Gage R&R for Continuous Data

1. Understanding Gage Repeatability & Reproducibility

Before we conduct a study, let's define the key terms. In Six Sigma, a "gage" isn't just the physical measuring tool; it's the entire measurement system—the device, the operator using it, and the procedure they follow. A Gage R&R study breaks down the measurement system's variation into two main components:

• Repeatability: Variation that occurs when the same person measures the same part multiple times with the same gage. This is often called Equipment Variation (EV), as it reflects the inherent precision of the measurement device itself.
• Reproducibility: Variation that occurs when different people measure the same part with the same gage. This is often called Appraiser Variation (AV), as it reflects the consistency between operators.

The following video provides an excellent visual introduction to these concepts and explains why controlling measurement error is critical to profitability and customer satisfaction in a manufacturing environment.

2. Setting Up a Gage R&R Study

Now that we understand the concepts, let's get into the practical steps of conducting a study. We'll focus on the most common type for non-destructive testing in manufacturing: the Crossed Gage R&R study.

A "crossed" study means that every operator measures every part multiple times. This is contrasted with a "nested" study, often used in destructive testing where a part can only be measured once by a single operator.

The preferred method for analyzing the data from a Gage R&R study is Analysis of Variance (ANOVA). While older methods exist, ANOVA is more accurate because it can distinguish the variation coming from the interaction between operators and parts.

This next short video explains the different study types and analysis methods.

Step-by-Step Planning

Proper planning is the most critical part of a successful Gage R&R study. Rushing this step can invalidate your results.

To summarize the key planning steps:

1. Determine Study Size: A common starting point is 3 operators, 10 parts, and 3 repeat measurements (trials) per part. This gives you a total of 3 x 10 x 3 = 90 data points.
2. Select Appraisers: Choose operators who normally perform the measurement as part of their job.
3. Select Parts: This is crucial. The 10 parts you select must represent the full range of variation your process typically produces. Include parts that are on the low end of the specification, on the high end, and in the middle. If you only use parts that are very similar to each other, you will artificially inflate the percentage of error attributed to your measurement system.
4. Randomize and Blind the Study: The parts should be numbered, but the measurements must be taken in a random order. The operators should not know which part number they are measuring, nor should they see their previous results. This prevents bias from influencing the measurements.

3. Collecting Data and Performing the Analysis

Once the study is planned, the next step is to collect the data systematically.

Data Collection

You will use a data collection sheet to record the measurements. Each row will contain the operator, the part number, and the measured value.

A third party should ideally manage the data collection, providing the parts to the operators in random order and recording the results without revealing them.

Performing the Analysis with Software

In any modern Six Sigma project, the Gage R&R analysis is performed using statistical software like Minitab. Your role as a Green Belt is not to perform the complex ANOVA calculations by hand, but to know how to input the data correctly and run the analysis.

The following video demonstrates exactly how to do this.

4. An Introduction to the Output

After running the analysis, the software will produce a set of graphs and tables. We will dedicate the entire next lesson to a deep dive into interpreting these results. For now, the goal is to familiarize yourself with what the output looks like and what the key terms mean.

This final video walks through the standard output of a Gage R&R study, explaining each component of the ANOVA and Variance Components tables.

The graphical output provides a quick visual summary of the results. The most important chart is often the Components of Variation chart. In a good measurement system, the bar for Part-to-Part variation should be much larger than the bars for Repeatability and Reproducibility. This indicates that most of the variation seen in the study comes from actual differences between the parts, not from measurement error.

Conclusion

In this lesson, you've learned the practical process for conducting a Gage R&R study for a continuous measurement. This is a non-negotiable step in the Measure phase for ensuring the integrity of your project data.

Key Takeaways:

• A Gage R&R study quantifies the variation from the measurement system by separating it into Repeatability (equipment variation) and Reproducibility (appraiser variation).
• A Crossed Gage R&R study using the ANOVA method is the standard approach for non-destructive measurements in manufacturing.
• Proper planning is essential, especially the selection of parts, which must represent the full process spread.
• The study must be conducted in a random and blind manner to prevent bias.
• Statistical software is used to generate analytical outputs, including numerical tables (like Variance Components) and graphical summaries.

Preview of the Next Lesson:
Now that you know how to plan a study, collect the data, and generate the results, our next lesson will focus entirely on interpretation. We will learn how to use the AIAG (Automotive Industry Action Group) guidelines to analyze the % Study Variation and Number of Distinct Categories to formally decide if a measurement system is acceptable, conditionally acceptable, or unacceptable for its intended use.