# Gage R&R Anova

## This article on Gage R&R Anova shall help reader understand the concept and methods involved in the Gage R&R Study in MSA exercise.

This article on Gage R&R Anova shall help reader understand the concept and methods involved in the Gage R&R Study in MSA exercise.

Gage R&R Anova

•Gage Repeatability and Reproducibility (R&R) is a tool and technique to understand the amount of variability that a measurement system (or counting technique) brings to the overall process.

•Gage R&R is first used in the Measure phase, specifically in a step called Measurement System Analysis (or Adequacy), also called MSA. Later, you perform Gage R&R again in the Control phase.

•In Measure, you gather data about the Y’s. These are the elements that the customer feels.

•By the time you get to Control, you know which are the critical X’s that affect your Y’s, and it is important to keep those X’s within spec limits. Thus, starting Control and for the life of your process, you measure the vital few X’s.

•The term measurement system and gage are used interchangeably, though technically people and the environment can also contribute to measurement system errors.

MSA_1

Data gathering steps

• Establish upper and lower spec limits (USL and LSL)

• Collect existing data or run experiments that produce data

• Count, or measure, the results that will be either good or defects

• Calculate process (or product) capability from the ratio of good to bad

Where does measurement come in?

• One part just barely made it to be classified as good

• The actual results are distorted by the measurements

• The measurement system looked at that part several times and sometimes called it good and other times bad

• Measurement systems distort reality

Our Objective

Our goal is to minimize the distortion caused by measurement systems.

MSA2

• When calculating a sigma value, you would hope that you are measuring the actual process quality.

• You are actually measuring the observed process quality, which has two parts: The real-life actual results and the quality of the gage (the measuring system).

• In an ideal situation, the gage is so good that it contributes very little to the total observed variation.

• The measurement system will almost always add to the variation observed, and it is our job to understand and minimize it as much as possible .

• Gage errors rarely occur at the midpoint between upper and lower spec limits. Even if you distort the real number, you are still going to be in the “good” region.

• The problems occur near the goal posts, the USL and LSL. There, we might mistake a good value for bad, or a defect for a good result.

• The result is additional variation, and what is worse, it is not even real.

• There are different causes of variation in the gage as the underlying true data. Or the same causes may contribute different percentages.

Sources of measurement system variation

Here are potential causes of measurement system variation:

> People

> Process

> Machine

> Scale

> Environment

• A fishbone diagram is a tool to brainstorm the effect of these causes on the observed data values.

2 Ways of going through MSA

– Gage R & R as % of Contribution

– Gage R&R as % of Tolerance

MSA_Contribution

MSA_Contribution_rules

MSA_Tolerance

MSA_Tolerance_rules

## Comments (0)

## Facebook Comments