Measurement Uncertainty (MU), Accuracy, Precision - Relationship?

[pagnonig]

Hi All,

I'm a bit confused about how Measurement Uncertainty relates to Measurement Accuracy and Precision.

Could anybody help me (examples will be much appreciated!).

Thank you.
Giuseppe

[Benjamin28]

Accuracy is the closeness of a measured value to it's true value.

Precision is the repeatability of those measurements.

Uncertainty is a scientific quantification of the amount of error in a measurement when you take into account the many different variables which can effect it.

You can not truly know the accuracy of a measurement without knowing the Measurement uncertainty and you can not calculate the measurement uncertainty until you know the precision and much more.

[pagnonig]

Originally Posted by Benjamin28

Accuracy is the closeness of a measured value to it's true value.

Precision is the repeatability of those measurements.

Uncertainty is a scientific quantification of the amount of error in a measurement when you take into account the many different variables which can effect it.

You can not truly know the accuracy of a measurement without knowing the Measurement uncertainty and you can not calculate the measurement uncertainty until you know the precision and much more.

Dear Benjamin,

thank you for the info.
Could you please suggest/post any document which explains, in simple terms, the concepts you summarized.

Giuseppe

[Umang Vidyarthi]

Originally Posted by pagnonig

Dear Benjamin,

thank you for the info.
Could you please suggest/post any document which explains, in simple terms, the concepts you summarized.

Giuseppe

Hello Pagnonig,

Benjamin has provided good answers here. I surmise from your post, that you perhaps are not well versed on the subject. I suggest you go through the MSA manual published by AIAG. Rummaging through the section E & F of chapter 1 will provide you with more clarity.

You may also benefit from the SPC newsletter, I am providing link for four very good articles :

http://www.spcforexcel.com/ezine/sept2007/sept_2007.htm

http://www.spcforexcel.com/ezine/oct2007/oct_2007.htm

http://www.spcforexcel.com/ezine/nov2007/nov_2007.htm

http://www.spcforexcel.com/ezine/dec2007/dec_2007.htm

Besides, you just google and find plenty of sites offering articles,books,and training. Here are some :

http://www.statgraphics.com/measurem...s_analysis.htm

http://www.minitab.com.mx/training/T...entSystems.pdf

http://www.qualitytrainingportal.com.../msa/index.htm

http://ezinearticles.com/?Measuremen...MSA)&id=801704

Hope this helps

Umang

[Benjamin28]

Giuseppe I don't really have such a document at hand myself. I would think that a physics teacher or a good beginner's book in metrology would be a good source for basic definitions and learning examples.

For examples of each concept I would use something along these lines.

Accuracy: Brad measures the outside diameter of a washer to be 0.20 inches, the true value of the OD is 0.20 inches and therefore his measurement is accurate. (so accuracy is described as a comparison of a measured value to it's true value)[it should be noted that the true value of a measurement can be infinitely small and therefore can never truly be known]

Precision: Brad uses a different instrument to measure the OD of the same washer and it reads 0.22" so he measures 10 more times and gets the same reading all ten times....it would be said that the instrument is very precise but not accurate.

Uncertainty: Brad uses his superior metrology skills to assess the uncertainty of measurements in his caliper as +/- 0.0005". In this way Brad can safely say that measurements made by this instrument utilizing a sound measurement method in a controlled setting will be within that interval 98% of the time.


These are pretty simplified examples, and as I said I'm sure you could find better ones from a physics teacher or a physics book but I hope this is of some help to you.

[pagnonig]

Originally Posted by Benjamin28

Giuseppe I don't really have such a document at hand myself. I would think that a physics teacher or a good beginner's book in metrology would be a good source for basic definitions and learning examples.

For examples of each concept I would use something along these lines.

Accuracy: Brad measures the outside diameter of a washer to be 0.20 inches, the true value of the OD is 0.20 inches and therefore his measurement is accurate. (so accuracy is described as a comparison of a measured value to it's true value)[it should be noted that the true value of a measurement can be infinitely small and therefore can never truly be known]

Precision: Brad uses a different instrument to measure the OD of the same washer and it reads 0.22" so he measures 10 more times and gets the same reading all ten times....it would be said that the instrument is very precise but not accurate.

Uncertainty: Brad uses his superior metrology skills to assess the uncertainty of measurements in his caliper as +/- 0.0005". In this way Brad can safely say that measurements made by this instrument utilizing a sound measurement method in a controlled setting will be within that interval 98% of the time.


These are pretty simplified examples, and as I said I'm sure you could find better ones from a physics teacher or a physics book but I hope this is of some help to you.

Thank you very much to both of you for your valuable help!
Giuseppe

[Umang Vidyarthi]

Originally Posted by pagnonig

Thank you very much to both of you for your valuable help!
Giuseppe

You are welcome. There is more material on the cove. Click on the forum discussion thread index (located on left hand top in third row, highlighted in blue). In the index, click on 'Measurement Uncertainty (MU)', to open the pandora box.

There is another valuable thread you may benefit from- 'Accuracy vs Precision-what is more important for process capability'.

Here : http://elsmar.com/Forums/showthread.php?t=22349

Umang