Measurement systems analysis

The measurement system analysis or measurement equipment capability analysis or test equipment capability analysis, short MSA ( English: Measurement System Analysis), refers to the analysis of the ability of measuring equipment and complete measurement systems in quality management and Six Sigma.

2.1 resolution
2.2 Method 1 (Eng. type- 1 study)
2.3 Method 2 (English type- 2 study, Gauge R & R study)
2.4 Method 3 (English type- 3 study, R & R study)

Measurement deviations

Generally one distinguishes five different categories of measurement deviations: accuracy, repeatability, reproducibility, linearity, and stability.

Precision, accuracy, bias ( engl. accuracy, bias)

The precision or accuracy is determined by repeated measurement of the same specimen. To this end, we compare the average of the measured values with the true characteristic value of the specimen. The difference between the mean value and the true value is called a systematic error ( bias german ). A statement about the accuracy ( correctness ) is then made of the measuring instrument basis of this difference.

Repeatability, reproducibility (English repeatability )

To determine the repeatability of the same test sample is measured by the same operator and with the same measuring equipment in the same place several times in rapid succession. The sample is, however, come again and again between the individual measurements. The standard deviation of the measured values is then a measure of the repeatability.

Reproducibility, traceability ( engl. battery characteristic )

To determine the reproducibility are on the same test items according to a defined measuring method measurements

By different operators,
In different places or
With multiple devices of the same type

Performed. Usually there are two or three operators who repeatedly measure the same parts, or one and the same operator repeated the same measurement process at different locations or with different devices. Within an investigation but is always changing only one of these three variables ( user, location, device). The measure of the reproducibility are then the differences between each of the operator ( or at any place or with any device ) observed mean values.

Stability ( engl. stability )

To investigate the stability of several measurements are in accordance with a defined measuring method with same equipment at the same place and by the same operator at fixed time intervals made of the same specimen. After each series of measurements, the mean of the measured values is calculated. The differences between the observed at different time points mean values are then used as a measure of the stability of the measuring means.

Linearity (English -linearity)

To study the linearity of the same operator and performed with the same measuring equipment at the same location and for a specified process measurements on several specimens, covering their characteristic values for the entire range of values expected in practice. Each sample is measured several times. Then, the average of the observed measurements is calculated for each sample. Then, the difference between the true value and the observed mean is calculated (see above accuracy ) for each specimen. Are these differences vary in size and these differences are so large that they are not simply explained as a random scattering, the behavior of the measuring instrument is not linear.

Each measurement system analysis is preceded by an analysis of the resolution of the measuring device used. Besides being used in practice mainly two research methods used: the method 1 (Eng. type- 1 study) and method 2 (English type- 2 study). The analysis of the data acquired in this is often performed with statistical software packages such as Minitab, qs -STAT or QS -1-2 -3-4.

Resolution

This method examines whether the measured value display (analog or digital) is shown reversed in relation to the tolerance accurately enough. Generally 5% tolerance limit are considered. (Eg: resolution 0.001 with a tolerance of 0.02)

Method 1 (Eng. type- 1 study)

This method examines the accuracy and repeatability of a measurement system. For the investigation of a normal with a known characteristic value will be used. The normal is 50 times ( 25 times) measured. The normal case is put back after each measurement. Based on the standard deviation of the measured values and the systematic error of measurement indices Cg and Cgk are then calculated. The computational steps for this are similar to those of process capability analysis; as " tolerance " a company-specific fixed percentage of the feature or characteristic dispersion tolerance is used.

Method 2 (English type- 2 study, Gauge R & R study)

This method examines the repeatability and precision of a measuring instrument (English repeatability and battery characteristic, so R & R, and Gauge R & R and Gage R & R, after engl. Ga [ u] ge, meter ) and will only be used if the measurement in accordance with procedures 1 has been classified as capable. Here, ten parts should preferably cover the entire variation range of the measured characteristic, two or three times by three different operators (or at three different locations and at three different devices of the same type ) is measured. None of this operator may see the results of the other operator. The parts should be measured at each passage in a randomized order, so that the operator can never remember the measurement result in the previous passage.

When the measurements are completed, an overall mean and an average span value is calculated (based on the differences between the largest and smallest measured value that the operator has determined for each part ) for each operator. The difference between the largest and smallest mean value operator allows a statement about the reproducibility; the overall mean value of calculated for each operator average span values is used to give evidence concerning the repeatability. Based on repeatability and precision, the total scattering of the measuring instrument is then calculated and correlated to the characteristic scattering or tolerance.

Method 3 (English type- 3 study, R & R study)

Process 3 is a special case of the method 2, which assumes that the operator can not influence the measuring device or the influence is negligible. Typical applications are automated measuring systems.

358917