Video není dostupné.
Omlouváme se.
Attribute MSA - Here is the simple way to do it...
Vložit
- čas přidán 28. 03. 2021
- A simple disecription of a Pass/fail MSA with no fancy Kappa statistics, lets keep it simple...
Here is my buy me a coffee link.
www.buymeacoff...
Design of Experiments for 21st Century Engineers - Minitab version www.lulu.com/e...
Drink tea and read the paper - www.lulu.com/e...
Design of Experiments for 21st Century Engineers - www.lulu.com/e...
Thx you so much for the course sir, so helpful, and easy to understand,
Glad you like it Kevin....
Would a 6 appraiser study (visual inspection) with 2 appraiser's % overall effectiveness at high 80% compared to the other 4 at high 90% be of concern enough to focus on the 2 with slightly lower scores? Even though the miss rate is zero & false alarm rate is OK?
Hi there - it all depends on what the risk of errors are, and that is for you to decide. If escapes go on to other parts of your system and the problem is kept in house, then low risk if the mistakes will be past to the customer this might be more of an issue...
@@paulallen5321 that's my thought exactly. Thank you. Although others are trying to tie me strictly to the AIAG manual. 🤦🏼♂️
Hi Paul,
We are quite often facing automatic camera systems in our processes. I am in a situation now where there is only one camera in the machine to check a cosmetic characteristic.
How would you perform a proper MSA in such case, considering that we only have one appraiser (one camera). An attributive gage R&R with one appraiser or an MSA type 1 would fit better?
Hi Jonatta - good to hear from you. You can still perform a Study its just that there is no reproducability and certainly a type study could be a good solution as long as this does not create bias. The 10 parts test is there to eliminate operator bias, so if there are no operators this is not a risk, run a type 1 study,,
Dear Paul,
First of all, thank you so much for your efforts and knowledge sharing.
Would you please cover the MSA for non replicable tests e.g. destructive tests, chemical reactions.
Kind regards
Great suggestion - i'll look into it....