Measurement Systems Analysis

Six Sigma is a problem solving approach that is based on facts and data.  It is imperative that we gather facts and data that can be relied on for making decisions about what to change and what to change to.   The devices that are used to measure and the way that people collect data are often sources of error.   Here are some common issues to be aware of when measuring and collecting continuous data.
The first issue is granularity, which is defined as the extent to which a larger entity is subdivided.  This issue is also referred to as discrimination or resolution.  For example, say we are gathering data about how long it takes for a task to be completed.  Time can be measured in years, months, days, hours, minutes, seconds, fractions of a second, and so on.  Our measurement system must be capable of measuring at a fine enough level of granularity so that the data will be useful for analysis.
The measurement system must be accurate, i.e. the facts must represent the truth.  If we use a scale to weigh an item, we must be sure that the reading that is obtained from the scale represents the true weight of the item.  We do so by calibrating the scale using a reference weight that is traceable to a known standard that is maintained by an organization such as the National Institute for Standards and Technology.
The measurement must be repeatable, i.e. the measurement device must give the same answer when the same thing is measured in the same way multiple times.    Also, people must be able to reproduce the measurement, i.e. the same person should get the same answer when measuring the same thing multiple times.  Gage repeatability and reproducibility studies are used to evaluate measurement systems for both gage error and human error.  These studies tell us how much of a specification or tolerance is consumed by errors made by the gage and the people using the gage.  We want the percentage of error to be small, i.e. 30% or less of the tolerance or specification that we are trying to maintain.  For example, if we are trying to hold temperature to plus or minus one degree, we need the measurement system error to be .60 degrees or less.
Measurement systems must be stable over time.  An example of a lack of stability would be a measurement device that wears out over time, such as a scale that has springs that stretch.
The measurement system must be linear, i.e. the system should perform consistently over the range of values being measured.  For example, a system that measures temperature should be equally accurate at cold and hot temperatures.
Your comments or questions about this article are welcome, as are suggestions for future articles.  Feel free to contact me by email at
About the author:  Mr. Roger C. Ellis is an industrial engineer by training and profession.  He is a Six Sigma Master Black Belt with over 45 years of business experience in a wide range of fields.  Mr. Ellis develops and instructs Six Sigma professional certification courses for Key Performance LLC.   For a more detailed biography, please refer to