Measurement System Analysis (With parts that have variation)




I have what i believe a rather uncommon case, where i need to perform Measurement System Analysis on testing bench which is used for testing other measurement devices (Water meters)

In this case i have:
Equipment: Testing bench (Water flow bench, with high accuracy water meter for master sample)

Parts: water flow measurement devices (1-10 parts on bench)

Operator: operators are commonly known to have influence to measurement results (No previous MSA was performed)

By using parts from production it is to my opinion impossible to differentiate variation between test bench and produced water meters. (because it is commonly known that both have variation)

Should we use some high-accuracy meter to replace parts from production or is there any other way to perform MSA correctly?

Thanks for any advice and please note i am new to measurement system analysis.

Bev D

Heretical Statistician
Super Moderator
welcome to the cove! :bigwave:

You are performing a functional MSA. This is different from a 'static' MSA where the parts themselves do not change (at least in the short run) as with most dimensions and geometrical features. Functional MSAs are used when the event or measurement is a function and therefore the characteristic of interest changes from trial to trial.

You will probably find this document on how to perform a Functional MSA helpful...
Last edited by a moderator:
Top Bottom