, trainings and projects related to Six Sigma management are inevitably touched, so that many people take Gage R&R Almost equivalent to MSA , And this is the focus of this article.
Here, "repeatability error" mostly refers to the same operator, using the same set of measuring equipment, to measure the same characteristics of the same component multiple times in a short time interval, and the consistency of the results obtained, that is, to emphasize all the measurement results Fluctuations that are all under the same conditions as possible, also called equipment variation (EV) In contrast, the "reproducibility error" emphasizes that the same characteristics of the same measurement component are measured multiple times under various measurement conditions that may change, and the results obtained are consistent. In the traditional case, reproducibility Errors are mainly caused by different operators, also known as personnel fluctuations (appraiser variation, AV) .
These two kinds of fluctuations are usually expressed by variance or standard deviation. The smaller the fluctuation, the better the measurement system. The traditional analysis method is often combined with the fluctuation of the production process, and it is realized by the variance analysis considering the random effect factor. The process of variance decomposition is shown in Figure-4 Shown.
Figure-4 Gage R&R Schematic diagram of traditional analysis methods
In the end, two important indicators for evaluating the accuracy of the measurement system are derived from this:
Among them, %Gage R&R Focus on evaluating the measurement effect of the measurement system on the overall process variation, and emphasize the resolution ability of the measurement system to improve the production process;
%P/T Focus on evaluating the measurement effect of the measurement system against product specifications, and emphasize the ability of the measurement system to distinguish tolerance limits.
%Gage R&R with %P/T It is two different aspects to evaluate the performance of the measurement system. A good measurement system should make these two indicators small enough at the same time. -1 It is the general reference and judgment standard for these two metrics.
Table-1 Gage R&R Reference judgment standard
So far, if you can understand the above analysis ideas and logic, then, in JMP With the help of modern statistical analysis software, you only need to collect and organize Gage R&R Analyze the required observation data, enter the software, and leave all the subsequent calculation processes to the software.
Next, we will use a simple case to discuss Gage R&R The operation process of the traditional analysis method is demonstrated, and the output result is briefly analyzed.
"Using a vernier caliper to measure the shaft diameter of a machined part, the known tolerance requirement is 45.5±0.5 , In order to analyze the accuracy of the measurement system:
- Representative draw 10 Parts and numbered
- Random Pick 3 Inspectors measure each part with their own calipers
- Disrupt the order of the parts and test the second Times
- Disrupt the order of the parts and test the third Times
- The measurement process schematic and the record data table are shown in Figure-5 Shown"
Figure 5 Schematic of the measurement process and record data sheet
JMP Gage R&R Analysis steps:
- Combined with JMP Data management related platforms, easily import and organize data as shown in the figure -6 Data table structure shown
Figure-6 Gage R&R Data entry and sorting
- According to Figure-7 Set, start "analysis > Quality and process > Variability / Counting and Drawing” platform
Figure-7 "Variability / Start the platform setting
- Figure-8 As shown, select the red triangle next to "Variability Gauge" > Measuring Tool Research > Measuring tool RR , Enter the tolerance interval, for this example 46.0-45.0=1
Figure-8 Start "Gage RR "analysis
Measuring Tool R&R Brief analysis of analysis output report:
Figure-9 Measuring tool R&R Analysis output report
From Figure-9 Gage shown R&R From the output report, the repeatability variance component is approximately 0.0011 , The reproducibility variance component is approximately 0.0035 , Two important indicators for evaluating the accuracy of measurement systems %Gage R&R=33.56% , %P/T=40.56% , Both greater than 30% Therefore, the accuracy level of the measurement system is not high, and measures should be taken to improve it, and improvement can be given priority to reducing reproducibility errors.
Further analysis, you can also find that in the reproducibility variation, "inspector* Component variation (0.3116)> "Inspector" variation (0.1708) , Which is related to the "measured value" variability graph and the mean "measured value" - Inspectors *" part " Interaction diagram ( Measuring Tool Research > Measuring tool RR Figure > Mean plot ) Presented in 3 Inspector's 4 , 5 , 10 The measurement results of the number components are consistent with the phenomenon of large differences.
Therefore, if in-depth study of the features of these components is carried out in the measurement standard operation training, it is possible to reduce the reproducibility error and improve the accuracy of the measurement system.
Figure-10 Variability graph of "measured value" and mean value of "measured value" - Inspectors *" part " Interaction diagram
The above is what we are going to introduce in this issue about MSA Introductory content, actually about MSA There are a lot of basic knowledge, such as the measurement system analysis of destructive experiments, the measurement system analysis of counting data, the measurement system analysis of using standard parts or fixed parts, and how to evaluate the measurement system when the two accuracy indicators are inconsistent. and many more. These contents are related to the operation of the software on the one hand, and on the other hand, we need to start with the methodology itself. We will discuss with you in subsequent articles.
