Understanding Measurement Error

Now, that we are aware that most measurement systems are in fact fallible, the next step is to explore exactly what makes them fallible. This concept has been studied by operations experts and the term measurement error has been given to it. This article explores the concept deeply as it is a primary tool in the tool-kit of any Six Sigma executive.

Decisions are Based on Variation Data: We need to re-iterate the fact that Six Sigma methodology does not like variation. It thinks that variation is the enemy of the process and must be eliminated. All decisions regarding how to modify the process are taken on the basis of observed variation. However, observed variation does not represent the true picture!

Break-up of the Variation Data: A deeper look at observed variation gives us the formula that describes measurement error:

Observed Variation = Actual Process variation + Measurement Error

Thus, the entire purpose of Measurement System Analysis is to systematically understand the different variation that the process is capable of exhibiting at different stages. Once the measurement error has been truly understood, it can be removed from the observed variation data. This will give us the actual process variation, on which effective decisions need to based upon.

Charting the Correct Data: One needs to understand, that when one piece of data is looked in isolation, the measurement error may look small and insignificant. However, the Six Sigma team and the process managers are usually looking at data from multiple variables charted on the walls of their control rooms. Here the small errors multiply themselves to manifest in the form of a big error.

It is therefore essential that a system be developed in which only the actual process variation is charted in front of the managers, instead of the observed variation. This will enable taking decisions which are much more sound than they otherwise would be.

May Arise Because of People or Instruments

Measurement error cannot be attributed to a single source. This is what makes understanding measurement variation all the more complex. We will discover the sources of measurement variation separately. However, at the current moment, it is enough to understand that measurement variation manifests itself in the incorrect reading of the instruments by the people and the instruments themselves. An effective measurement system must get rid of both of these errors.


❮   Previous  Article Next  Article   ❯


Authorship/Referencing - About the Author(s)

MSG team comprises experienced faculty and professionals who develop the content for the portal. We collectively refer to our team as - “MSG Experts”. To Know more, click on About Us. The use of this material is free for learning and education purpose. Please reference authorship of content used, including link(s) to ManagementStudyGuide.com and the content page url.