What is Gage Repeatability and Reproducibility? Terms, Properties, Characteristics

What is Gage Repeatability and Reproducibility?

Gage Repeatability and Reproducibility (GR&R) is a statistical method used in Six Sigma to assess the variation in measurement systems, specifically those involving gauges or other measurement tools.

Before the project team starts collecting data for the Measure phase activities using the defined measurement system, it is quintessential to evaluate or analyze the measurement system’s accuracy level. This is called Measurement System Analysis (MSA).

The accuracy of statistical methods depends on the accuracy of the data which in turn depends on the accuracy of the measurement instrument or method used to collect data. If data is collected using a defective measurement instrument/method, it can lead to the inclusion of significant measurement errors in the data. Therefore, conclusions made using such defective data are often inaccurate and misleading which may lead to false signals on control charts. In such cases, an effective process may be misunderstood as an incapable process.

When the source of variation lies in the measurement system, the project team may require a significant amount of time and effort in fixing and controlling the process. The two important types of measurement errors are systematic errors and random errors. Systematic errors occur when the measuring instrument is not calibrated properly. Random errors occur due to factors such as operator variation, instrument variation, environmental changes, and time-to-time variation.

The total variability that is included during the measurement can be attributed to two factors, namely variability due to the part of the product being measured and the variability attributed to the measurement or gauge.

Measurement System Analysis (MSA) is also called Gauge Repeatability and Reproducibility abbreviated as Gauge R&R or simply R&R. The objective of MSA is to determine the part of the variation in data resulting from variation in the measurement system. Certain statistical methods are used to separate the two components of the total variability and to determine the gauge capability.

Terms of Measurement System

Before continuing our discussion on MSA, it is important to define a few terms as follows:

• Gauge
  
• Bias
  
• Resolution

Gauge

Gauges are devices or instruments of predefined dimensions. These are used to check whether a product meets the stated expectations or not.

Bias

Bias is the difference between the average value of the measurements and the true value/reference value of the part.

Resolution

The resolution of a measurement refers to the number of digits of precision needed in the measured value.

Properties of Measurement System

A measurement system can be evaluated and the measurement errors can be assessed if the project team knows what the properties of a good measurement system are. Therefore, let us first learn about the properties of a good measurement system. They are:

• Accuracy/Bias
  
• Precision

Accuracy/Bias

Accuracy refers to the difference between the average of measurements made on a part (observed value) and the true value/reference value/master value of that part. The bias tells the project team how well is the given measurement as compared to the reference value. If the true value is unknown, it can be calculated by averaging several measurements with the most accurate measuring equipment available.

Linearity

Linearity tells the project team how accurate are the measurements throughout the expected range of measurements. Linearity is also defined as the measure of the consistency of bias over the range of the measurement device. For example, when the blood sugar level of a person is between 0 and 250, a glucometer shows a deviation of ± 2.5%. However, the same glucometer shows a deviation of ±4% when the blood glucose levels are more than 250. Therefore, the bias here is non-linear.

Stability

Stability refers to the capacity of a measurement system to produce the same/consistent values over some time when measuring samples. Stability means the absence of variation attributable to special causes in a control chart. However, common cause variation can still be present.

Precision

Precision means that the measurements were done repeatedly to yield consistent results. In other words, precision can be defined as the degree to which repeated measurements tend to agree with each other.

Measurement errors are estimated using accuracy and precision. The different components of accuracy and precision.

Repeatability

A process is said to be repeatable when the same appraiser can measure the same part or sample several times with the same measuring device and get the same value every time.

Reproducibility

A process is said to be reproducible when different appraisers can measure the same part or sample several times with the same measuring device and get the same value every time.

Characteristics of Measurement System

The characteristics of a capable measurement system are as follows:

• Statistical stability
  
  
• The variability of the measurement system must be smaller than the process variability.
  
  
• The variability of the measurement system must be smaller than the specification limits (tolerance).
  
  
• Low resolution

There has been a rapid development in automation technology that affects automotive manufacturing processes. The manufacturing processes are usually massive and efficient. Given these developments, automotive companies and suppliers demand measurement systems with accurate and automated measuring instruments.

A lot of problems (including financial losses) can accrue if these organizations use problematic instruments that are inaccurate and unstable. Therefore, it is required that these organizations use an effective MSA.