Search the Elsmar Cove!
**Search ALL of Elsmar.com** with DuckDuckGo including content not in the forum - Search results with No ads.

Gauge R&R on multiple dimensions using 3D measurement system

#1
Hello all,
This is my first post here, so please be kind :D

I am trying to analyze my data on minitab, and was hoping to get some insight.
Background:
We have a 3D measurement system, that can scan a whole part. We have setup an "inspection program" that measures 18 different dimension.
As part of the Test Method Validation protocol, we will be doing gauge r&r on this inspection program.
Once the part is loaded in the system, the system takes automated measurement without operator intervention.

10 parts being measured 5 times each by one operator, one gauge.
each part will have 18 individually identified dimensions.

I can perform gauge r&r on each dimension, but i dont want to do it 18 times for each dimensions.
I was wondering if any of you here have come across a similar situation or have suggestion on how you would approach this.
Thanks
 

Ronen E

Problem Solver
Staff member
Super Moderator
#2
I can perform gauge r&r on each dimension, but i dont want to do it 18 times for each dimensions.
If you don't, it implies an assumption of uniform variability among the measurement of these 18 different dimensions.
Are you okay with this assumption?
I would have gone through the full exercise at least once (if it's a new CMM) to get an idea whether that would be a reasonable assumption going forward (i.e. for future studies on other parts etc.). But even then, such assumption would have to be made explicit on each and every study.
 

Bev D

Heretical Statistician
Staff member
Super Moderator
#3
A Guage R&R is not intended to be a ‘check the box’ exercise. You might save a little time with your approach but if you miss something you’ll spend a lot more time and money trying to correct it. There are so many things that can go wrong even with (and maybe especially with) a CMM. Approach angles, size of the ruby, basic errors in programming, poor fixtures, etc. I’ve experienced this desire to cut corners too many times with catastrophic results. (The worst was with a CMM on an engine part that caused mid air shutdowns of the engine) It’s like running a red light - you won’t always get T-boned but when you do...
 
#4
I am sorry I was not clear with my request.
I want to do gauge r&r on all 18 measurements. The system automatically measures those dimensions.

I was wondering, if minitab could process all 18 of them. I know how to do one at a time. I am fairly new to minitab.
Thanks.
 
Last edited:

optomist1

A Sea of Statistics
Trusted
#5
Gamma, what version of MT are you using? To confirm you have parts to be measured via CMM, on each part you wish measure 18 different features or part dimensions? Also with the CMM are there any fixtures or gages to be used?
 

Miner

Forum Moderator
Staff member
Admin
#6
I am sorry I was not clear with my request.
I want to do gauge r&r on all 18 measurements. The system automatically measures those dimensions.

I was wondering, if minitab could process all 18 of them. I know how to do one at a time. I am fairly new to minitab.
Thanks.
The easiest way would be to put all your data into a single worksheet in Minitab. Part numbers and Operators in the first columns followed by 18 columns of data (one for each dimension). You can then run the study 18 times, or create a macro using the following command 18 times total.

GageRR;
Parts 'Part';
Opers 'Operator';
Response 'Measurement 1';
Studyvar 6;
Tolerance 5.
GageRR;
Parts 'Part';
Opers 'Operator';
Response 'Measurement 2';
Studyvar 6;
Tolerance 8.

You will need to change the values shown in red to match the column names for your 18 measurements and the corresponding tolerances.
 
#7
Miner, thanks for the tip. I did that manually for my first inspection program.
I will try the macro for the second inspection program.
 
Top Bottom