SPC Control Charting Rules

[@TheEngineeringToolboxChannel]

Thank you all for watching! In the description are time quick links to different parts of the video. Feel free to ask questions! I reply to ALL comments and questions as best I can. 🙂 If you enjoyed this video be sure to subscribe and let me know what you would like to see more of! I am happy to take suggestions.

[@KyleGillis]

This 10 min vid contained everything from a 4 month class I had last year lol.

[@TheEngineeringToolboxChannel]

Haha! Colleges gotta make their money somehow!

[@babybats562]

Thank you so much for your videos, they are really useful as they explain simply the whole concept along with working examples

[@baselanz9854]

Thanks for the great video, what does IAR stand for?

[@TheEngineeringToolboxChannel]

In A Row. 👍

[@_M_D_M_]

Great Work

[@4abhisheksrivastava]

Thanks for the great video. I have one question. When you say Rules 1-4 can be applied to mR, R and S charts, do you mean only Nelson? How about Western Electric?

[@wmouse5158]

Well explained! Could you please explain more about chart # 8.

[@Error-oi9ms]

Very interesting to know that nelson extended the rules up to 8. I thought all rules are introduced by Shewhart. Just wanted to know that after we build charts, can we automate to detect patterns and shifts. Or manually do we have to detect the patterns and shifts?

[@TheEngineeringToolboxChannel]

It can be automated! There are stats programs like MiniTab built with these techniques in mind. I also have mutiple videos on my channel to show you to do it in excel! Check them out! The only manual part of SPC is setting up appropriate sampling plans and deciding which rules you want to use/consider in your analysis.

[@Error-oi9ms]

@@TheEngineeringToolboxChannel Thanks, I will checkout MiniTab.

[@brudo5056]

Ok full explanation of most of the validation/interpretation rules. But I hope one day someone can make a ‘good’ video about the correct continuation and maintenance of a control chart... what to do with the parameters (average, standard deviation, control limits, outliers etc) in a new chart and the dataset of all the previous charts etc etc... or can I find this somewhere in a good publication? Greetings and my search goes on ;-)

[@TheEngineeringToolboxChannel]

Control charts and their parameters should be started or “reset” from the point the process was established and determined to be stable. If the process is changed you have to restart as well. I.e. verify control and reestablish the parameters for the process. Hope this helps!

[@John5ive]

You are awesome...

[@envcas31]

Whooo! Keep up the good work!!!!

[@TheEngineeringToolboxChannel]

Env Cas31 knew you’d like this one! Thanks!

[@envcas31]

Hehe yeah! I see you added some other ones, ill go through em tomorrow at work. Thanks a lot again my friend

[@envcas31]

Hey, friend. I know we spoke about the removal of special cause variations from the data set, but my question is when do you stop? I was practicing with my technicians with some data points we collected from Temperature and Relative Humidity in the labs, and we noticed we had SCV right off the bat. We discussed the causes for these signals and determine the were cause by factors beyond our control and the data went back to "normal" shortly after. We removed those points but we kept seeing more SCV. If we were to remove all the points that were SCV, our data points would drop for 31 all the way down to about 16. FYI, we did removed them one by one and recalculated the limits each time. Is there a limit to how many points you can remove?

[@TheEngineeringToolboxChannel]

There is no limit to the amount of points that you can remove but you have to be able to assign a cause and CORRECT the issue when removing points.. The other thing, is that if you have that many points out of control in your data, the whole process may be out of control and monitoring the process with the chart will do you no good because you dont have a predictable process. Also, A rule of thumb to keep in mind is that 30 data points are required to make a reliable control chart...sounds like you only have 16 left? I would say that is not enough. And What type of control chat are you using? This could also be a sampling issue. Would you be able to send me the data? [e-mail]

[@envcas31]

@@TheEngineeringToolboxChannel Actually its more like 49 points I believe, and after correcting the SCV there were still points spiking outside of 3Sigma. Sure I Can send you the Excel sheet. It'll be from my work email : [e-mail] just in case its in your junk.

[@justluke8730]

8:22 cause 6: hunting season is more of an annual tradition ;)

[@eriquedemans5178]

Please help A quality control engineering is considering the optimal design of a X ̅ chart. Based on his experience with the production process there is a probability of 0.03 that the process shifts from an in control to an out-of-control state in any period. When the process shifts out of control it can be attributed to a single assignable cause, the magnitude of the shift is 2σ. Samples of n items are made hourly and each sampling costs $ 0.50 per unit. The cost of searching for the assignable cause is $25 and the cost of operating the process in an out-of-control state is $300 per hour. (a) Determine the hourly cost of the system when n = 6 and k = 2.5 (b) Estimate the optimal value of K for the case n = 6 (c) Determine the optimal pair of n and k.

[@TheEngineeringToolboxChannel]

I do not have enough information or know enough about your coursework to answer this question. Sorry...I recommend talking with your instructor for help.