# Comparing 2 gages

Six Sigma – iSixSigma Forums Old Forums General Comparing 2 gages

Viewing 8 posts - 1 through 8 (of 8 total)
• Author
Posts
• #40136

Lomax
Participant

I have a study in which I would like to compare 2 gages. The current “talk” is that group “A” uses one gage to measure a part and then group “B” is using another gage (both designed and made internally) and group “B” rejects parts that group “A” accepts. I’ve performed gage R&R’s before but what would be the best approach in this situation?
Thanks,
Neil

0
#124235

sathish chandran A R
Member

Hi,
The issue here is lack of operational definition; get the answers for the following;
1) What is the parameter to be measure.
2) How exactly you need to measure it.
3) Which instrument is the right instrument to use.
4) how to use the instrument to measure
5) arrive at the procedure for measurement.
6) Train all concerned on the same.
7) Establish frequency of claibration etc.
regards
sathish

0
#124236

Randa
Participant

Neil,
Firstly, are the two gauges calibrated? Secondly, as the test is non-destructive as you can measure the same part on both gauges, a paired t-test would tell you whether they are reading the same. Thirdly, if the gauges are different, but are both calibrated, then the meausurement process for the two must be different. Consider the problem then as a mini 6S issue – i.e. define the gap, measure (map the process), etc.
Randa

0
#124242

“Ken”
Participant

Neil,
Would it be possible for you to measure a range of samples on both gages?  In otherwords, could you select say 5 to 10 samples across the expected range of measures, then measure each part on gage 1 and measure each part again on gage 2.
If you can do this type of paired evaluation, then you could prepare a scatterplot of the paired measures, hopefully regress a linear relationship, assess the quality and confidence bounds for the linear relationship, and determine the bias between the two measurement systems.
Just a few thoughts before I call it a day…
Ken

0
#124304

MLA
Participant

First thing to do:
Get rid of one of the measurement gages. Your letting them add variation to the measurement process. Then do the gage RnR on the ONE measurement tool
mla

0
#124330

jediblackbelt
Participant

How about trying to do a DOE on the gages and operators.  You could then come up with something to show whether or not it is a difference due to operators or gage or the interaction between the two.
Just a suggestion too is that if you do a paired T-Test to also use only one operator.  Otherwise you are throwing in a variable that you can’t account for in order to determine your gage bias.

0
#124477

Lomax
Participant

Folks,
Thanks for your input. The problem is:
1. The instruments are “home made”
2. One is in a Cleanroom while the second is on the manufacturing floor.
3. I was thinking of performing a paired “t-test” as many suggested.
4. The “measurement” isn’t exactly what most people are used too (spring preload aka spring droop)
I really can’t get rid of one of the instruments (not practical) and I really can’t use it on the production floor and then bring it into the cleanroom (again not practical). Having samples is not a problem and I really think the paired t-test will be the best approach.
Thanks to all for the input,
Neil

0
#124953

jormonal
Participant

Neil:
I agree with sathish chandran: you must know EVERYTHING about yor measure system. In the other hand I think ‘home-made’ instruments is not an issue as long as they were built to measure the same kind of things. By the way:
1. (Perhaps a silly question) Are the instruments made WITH THE SAME MATERIALS?.
2. What kind of UNITS your instruments measure?; are the variables CONTINUOUS or DISCRETE (go-nogo like)
3. You wrote ‘ one is in the Cleanroom while the second is on the manufacturing floor’: Are the ENVIRONMENTAL conditions THE SAME for both instruments? (humidity, temperature)

0
Viewing 8 posts - 1 through 8 (of 8 total)

The forum ‘General’ is closed to new topics and replies.