Repeatability and Reproducibility Studies (Gage R&R)

GLP2

Duration : 1.0 day(s)

:: Course Summary

This session discusses how to set up repeatability and reproducibility studies (R&R) and how to analyze the results with the appropriate statistical methods. The main goal of R&R studies is to determine which sources of variability contribute the most to the overall process variation and how to decide if the measurement system is acceptable or not.

:: Learning Objectives

Upon completion of this course, participants will be able to:

To understand the issues involved in repeatability and reproducibility (R&R) studies

To differentiate between the available experimental designs in R&R and to choose between them

To learn how to estimate the repeatability and reproducibility in a measurement process

To master the available statistical and graphical tools for summarizing and presenting study results

To make a decision on the acceptance of the measurement system

:: Target Audience

This applied training session in statistics is aimed at:

Laboratory, production line and quality control personnel

Anyone who must determine the accuracy and precision of gages, instruments and operators in a measurement process

:: Prerequisite

Working knowledge of methods in descriptive statistics and of confidence intervals is required. Participants should have attended Fundamental Tools in Statistics or have a similar background. Furthermore, it is important that attendees possess an excellent knowledge of the analysis of variance technique for crossed and nested designs.

:: Topics Covered

Introduction to R&R Studies
- Discussion: R&R Case Study
- Scope of R&R studies
- When Should an R&R Study be Performed?
- Positioning R&R Studies in Good Laboratory Practices (GLP)
- Characterizing the Factors Affecting the Measurement Process: Instrument, Operator, Equipment, etc.
- Understanding the Notions Involved in R&R Studies: Repeatability, Reproducibility, Stability, Bias, Linearity
- General Statistical Approach to R&R Studies
Implementation of R&R Studies
- Identifying the Possible Sources of Measurement Variability:
- The Approach of Statistical Designs
- Available Statistical Designs
- Determining the Number of Levels to Use and the Number of Operators
- Sampling and Sample Size: Understanding the Impact of Sample Size on the Scope of the Results
- Randomization: Why and How?
- Application to the R&R Case Study
Statistical Analysis of R&R Studies
- Statistical Methods and Calculation of Repeatability and Reproducibility
- Interpreting the Results
- Overview of the Analysis of Bias and Linearity
Conclusion
- Taking a Decision Based on the Accuracy of the Measurement Process based on the R&R Study Results

:: Course Content

Whether in a laboratory or a production line, taking measurements is a necessary part of any process or system. An adequate level of precision of the system is critical as it determines the capacity of the system to detect differences between the parts, samples, etc.
It is therefore necessary to quantify the detection capability of the measurement process which is achieved by quantifying the variability therein. The total variation can be further decomposed into variation due to the instrument, variation due to the operators and variation due to the parts or samples.
Although variations in the system can be caused by many different phenomena such as calibration error, stability and linearity of the instrument, typical R&R studies focus on the evaluation of repeatability and reproducibility of the process.
The main objectives in performing an R&R study are to identify and quantify the absolute and relative contribution of each source of variation, to decide if the measurement process is adequate or not and, if not, to correct the errors by recalibrating the instrument, training the operators, other mathematical corrections, etc.
Different fields often have different incentives for performing R&R studies. In quality control and quality assurance, the goal is to reduce the variability in the system and, consequently, the variability in the product. In R&D, the accuracy and reliability of the instrument and measurement system are the driving force in performing this type of study. This training course studies the statistical aspects involved in repeatability and reproducibility studies: the different sources of variation, the experimental designs for these studies, the statistical analyses required to quantify repeatability and reproducibility and the interpretation and presentation of results.