Measurement System Capability with a Continuous Measurement System

Miner

Forum Moderator
Staff member
Admin
#11
You analyze the raw data from a single run. Do not subgroup the data, but analyze individual measurements taken at specific intervals.

Your understanding of the interpretation is correct.

If this analysis tells you that you must wait 55 sec. between samples to see true product variation, you have two options: 1) use an I-MR chart, or 2) collect 1 measurement every 55 seconds and group x number of them into a subgroup. Normally, the period of autocorrelation is much greater than 50 seconds, so it usually makes more sense to use an I-MR chart.

Regarding Repeatability, the variation that you see during the 50 second interval is pure repeatability variation. As you describe your measurement device, there is no reproducibility variation.

Once you have this, you can perform your capability analysis. Just remember that each "part" is 50 seconds long. Allow at least 50-55 seconds between measurements for a subgroup, and longer times between subgroups.
 
Elsmar Forum Sponsor
E

Elwood

#12
Thank you, this is helping a lot.

Ok, so I would not want to keep collecting data at the 5 second interval for 55 seconds to get 11 data points and average them into one 'part?' I should take just one data point, one single measurement, every 55 seconds, correct?

Once I conduct a capability study I want to set up a control plan, do I continue to collect data at the rate of 1 measurement/per required sampling frequency to monitor the process for control purposes? Or can I return to a shorter sampling frequency? Would returning to a shorter sampling frequency for SPC increase the chance of responding to a false out of control signal if the process autocorrelation interval is longer than the sampling frequency?
 

bobdoering

Stop X-bar/R Madness!!
Trusted Information Resource
#13
If you can get to the measurement device when the line is down, place a series of cable segments in its view. Repeat as necessary for your evaluation. You can do it at two different times as your reproducibility test (since it is not really operator dependant), or if you feel as a purist, have a second person load the cable. You will be making some assumptions with this approach, such as the effect of a moving sample versus a static sample is statistically insignificant. But, it should tell you if you your gaging is anywhere near effective. :cool:
 

bobdoering

Stop X-bar/R Madness!!
Trusted Information Resource
#14
Once I conduct a capability study I want to set up a control plan, do I continue to collect data at the rate of 1 measurement/per required sampling frequency to monitor the process for control purposes? Or can I return to a shorter sampling frequency? Would returning to a shorter sampling frequency for SPC increase the chance of responding to a false out of control signal if the process autocorrelation interval is longer than the sampling frequency?
You need the results of your capability study to determine the check frequency. The capability study will tell you the type of variation you have (distribution), and the nature of the variation. How you address it (adjustment) will determine the necessary frequency. You can always sample more frequently if you wish - especially if you have the memory to store a lot of data. :cool:
 

Miner

Forum Moderator
Staff member
Admin
#15
Thank you, this is helping a lot.

Ok, so I would not want to keep collecting data at the 5 second interval for 55 seconds to get 11 data points and average them into one 'part?'
Correct.

I should take just one data point, one single measurement, every 55 seconds, correct?
Correct. If you want to subgroup the data, subgroup these individual measurements taken every 55 sec.

Once I conduct a capability study I want to set up a control plan, do I continue to collect data at the rate of 1 measurement/per required sampling frequency to monitor the process for control purposes?
Yes, or at a longer frequency.

Or can I return to a shorter sampling frequency? Would returning to a shorter sampling frequency for SPC increase the chance of responding to a false out of control signal if the process autocorrelation interval is longer than the sampling frequency?
Yes. The false alarm rate will be significantly greater because there is no short term product variation in the control limits, only measurement error.
 

Miner

Forum Moderator
Staff member
Admin
#16
If you can get to the measurement device when the line is down, place a series of cable segments in its view. Repeat as necessary for your evaluation. You can do it at two different times as your reproducibility test (since it is not really operator dependant), or if you feel as a purist, have a second person load the cable. You will be making some assumptions with this approach, such as the effect of a moving sample versus a static sample is statistically insignificant. But, it should tell you if you your gaging is anywhere near effective. :cool:
Normally, I would agree with you because many automated tests do have a Reproducibility component due to multiple stations or from variability introduced by loading the tester. However, in the situation, these possibilities do not exist and there is no Reproducibility component. No need to artificially inflate the measurement variation by adding it.
 

Bev D

Heretical Statistician
Staff member
Super Moderator
#17
on the other hand SPC often works just fine even on the presence of relatively large measurement error. So I wouldn't hesitate to start taking time series data (as suggested by Bob D) to begin understanding your largest components of variation, rational subgrouping scheme and sample size/frequency.
 

Miner

Forum Moderator
Staff member
Admin
#18
I would agree with this statement except in the situation where autocorrelation exists. If all samples within the subgroup are taken within the period of autocorrelation, the control limits are established using only measurement variation. Because of this, the false alarm rate is raised dramatically, and the chart reacts to the short term variation that would normally be part of the within subgroup variation.
 

Bev D

Heretical Statistician
Staff member
Super Moderator
#19
Miner - yes of course, which is why I recommended he start sampling to determine how to rationally subgroup. sequential pieces is not the only way to subgroup and in the presence of autocorrelation a very bad choice. The fact that autocorrelation might very well exist had already been established. and autocorrelation has nothign to do with measurement error. my post was to suggest that the OP didn't have to wait on the MSA to begin the necessarry work to determein how best to apply SPC...
 

Miner

Forum Moderator
Staff member
Admin
#20
... and autocorrelation has nothign to do with measurement error.
Not true. Within the period of autocorrelation, all variation observed is measurement variation. How can you say it has nothing to do with it?

...my post was to suggest that the OP didn't have to wait on the MSA to begin the necessarry work to determein how best to apply SPC...
Agreed, but the OP DOES need to understand the autocorrelation first, so he does not subgroup on it.
 
Thread starter Similar threads Forum Replies Date
A Yearly MSA (Measurement System Analysis) & Capability Study - TS 16949 Requirements Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 2
M MSA Chicken and Egg - Process Capability and Measurement System Capability Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 1
Ron Rompen MSA on automated measurement system - Multiple Step Vision System Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 3
G Gauge R&R on multiple dimensions using 3D measurement system Capability, Accuracy and Stability - Processes, Machines, etc. 6
Emran.mi Measurement system analysis - Can you help me about implementation MSA for CMM device Manufacturing and Related Processes 2
N Justifications for not performing MSA (Measurement System Analysis) Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 6
T ISO 13485 8.2.5. Monitoring and measurement of quality management system processes ISO 13485:2016 - Medical Device Quality Management Systems 1
H ISO13485:2016 Measurement System - Methods General Measurement Device and Calibration Topics 4
E Measurement System Analysis/validation to new factory Manufacturing and Related Processes 2
S IATF 16949 7.1.5.1.1 Measurement System Analysis (MSA - Die casting) IATF 16949 - Automotive Quality Systems Standard 5
T Method to assess Measurement System of a Continuous Process Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 2
K Measurement system analysis "shall be conducted to analyze the variation..." Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 3
B Camera Vision System Measurement System Analysis (MSA) Reliability Analysis - Predictions, Testing and Standards 3
G Where to obtain training to program the Keyence Measurement System IM6120 Training - Internal, External, Online and Distance Learning 3
E Measurement System Analysis (With parts that have variation) Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 1
L IATF 16949 7.1.5.1.1 Measurement System Analysis (MSA) - Visual Inspection IATF 16949 - Automotive Quality Systems Standard 14
W Gage R&R where Device Under Test (DUT) is part of the Measurement System Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 1
C MSA (Measurement System Analysis) - Instrument Selection Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 4
C MSA (Measurement System Analysis) - Dimension Selection Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 5
T Automated and Manual Measurement System Comparison (Correlation) Quality Tools, Improvement and Analysis 3
S MSA (Measurement System Analysis) - Process Tolerances Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 9
P Vision Measurement System - Software Program Qualification Qualification and Validation (including 21 CFR Part 11) 6
Y Gage R&R for multiple measurement system Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 1
D Signal Detection-If Ppk>1, compare measurement system to process. Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 1
S MSA (measurement system analysis) on Flammability tester? Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 2
B Selecting Parts for an MSA (Measurement System Analysis) Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 7
M A question on Measurement System Analysis - Stability to measure got Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 2
T MSA on Camera Vision System and other Process Measurement Systems Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 3
G MSA (Measurement System Analysis) and Dimensional Tolerances Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 1
C Measurement System Artificially Creates a Skewed Distribution Inspection, Prints (Drawings), Testing, Sampling and Related Topics 16
F MSA - Measurement System Analysis Attribute Data Crosstab sheet APQP and PPAP 1
R Destructive Non Reproducibility MSA (Measurement System Analysis) Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 8
M MSA (Measurement System Analysis) on a Photospectrometer Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 9
T MSA (Measurement System Analysis) Competency Test Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 11
J How do you perform a S3 Stability Study for a Non-Replicable Measurement System Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 1
B Surface Finish Equipment MSA (Measurement System Analysis) Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 5
A Is MSA useful to prove a Measurement System after Relocation? Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 2
B Very interesting article concerning MSA (Measurement System Analysis) Book, Video, Blog and Web Site Reviews and Recommendations 8
B Wait or not wait? MSA (Measurement System Analysis) Actions Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 2
B MSA (Measurement System Analysis) - Results and Actions Taken Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 12
A MSA (Measurement System Analysis) for Characteristics with One Tolerance Limit Capability, Accuracy and Stability - Processes, Machines, etc. 10
C Camera Vision Measurement System Gage R and R Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 7
W Anyone using Keyence Image Measurement System Calibration and Metrology Software and Hardware 2
C Measurement System to measure Hole Diameters, True Positions, Surface Profiles, etc General Measurement Device and Calibration Topics 4
C Attribute Measurement System Analysis (MSA) Template Document Control Systems, Procedures, Forms and Templates 0
H Measurement System in Control if the Equipment requires Daily Calibration? Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 3
D MSA (Measurement System Analysis) on a Bore Measurement - Good or bad result? Six Sigma 9
K Effectiveness of ISMS (Information Security Management System) Controls Measurement IEC 27001 - Information Security Management Systems (ISMS) 3
M Is MSA (Measurement System Analysis) Required for Degradation Testing? IATF 16949 - Automotive Quality Systems Standard 1
M MSA (Measurement System Analysis) - Do we have control of all states in the chart? Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 5

Similar threads

Top Bottom