How to Perform Measurement Systems Analysis

Measurement systems analysis (MSA), or gage reproducibility and repeatability (R&R), quantifies measurement variation from inspectors (reproducibility) and the gage itself (repeatability). This returns a measurement of the gage's precision, or ability to get the same result consistently from the same part. Precision is the counterpart to accuracy, or the gage's ability to get, on average, the correct measurement from a part. Gages must be precise as well as accurate to perform their jobs. While ISO 9001:2008 required only accuracy (as ensured by calibration), the automotive ISO/TS 16949 standard requires MSA as well.

• Know the difference between calibration (which ensures accuracy) and MSA (which measures precision)
• Know the meaning of repeatability (the ability of a gage to get the same measurement from the same part, when used by one inspector) and reproducibility (the ability of the gage to get the same measurement from the same part when used by different inspectors). That is, are the results user-dependent? It will depend on the nature of the gage
• Know the meaning of linearity and bias. Bias is a systematic deviation from the part's actual dimension, while linearity reflects bias as a function of the magnitude of the measurement
• Know how gage variation affects the risks of rejecting acceptable parts and accepting nonconforming ones, as well as the power of statistical process control (SPC) charts to detect process shifts
• Know how gage variation affects the process performance index or process capability index, and how to get the true process performance index if the gage variation is known
• Know how to perform a reproducibility and repeatability (R&R) study to quantify the gage variation. This includes not only the mathematical calculations (which are easily handled by software packages like Minitab and Stat Graphics) but also sampling requirements and practices to ensure useful results
• Know how to implement possible remedies for non-capable gages

• Graphical images will be used to illustrate the concepts of accuracy (as assured by calibration) and precision (as quantified by MSA)
• Figures will also illustrate the chances of accepting nonconforming parts and rejecting good ones, as functions of the part's actual measurements and the gage variation
• Equations will be provided for the effect of gage capability on statistical process control (SPC) and process capability estimation
• Gage variation reduces the power of control charts to detect process shifts, although it does not increase the false alarm risk
• Gage variation also reduces the process capability estimate although, if the gage variation is known from MSA, its effect can be removed mathematically
• Common sources of gage variation from a perspective similar to that of the cause and effect diagram
• Gage study requirements: number of parts, number of inspectors, and randomization aspects
• Average and range, and analysis of variance (ANOVA) methods for calculation of the repeatability and reproducibility components
• Remedies for non-capable gages
• A table of the d*2 factors (necessary for performance of the calculations in Excel) will be provided

MSA is a major and vital component of the control plan for a manufacturing process, which means that everybody involved in process planning (e.g. advanced quality planning) and process management needs to understand what it does and how it works. This includes understanding of the risks from poor gage capability (regardless of the gage's calibration status) and potential actions to address situations of this nature.

A control plan is a mandatory part of advanced quality planning (AQP), and it defines the required measurements for critical to quality (CTQ) part characteristics along with the required gages and instruments. This in turn defines the calibration and measurement systems analysis (MSA) requirements. Calibration and MSA ensure (respectively) that gages are accurate and that they are sufficiently precise for their roles in the control plan.

Quality managers, engineers, and inspectors with responsibility for process measurements and also advanced quality planning (AQP), R&D engineers, Six Sigma professionals, Auditors, Regulatory affairs, noting that MSA supports the control plan.

William A. Levinson, P.E., is the principal of Levinson Productivity Systems, P.C. He is an ASQ Fellow, Certified Quality Engineer, Quality Auditor, Quality Manager, Reliability Engineer, and Six Sigma Black Belt. He is also the author of several books on quality, productivity, and management, of which the most recent is The Expanded and Annotated My Life and Work: Henry Ford’s Universal Code for World-Class Success.