Intro: Welcome to the E6S-Methods podcast with Jacob and Aaron, your source for expert training, coaching, consulting, and leadership in Lean, Six Sigma, and continuous improvement methods. In this episode number 49, “Rule out the Ruler” part 1 of our Measurement Systems series, we discuss some of the basic measurement system terminology, bias, linearity, stability and more... Here we go.
***Rule out the Ruler - Measurement (Decision) Systems***
Objection 1: This is a tried and true measurement system we've been using for more than 30 years, this is the industry standard.
Counter 1: Common sentiment, and it may be true. Often find uncontrolled details that create variabilty, even from an ASTM standard method.
Objection 2: This is just a complicated way of telling us what we already know. The data is no good.
Counter 2: Perhaps, and it gives you some direction to make an improvement.
Objection 3: Data may be no good, but this is what we have to live with.
Counter 3: This may also be true. Sometimes the output of the MSA is to live with a less reliable system, or eliminate the measurement and find another means to get what you need.
Measurement systems terms & definitions (Decision Systems)
I Measurement System: Combination of Instrument, Procedures/Methods, and User/Operator
a. A measurement system is a process in itself, and often its variation can be explained by external factors (i.e. X’s, or inputs), much in the same manner that the processes they measure can vary. (Man, Machine, Mother Nature, Methods, Materials)
i. QC Gauge example:
1. Situation where the process spec & variation shrunk so much. Only measurement system variation was being captured & reported. Total variation was primarily the gauge, not the process. The same process also was extremely sensitive to temperature swings within the room; statistically determined during the gauge study. (8 Deg F swing created a measurement system variation that ate up entire spec limit for the product). Any shift in the actual production process would be missed. Reported “certified” output values may only represent the noise in room temperature fluctuations
II Accuracy vs. Precision
a. Accuracy – Measurement comparison to actual known standard value
i. Bias – reading offset from a known value.
“I know my home scale shows a bias, because when I step on it, it says I weigh 5 pounds more than I actually do.”
ii. Linearity – A change in bias over the scale’s measurement range.
“I know my home scale shows linearity, because when my wife steps on it, it says she’s 10 pounds more than she actually is, (vs. only a 5 pound difference for me).”
iii. Stability – changes is bias over time.
1. Example: Standard everyday measurements, varying either randomly or trending.
2. Stability issues may indicate uncontrolled external factors (like temperature, expiration, or wearing components.)
b. Precision – Ability to consistently measure the same value
i. Repeatable – variation within a single user/operator
ii. Reproducible – variation between users/operators
Outro: Thanks for listening to episode 49 of the E6S-Methods Podcast. Stay tuned for episode number 50, part 2 of “Rule out the Ruler,” where we continue our discussion on Measurement Systems, explaining the concepts of Repeatability and Reproducibility. Subscribe to past and future episodes on iTunes or stream us live on-demand with Stitcher Radio. Have an idea for an episode? Contact us! Follow us on twitter @e6sindustries. Join a discussion on LinkedIn. Find outlines and graphics for all shows and more at www.E6S-Methods.com. “Journey Through Success”