MSA (Measurement Systems Analysis) - Is MSA worth doing?

A

Atul Khandekar

A recent discussion thread has raised a few questions in my mind. I would like to request you all to please take some time and post your views & opinions:

1. Do you think MSA (R&R + all the other studies) is a useful (value adding) activity? Or is it just a pain in the ****, a necessary evil and a waste of time & efforts? Why?

2. What are the process situations when MSA is a MUST DO? What are the circumstances in which it just doesn't make sense?

3. Do your customers mandate MSA studies just for the sake of it?

Thanks.

-Atul.
 

LMO

Quality
A recent discussion thread has raised a few questions in my mind. I would like to request you all to please take some time and post your views & opinions:

Atul,

In response to your questions?

You!
1. Do you think MSA (R&R + all the other studies) is a useful (value adding) activity? Or is it just a pain in the ****, a necessary evil and a waste of time & efforts? Why?

Answer!
Yes R&R studies are useful. The old ways. Now we have bias, linearity and a whole bunch of other Mojo terms and formulas. The folks up in detroit have nothing better to be then try to write training books, which they tried and in my opinion with the mSA did not do a very good job of it. They should not be trying to train the supply base on basic information. They act like they are the only ones who how to do something. I was resently at the local ASQ dinner where a local professor did some training on the MSA. What a pain in my back side.

Yes, it is a pian in the ****!

R&R is not a waste of time, but some of the requirements of the MSA are!

You!
2. What are the process situations when MSA is a MUST DO? What are the circumstances in which it just doesn't make sense?

Answer!
I have both automotive and commerical customers. All of my equipment is used for both.

Not sure about the rest. I need to study it a little more! :eek:

You!
3. Do your customers mandate MSA studies just for the sake of it?

Answer!
Yes, they have taken the belief that their suppliers should be TS 16949 or at least in their Supplier Manual s require all the automotive BS.
Thanks.

The former big three now big two, I just give up on them :truce:

Cost reduction every year etc, that is why transplantes are beating them.

I am a tier 2 supplier currently and have been a tier 1 for many years.

:bigwave:
 
A

Al Dyer

Good topic for thought Atul!!!

1: YES (even though it can be a pain)

Very usefull and important if we want to know how our gages perform.

Say we all got paid by the stacked width of $100 bills. We wan't to use the most accurate gage possible!!!

2: Any dimensional measurement

When doesn't it make sense? When the customer calls and requests a piece of plastic, they dont' care what type, size, purpose etc...

3: NO

They request it because whatever item is supplied needs to mate with parts other suppliers submit and make the entire system work.

Interchangability
Form
Fit
Function
Safety

Etc...

Al...

:bigwave:
 

howste

Thaumaturge
Trusted Information Resource
Re: MSA - How much is it worth?

I think it is a value added activity when it is applied with some thought. When I buy a custom gage with a specific application, I require an acceptable R&R before I pay the supplier for it. When critical dimensions are being measured, it's very important to know the limitations of the measurement system.

My problem with it is when it is just required to check off a box. Customer SQEs sometimes require studies on dimensions that just aren't that critical. I have included R&R studies in PPAPs where the total error is <1%. My experience has been that tier 1 SQEs generally are a little more knowledgeable about what makes sense than many down the supply chain. The further you get from the source it seems, the more the requirements are just passed on without thought.
 
C

chalapathi

It is a good question and we need to ask this question repeatedly.
Measurements are the basis of all decisions - acceptane, process control etc. Error, if high can influence these decisions and there is no way to know what we are doing is right. It is a proactive way and it is really difficult to quantify the value addition of this. The company will realize, when a critical customers rejects material etc. In six sigma, we do MSA before analyzing the data.
Comming to implementation problems. These are created by the Not so effective MSA manual and the lack of understanding. The issue the quality of training and material. Further, the MSA manual is not giving practical tips for implementation. For example, attribute MSA, guidelines on how to collect samples is totally missing. Some people are using softwares. They give good reports and without propoer understnading you can get best of them.
Another concern is that many TS companies are doing MSA to meet the audit requirements and not as a good practice. It is again lack of awareness.
 

Marc

Fully vaccinated are you?
Leader
The neccessity of Gage R&R is situation specific

Atul Khandekar said:
1. Do you think MSA (R&R + all the other studies) is a useful (value adding) activity? Or is it just a pain in the ****, a necessary evil and a waste of time & efforts? Why?
This depends entirely upon the tolerances and overall requirements. Volume can also play in as a factor. Does a carpenter need to do gage R&R studies on each of their measurement instruments?

So to me - It depends upon the specific situation.

Of importance to me in *almost* all situations is the state of calibration of the instrument plus the training of the person using the gage.

Atul Khandekar said:
2. What are the process situations when MSA is a MUST DO? What are the circumstances in which it just doesn't make sense?
When the tolerances and requirements are critical. Again - It is situation specific.

Atul Khandekar said:
3. Do your customers mandate MSA studies just for the sake of it?
NA to me, but I think it's a misunderstood fad in many situations.

By the way - What thread brought this to your thoughts for discussion?

My memory is that QS-9000 played a big role in advancing the 'importance' of gage R&R over simply ensuring an instrument is in calibration and that the operators are trained to use it.


See these threads for some thoughts:
Tape Measures & Scales - Part Tolerance - Gage R&R
Tape Measure Calibration
Reasonable calibration? 50 foot measuring tape and other basic M&TE
Calibration of Tape Measures used for reference
 
A

Atul Khandekar

Marc said:
...By the way - What thread brought this to your thoughts for discussion?
Ah...Don't quite remember - it was 2003 - my own understanding of the topic was (& still is ?) rather limited and I was trying to get peoples' perspective.

Happy 2006!
 

Jim Wynne

Leader
Admin
Marc said:
Of importance to me in *almost* all situations is the state of calibration of the instrument plus the training of the person using the gage.

:agree1: Yes, except I would say "...training and experience."
 
C

chalapathi

Many people believe that, if the calibration is done properly we are taking good care of the gages. However, the five errors - Bias, Repeatability, Reproducibility, Linearity and Stability - are independed. If Bias is acceptable does not mean that the Repeatability error also need to be less.
I will explain this point: Bias is Xbar - Reference. You take five repeated measurements on the reference sample. Take the Xbar of the four readings and compute the Bias.
There is variation among the five readings (this is repeatability error). An extreme case, two the four values are very high values and two are very low values. Now the Xbar is very close to reference value and hence Bias is very less and within the acceptable limit. However, the range (repeatability error) is very high.
Hope I could make the poin clear.
Happy New Year 2006
 
Top Bottom