Practical MSA Advice

welcome to complexity made simple my name is paul allen and before we get on to today's video just a reminder of how you can support the channel please subscribe please leave comments on all the videos because that really does help but if you want to support the channel further there is a donation page please go to buy me a coffee and leave a donation which is always fantastically helpful keeps the channel keeps the channel moving and allows me to make all of this fresh material but if you want a textbook that contains all the good stuff from the videos please click on the link and buy drink tea and read the paper from lulu.com all of these things really help the channel thank you to all the people that continue to buy this textbook it really is fantastically helpful now let's get on to today's video welcome to the latest video and in this video newsletter what we're going to be looking at is some details of how to conduct an msa correctly now i'm not going through the msa process i'm dealing with some important points about measurement system analysis so we're going to be looking at you might call it gauge r and r it's the technique we're going to be looking at measurement system [Music] analysis and we'll just call them important points now this follows from a an email that i received by a viewer who's sent me a question about this is msa and he sent me two questions and he said number one should i round up should i round up so what he's got he's got an additional decimal place one more decimal place than the tolerance that he has and he wants to know whether that's a bit too being a bit too anal a bit too specific should he cut it off and just ignore the last decimal place and the other question is should he select defects it actually says it's actually out of tolerance parts he doesn't say defects what he's actually suggesting is that he actually measures something quite a bit bigger than what the tolerance is aiming for because he's been told he's got to include defects in the msa so let's answer these two questions well in both cases the answer is no should you round up now if you remove decimal places removing decimals makes things worse and in fact the advice is you should always have one more decimal and the tolerance so in other words if you are measuring hundredths of a mil that's your tolerance your measurement system should have an extra place on there and it should measure thousands of a millimeter that's really important that helps to get good rnr then this question should he select defects or out of tolerance well the answer here is no what the measurement test is doing what the gauge rnr is doing is it's asking the question is your measurement system capable of seeing difference for the group of data that your manufacturing process is producing so what you should do you should randomly select 10 pieces and this is most important randomly so what you're going to do your manufacturing process is producing this now this might be a capable process might have no defects in it so it might look like this what what you're asking the question is is my measurement system capable of seeing different size along that scale so if that's what your your manufacturing system creates if you randomly select out of that that's what your measurement system will be measured against and that's what your measurement system should be measured against if you deliberately start creating parts out here to create defects you will completely ruin the validity of the msa the msa is measured against the process that it's it's working with and of course one of the problems that you see over time if you get good at improving your process and you drive the variability in the more you drive the variability in the more likely it is that your msas will start to fail on a regular basis now that isn't because your msa your measurement system is poor it is because your manufacturing system is getting better and better so there is a point where your your msas will fail but you will still say the measurement system is okay to be used because what it can still see it will still see the process beginning to wander backwards and forwards it will still tell you that change is happening it's just not very good at picking out the specific size at a specific moment but there's there's the answer to two really important details please don't round up that's one of the worst behaviors ever if it's measured you write it down you never ever drop data never let the computer do that later if it wants to but always collect all the decimal places and the other one always randomly select from the manufacturing process that the measurement system is being used on because that is what the measurement system analysis is doing you are you are you are measuring it against the task that you're trying to conduct please select the parts randomly otherwise you ruin the result it's no longer valid if you start offering up random out of tolerance results there is your measurement system advice folks please use your measurement systems wisely your measurement system analysis wisely don't panic about bad results use your common sense and if you use your common sense you'll stop yourself spending bucket loads of cash that really isn't necessary and if you need any advice please drop me a line i'm more than happy to help you with any msi results that are confounding you

Channel: Paul Allen

Views: 263

Keywords: lean, six sigma, Six Sigma greenbelt training, Six Sigma Blackbelt Training, Shewhart, Juran, Deming, Taguchi, SPC, MSA, FMEA, DOE, X bar chart, Wheeler, Janam Sandhu, Mrnystrom, Gemba Academy, Full Factorial, Central Composite Design, Ronald Fisher, Hypothesis Test, p value, minitab, https://youtu.be/QH984PnwRDE, https://youtu.be/f_fjqCpd67Q, https://youtu.be/AGJ1QYI2B4c, https://youtu.be/gsD8V2_eZ0A, https://youtu.be/mM6EyMvvAKk, quality hub india, https://www.youtube.com/watch?v=um9Ly-ZsZUc

Id: yN811BjEA54

Length: 8min 49sec (529 seconds)

Published: Tue Oct 19 2021