Measurement Study

Custom Search

Get Free Six Sigma Tools Here

Variable Gage R&R Study


Purpose

The purpose of the Gage R&R measurement study is to analyze the scope and types of error produced when a variable based measurement system interacts with its environment.

The measurement study focuses on two types of error with the greatest effect on the measurement system, repeatability and reproducibility, accounting for the variation introduced by one measuring instrument and different operators.


Anatomy

Measurement Study

Reference: Measurement Systems Analysis (Chrysler, Ford, GM)

Terminology

A. A continuous characteristic (CTQ, CTP, etc.) to be measured with an instrument.

B. The measuring instrument that we want to verify.

C. A representative group of personnel who normally operate the instrument.

D. Data collection and analysis of the results with a statistical software.


Major Considerations

A variable Gage R&R measurement study is a carefully planned way to study a measurement system. This tool looks at the variation in measurement data employing one measuring instrument and several operators. In the case that it is necessary to analyze other factors, such as different equipment, measuring methods etc., a design of experiment study would be required.

Application Cookbook

Although the number of operators, trials and parts may vary, the following steps are the most common for a Gage R&R measurement study.

1. The instrument must have an adequate resolution that allows at least one-tenth of the expected process variation of the characteristic to be read directly. For example, if the characteristic's variation is 0.001, the equipment should be able to "read" a change of 0.0001.

2. Identify three operators who will participate in the study. These operators must be familiar with the instrument and the measuring method.

3. Obtain 10 samples (e.g. parts) that represent the actual or expected range of process variation. Number the parts and mark the area to be measured on each part.

4. Calibrate the gauge, or assure that it has been calibrated.

5. Have the first operator measure all the samples once in random order (another technique is to reduce human bias by blind sampling, where the operator does not know the identity of the parts being measured).

6. Record the values obtained on a data collection sheet or directly in a computer file.

7. Have the second and then the third operator measure all the samples once in random order.

8. Repeat steps 5-7 until finishing 3 trials.

9. Check for outliers in the data, if possible on the spot, in order to repeat these readings (wrong adjustments, handling errors, dirt on the samples or instruments, reading errors, etc.). A good way to find outliers is out-of-control points on the range chart.

10. Analyze results using Minitab Software (Stat>Quality Tools>Gage R&R Study).

Note: The preferred method of measurement study analysis is the ANOVA. To use the Gage R&R function, Minitab needs three column variables: Part numbers (sample names or numbers for each observed measurement), Operators (operator names or numbers for each observed measurement) and Measurement data (observed measurements).

To generate part and operator numbers, the function Simple Set of Numbers can be used (Calc>Make Patterned Data>Simple Set of Numbers

11. Determine follow-up actions if necessary. For example, improve/clarify measuring method, organize training for the operators, find another instrument, etc.

Attribute Gage R&R Study


Purpose

An attribute Gage R&R measurement study evaluates the consistency between measurement decisions to accept or reject.

Anatomy

Measurement Study

Reference: Measurement Systems Analysis (Chrysler, Ford, GM)

Terminology

A. A sample of operators who usually take the measurements.

B. A characteristic (CTQ, CTC,CTD, CTP) measured with attribute data.

C. Data collection and analysis of the results.


Major Considerations

An attribute-based measurement system, such as a go/no-go gage, cannot indicate how good or how bad a part is. It can only indicate if the part is accepted or rejected.

Although this Gage R&R permits an evaluation of the measurement system used to obtain data, it is recommended to try and find a measurement system which provides continuous data.

Application Cookbook

Although the number of operators, trials and parts may vary, the following steps are the most common.

1. Identify two operators who will participate in the study. These operators should be selected from those who normally take the data.

2. Obtain 20 samples (e.g. parts) that fall within the range of possible outcomes of the process. For example, when selecting parts for evaluating a go/no-go gage, it is desirable that some of the parts are slightly below or above the standard.

3. Have the first operator measure all the samples once in random order (blind sampling, where the operator does not know the identity of the sample, is often done to reduce human bias).

4. Record the values obtained on a data collection sheet or directly in a computer file.

5. Have the second operator measure all the samples once in random order.

6. Repeat steps 3-5 until finishing 2 trials.

7. Analyze the results and determine follow-up action if necessary.

An Excel spreadsheet can be used to analyze the results. The measurement system is acceptable if the majority of the measurement decisions agree. If the measurement decisions do not agree, the measurement system must be improved and re-evaluated.

Some improvement methods are:

  • establish a standardized measuring procedure,
  • provide training to the operators,
  • change the go/no-go gage or
  • find an alternative measurement system which can obtain continuous data.

From Measurement Study to Six Sigma Tools.

Recent Articles

  1. Quality Control Plan - Simple Method To Create Effective Plans

    May 10, 16 09:24 PM

    A Quality Control Plan is a documented description of the activities needed to control a process or product. The objective of a QCP is to minimize variation.

    Read More

  2. Get Free Six Sigma Tools

    May 10, 16 08:49 PM

    The Largest Collection of Free Six Sigma Tools and Training on the Web!

    Read More

  3. Weibull Distribution

    May 10, 16 07:28 PM

    The Weibull distribution is applicable to make population predictions around a wide variety of patterns of variation.

    Read More