Do we still need to Calculate Measurement Uncertainties (MU)?

I talked to one of the Lab experts who is also the assessor of ISO/IEC 17025:2005 & has more than 3 decades of experince.
One of the thing that he assert is that in some cases measurement uncertainty/ Precision (repeatability/ reproduceability) is given in the standard method so we don't need to re calculate it for those tests, he also asserts that this measurement uncertainty we can report to the customer & it is accepted by the accrdediation boards.
Do we still need to calculate the measurement uncertainties for tests & we can rely on this statements?
My question is important since sometime the repeatability/ reporoduceability can be differnt from the one stated in the standard test method.

Re: Do we still need to calculate measurement uncertainties?

In many cases this may be true, such as for the ASTM standards where a precision and bias statement is included. However, not all standards have such statements or equivalents and in those cases you very likely will need to calculate your uncertainty.

The best thing, if there is any doubt, is to calculate the uncertainty.....and to ask your chosen accrediting body, which I presume is SCC.

Re: Do we still need to calculate measurement uncertainties?

I would echo Hershal on this topic. Many standard methods do not require calculating UOM if you follow the method to a t and will typically contain a statement to that effect. If you can get away without the added work of uncertainty calculations go for it. And if you're still doubtful, ask your AB.

Re: Do we still need to calculate measurement uncertainties?

just to understand the logic since ...though uncertainties given in ASTM std but our repeatability & repeatability is different from the one that is given. Our analyst is diffrent & though our equipment is same as mentioned in the std but its accuracy & precision maybe different.
My question is logically how do we justify using the same MU as ASTM std refers?

Re: Do we still need to calculate measurement uncertainties?

I have to admit I was intrigued by the question at hand. As I work for a metals testing laboratory we apply the note within ISO 17025 and yet we certainly do not quote a numerical value for the MU:

In those cases where a well-recognized test method specifies limits to the values of the major sources of uncertainty of measurement and specifies the form of presentation of calculated results, the laboratory is considered to have satisfied this clause by following the test method and reporting instructions.

Interestingly enough this statement has been debated quite extensively...in particular is the question of whether a method can provide a numerical value for measurement uncertainty which can be quoted. You can read about some of this here...

http://www.astm.org/cgi-bin/SoftCart.exe/SNEWS/OCTOBER_2005/ullman_oct05.html?L+mystore+jyuv9248

The more interesting statement I found within this debate is:

"A22.2 It is neither appropriate for, nor the responsibility of, the test method to provide explicit values that a user would quote as their estimate of uncertainty. Uncertainty values must be based on data generated by a laboratory reporting results using the test method.

So, to me that would seem to be an answer...but I would look for something more definite and official. If you research further please do post your findings here for the rest of us to read.

Re: Do we still need to calculate measurement uncertainties?

We are also an L-A-B accredited test lab and we do not calculate measurement uncertainty for the majority of our testing.

What we chose to do is chart our properties in Excel and use action limits and warning limits. The action limits provide an estimate of uncertainty at approximately the 99.7% level of confidence and the warning limits provide approximately a 95% level of confidence. Our auditor has given this her blessing.

The A2LA Guide for the Estimation of Measurement Uncertainty in Testing is what we used to set this up. Is was written by Thomas M. Adams. You might want to take a look at this excellent guide.

Re: Do we still need to calculate measurement uncertainties?

Ken K said:

We are also an L-A-B accredited test lab and we do not calculate measurement uncertainty for the majority of our testing.

What we chose to do is chart our properties in Excel and use action limits and warning limits. The action limits provide an estimate of uncertainty at approximately the 99.7% level of confidence and the warning limits provide approximately a 95% level of confidence. Our auditor has given this her blessing.

The A2LA Guide for the Estimation of Measurement Uncertainty in Testing is what we used to set this up. Is was written by Thomas M. Adams. You might want to take a look at this excellent guide.

Click to expand...

but the question is, is it accpetable to the accreditation body?