How to classify measuring fixtures, and what aspects and methods of measuring system analysis are carried out according to these classifications.

MSA (Measurement System Analysis) concept

In daily production, we often analyze the state of the process, the ability of the process and monitor the changes of the process according to the measured data of the process parts. So, how to ensure that the analysis results are correct? We must ensure the accuracy/quality of the measured data from two aspects. First of all, we should use MSA method to evaluate the measurement system that obtains measurement data. The second is to ensure the use of appropriate data analysis methods, such as SPC tools, experimental design, variance analysis, regression analysis, etc.

MSA(MeasurementSystemAnalysis) uses mathematical statistics and charts to analyze the resolution and error of the measurement system, evaluate whether the resolution and error of the measurement system are compatible with the measured parameters, and determine the main components of the measurement system error.

The error of measurement system is characterized by statistical characteristics: deviation and variance of measurement data of measurement system running under stable conditions. Bias refers to the position of measured data relative to the standard value, including bias, linearity and stability); Measuring system; Variance refers to the dispersion degree of measurement data, also known as R &；; R, including the repeatability and reproducibility of the measurement system.

Generally speaking, the resolution of the measurement system should be one tenth of the change in the process of obtaining the measured parameters. The deviation and linearity of the measuring system are determined by the calibration of measuring tools. The stability of the measuring system can be monitored by repeatedly measuring the average range control chart of the same quality characteristics of the same parts. The repeatability and reproducibility of the measurement system are determined by Gager & R research.

The data used for analysis must come from the measurement system with appropriate resolution and measurement system error, otherwise, no matter what analysis method we adopt, it may eventually lead to wrong analysis results. In ISO 100 12-2 and QS9000, there are corresponding requirements for the quality assurance of the measurement system, requiring enterprises to have relevant procedures to verify the effectiveness of the measurement system.

The measurement system is characterized by F and S, and its evaluation methods include small sample method, dual method and linear method.

The basic content of MSA

Data is obtained through measurement, and measurement is defined as: assigning measurement to specific things to express their relationship about special characteristics. This definition was first given by C.Eisenhart. The assignment process is defined as the measurement process, and the assignment value is defined as the measurement value.

As can be seen from the definition of measurement, in addition to specific things, there should also be measuring tools, qualified operators who use measuring tools, prescribed operating procedures, and some necessary equipment and software involved in the measurement process, and then combine them to complete the assignment function and obtain measurement data. Such a measurement process can be regarded as a data manufacturing process, and the data it produces is the output of the process. This measuring process is also called measuring system. Its complete description is a collection of instruments or measuring tools, standards, operations, fixtures, software, personnel, environment and assumptions used for quantitative measurement or qualitative evaluation of measured characteristics. The whole process of obtaining measurement results is called measurement process or measurement system.

As we all know, measurement is one of the six basic quality factors (people, machines, materials, operation methods, measurement and environment) that affect the change of product quality characteristic values. Different from the other five basic quality factors, under the comprehensive action of the five basic quality factors, the influence of measurement factors on the characteristic value of process quality has nothing to do with the machining process, which makes it possible to study the measurement system independently. And correct measurement is always the first step of quality improvement. If there is no scientific evaluation method and effective control of measurement system, the basic premise of quality improvement will be lost. Therefore, measurement system analysis has become the only way for enterprises to achieve continuous quality improvement.

In recent years, measurement system analysis has gradually become an important work of enterprise quality improvement, and both business and academic circles have paid enough attention to measurement system analysis. Measurement system analysis has also become one of the elements of the quality system QS9000 of the three major American automobile companies, and it is also an important content of the 6σ quality plan. At present, the 6σ continuous quality improvement planning model represented by General Electric (GE) includes: definition, measurement, analysis, improvement and control, which is abbreviated as DMAIC.

From the point of view of statistical quality management, the analysis of measurement system essentially belongs to the category of variation analysis, that is, to analyze the variation caused by the measurement system relative to the total variation of the process, so as to ensure that the main variation of the process comes from the process itself, not the measurement system, and the measurement system capacity can meet the requirements of the process. The analysis of the measurement system aims at the stability and accuracy of the whole measurement system, and it is necessary to analyze the position change and width change of the measurement system. The position change includes the deviation, stability and linearity of the measurement system. Width variation includes repeatability and reproducibility of the measurement system.

Measurement systems can be divided into two types: counting type and metering type. The measurement system can give specific measurement values after measurement; Counting measurement system can only give qualitative measurement results. The analysis of "metrology" measurement system usually includes deviation, stability, linearity, repeatability and reproducibility (R&R for short). In the actual operation of measurement system analysis, it can be carried out simultaneously or selectively, depending on the specific use.

The analysis of "counting" measurement system usually adopts hypothesis testing analysis method to make a decision.

Statistical characteristics of MSA

1. The measuring system must be under statistical control, which means that the changes in the measuring system can only be caused by common reasons rather than special reasons. This can be called statistical stability.

2. The change of measuring system must be less than the change of manufacturing process.

3. The variation should be less than the tolerance zone.

4. The measurement accuracy should be higher than the process deviation and tolerance zone. Generally speaking, the measurement accuracy is one tenth of the process deviation and tolerance zone.

5. The statistical characteristics of measurement system may change with the change of measurement items. If so, the maximum variation of the measuring system should be less than the smaller value in the process variation and tolerance zone.

Indicators of management transaction agreement

1. Repeatability of measuring tools: refers to the change of measured values (data) obtained when the same evaluator uses the same measuring tool to measure the same characteristics of the same part for many times.

2. Measuring tool reproducibility: refers to the change of average measurement value when different evaluators measure the same characteristics of the same part with the same measuring tool.

3. Stability: refers to the total variation of the measured value obtained when the measuring system measures the single characteristic of the same benchmark or part within a certain duration.

4. Bias: refers to the difference between the average value obtained by the same operator using the same measuring tool to measure the same characteristic of the same part for many times and the average value obtained by measuring the same characteristic of the same part with a more precise instrument, that is, the difference between the observed average value of the measurement results and the reference value, which is what we usually call "accuracy".

5. Linearity: refers to the deviation change of the measuring system within the expected working range.

MSA time series

1). The newly produced product PV is different;

2). New instruments are different from electric vehicles;

3). New operators have different AV;

4). Wear-prone instruments must pay attention to their analysis frequency.

Analysis of1.r&

Determine the research object of main variation forms.

Use the method of "full scale and average" or "analysis of variance" to analyze the measuring tools.

The random sampling of the tested materials in the process should be a unified process.

Select 2-3 operators to measure 65,438+00 parts without knowing it. Testers will record the data read by operators to study its repeatability and reproducibility (operators should be familiar with and understand the general operating procedures to avoid affecting the reliability of the system due to inconsistent operation), and evaluate the proficiency of measuring tools to different operators.

The accuracy of measuring tools used for important characteristics (especially those specified by special symbols) should be110 of the tolerance of the measured article (that is, the minimum scale should be able to read110, in which the process is poor or the specification tolerance is small; If the reading accuracy of measuring tools required in the process is 0.0 1m/m, the measuring accuracy should be 0.00 1m/m) to avoid the lack of resolution of measuring tools, and the measuring accuracy for general characteristics should be 1/5 of the tolerance of the measured object.

After the test, the tester will calculate the repeatability and reproducibility data of the measuring tool, as shown in Appendix 1 (R&; R data sheet), Annex II (R&; R analysis report), according to the formula to calculate and make a -R control chart or directly use the table calculation.

2. Result analysis

1) When the variation value of repeatability (AV) is greater than reproducibility (EV):

The structure of measuring tool needs to be strengthened in design.

It is necessary to improve the way of clamping measuring tools or positioning parts (inspection points).

Measuring tools should be maintained.

2) When the variation value of reproducibility (EV) is greater than repeatability (AV):

Operators should strengthen the education on the operation methods and data reading methods of measuring tools, and the operation standards should be clearly formulated or revised.

Some fixtures may be needed to help the operator use the measuring tool more consistently.

After the measuring tools and fixtures are delivered to the factory for repair and calibration, their calibration frequency should be analyzed and recorded.

Steps of MSA

The evaluation of measurement system analysis is usually divided into two stages:

1. Stage 1: Verify whether the measuring system meets the requirements of its design specifications. There are two main purposes:

(1) Determine whether the measuring system has the required statistical characteristics, which must be done before use.

(2) Find out which environmental factors have a significant impact on the measurement system, such as temperature and humidity, so as to determine the space and environment for its use.

2. The second stage

The purpose of (1) is to verify that once the measurement system is considered feasible, it should continue to have appropriate statistical characteristics.

(2) The common one is "measuring tool R &；; R "is one of them.

Analysis of MSA measurement system

I. Introduction of Measuring System

Basic concepts of 1 and MSA

2. Why should we consider measuring system variation?

Source of data variation

The influence of error factors

3. The importance of 3.MSA

Second, the statistical characteristics of the measurement system

1, acceptable measurement system

Influence on total variables

Impact on production specifications