Cg, Cgk calculation: measuring instrument capability index | measurement system 1-1.3| IHDE Academy

[@pijavka123]

The best explanation of Cg, Cgk I found online. Clear and simple. Great work!

[@ihdeacademy]

Thank you for the positive feedback. My pleasure.

[@edgarivanbermudezrodarte9718]

Hi, great video ! I have a question, where can be found the reference about cg and cgk need to be equal or higher that 1.33 ? I have look for it in AIAG MSA manual and SPC with no success

[@ihdeacademy]

@@edgarivanbermudezrodarte9718 Hi, thank you for your question. Different companies have different requirements regarding Cg and Cgk. You may Google 'Quality Management in the Bosch Group; Technical Statistics 10. Capability of Measurement and Test Processes.' There, you will find the capability criterion: Compliance with specified minimum values for Cg and Cgk. The current release of [CDQ 0301] is binding for these minimum values. At the time of the publication of the present issue of booklet 10, the following limits apply: Cg ≥ 1.33 and Cgk ≥ 1.33.

[@deepan16]

Hi, How did we finalize the reference value Xm of 60.1030?

[@ihdeacademy]

Hi, thank you for your question. At 2:15, it is mentioned that a part from the series that has been calibrated serves as the 'normal' reference part. A reference value of xm = 60.1030 is specified, i.e., measured for the diameter. You may remember that in the context of quality control or manufacturing, a calibrated part would be one that has been measured and confirmed to match certain specifications very closely, serving as a standard or benchmark for other parts.

[@lakshmipathi8340]

Hello, i have one doubt if we get cg cgk value more than that >1.33,is it ok or not ok

[@ihdeacademy]

Hello, Thank you for your question. The best advice I can give is to: (A) Ensure that the measuring instruments are properly calibrated (ISO/IEC 17025), and that the data collected for the calculation of Cgk is representative of the process being measured (ISO 3951 Series, ISO 16269 Series). (B) As stated in the tutorial, depending on the applicable customer specifications, guidelines, or standards (MSA, ISO 22514-7, ISO 10012, VDA Volume 5, ANSI/ASQ Z1.9), there may be different calculation methods and specifications for certain values. If you are sure that (A) is fulfilled and you are in doubt whether you can achieve a Cpk value of at least 1.33, then you should involve your customer to agree on (B). As you can see, there is no simple yes or no answer to your question.

[@liseaimejaychou]

Hello, I have a question, 25 mesures are done on the same part, right?

[@ihdeacademy]

Hi, that is correct. All tests are carried out on the same part by the same person with the same measuring instrument.

[@deanopenn]

what happens if bias fails but cgk is good in a type 1 study

[@ihdeacademy]

Hello, Thank you for your question. I hope you find the following explanation helpful: If bias fails but Cgk is good in a Type 1 study, it indicates that while the measurement system can consistently replicate the same results (good precision), it systematically deviates from the true value (poor accuracy). This scenario suggests that although the gauge can reliably distinguish between close measurements, any conclusions drawn from the data may be systematically shifted or incorrect due to the bias. Steps to Address the Issue: Calibration: The measurement system should be recalibrated to align it closer to the true value, reducing the bias. Root Cause Analysis: Investigate why the bias occurred. This could involve examining the measurement device, operator errors, or environmental factors. Repeat the Type 1 Study: After addressing the bias, the Type 1 study should be repeated to ensure the bias has been effectively corrected and to reevaluate the Cgk. Monitor and Control: Continuously monitor the measurement system for any future occurrences of bias and implement regular checks to ensure ongoing accuracy and precision.

[@deanopenn]

@@ihdeacademy Hi thanks for a detailed response, so based on this, you would fail the study based on Bias having a p-value less than 0.05? even though Cg and Cgk is good.

[@ihdeacademy]

Hi, Thank you for your follow-up question. Yes, the study would fail based on a bias having a p-value less than 0.05, even if Cg and Cgk are good. The p-value indicates the statistical significance of the bias. A p-value less than 0.05 means the observed bias is unlikely to be due to random chance, indicating a significant systematic error in the measurement system. While Cg and Cgk demonstrate good precision (i.e., the measurement system can consistently produce repeatable results), a significant bias suggests that the measurements are not accurate. This means the measurement system is not reliable for making accurate decisions, as the systematic deviation from the true value could lead to incorrect conclusions. Significant bias undermines the reliability of the measurement system. Therefore, corrective actions are necessary to address and eliminate the bias, as suggested in my previous answer, before the system can be deemed acceptable.