Hey buddy, I am a mechanical engineer doing my internship at Supplier Quality department. I have to say that Sampling plans and incoming inspection was totally another world that what I learned at school. This video really helps me out to do a better job and understand the job of my peers. Excellent! Interns: we are gonna get that engineering position, do not stop dreaming and working!!
You covered all the points beautifully. You clearly explained up arrow, down arrow & 100% inspection note below the table with such clarity. Thank you for sharing the knowledge!!
Wow, thanks so much Alfredo! I get alot of positive comments, and I always appreciate when that feedback is specific like yours!!! So thank you so much!
Hey OEM QE here ; COuple of Questions - How do we define the AQL ? Who defines it generally? - How do we decide if we use single , double or multiple sampling plan ?
Hey Pranay!!!! Okay, so the AQL associated with your sampling plan should be based on the criticality/risk associated with the feature that you're inspecting for. High risk/high criticality features should have low AQLs, because we want the occurrence rate (defect rate) for that defect to be low. Low risk failure modes can tolerate a higher AQL. But ultimately you determine the AQL. You are also the person who gets to decide between single/double/multiple sampling plans. This basically comes down to the pros/cons of each plan. Double sampling has, on average, a lower average sample size, than a single plan. However it can be harder to administer (aka people can make more mistakes).
@@CQEAcademy Thanks for the quick reply. I own the suppliers so I give them AQL. What AQL should I assign to PC and SCs and what should be for all other measures. Like when a supplier is proving capability, on what basis should i give them the AQL?
@@pranayrkr Hey Pranay, okay so again that has to come from you, and the defect rates that you are willing to accept from a risk perspective. High risk quality attributes should have low AQLs, and then the supplier should design and control their process to ensure a capability level that is sufficient to achieve that AQL.
The video was easy to follow and understand it's a great help! But I am just confused in this part though where did the parameters from the multiple sampling came from? I mean its sample size, acceptance number and rejection number. How can we find it? Thank you
Your videos are very good. They are knowledgeable, clear and precise. I really like the slides that you use. Could you please make a video on MSA as well. Thank you so much!
Great explanation!!! I just have one question regarding the double sampling... what does the black square mean on some fields? for example H with 0.25 AQL
Dear teacher, I really love your videos, may I ask you a question? What the meaning of ASL data? for example, when we choose ASL criteria 1, does it mean the CPK is 1? Looking forward for your answer, teacher.
Very nicely presented! I have question on AQL - Does the AQL % represent the actual % defects? So if an AQL =1% does this mean the acceptable % defect by the customer is 1%?
Hey There VK, that's a great question, and honestly it's a long answer. Sooo, the ANSI Z1.4 standard talks alot about the definitions of AQL, and the common miss-interpretations of it. Technically, AQL is commonly defined as the worst tolerable process average that is still considered acceptable. So you're right to think about it as an on-going, average defect rate. But I don't think it should be described as an "acceptable" defect rate. It's the worst defect rate that would be acceptable. You can set an AQL at 1.0, but you should desire a defect rate much lower than 1%. Does that make sense? Sorry for the long reply.
Hi sir, Thanks for your excellent explanation. And I have a question, did the Double sampling plans Table have a Tighten and reduce like a single sampling plans
Hey Andy, if you have free time would you explain or upload videos explaining about acceptance sampling for Variable Sampling (ANSI/ASQ Z1.9) or this table is the same like ANSI Z1.4? Thank you! btw I love your videos :)
Thanks so much Nur! I've added that idea to the list of video ideas. In all honesty though, Z1.9 is very minor (in relationship to Z1.4) both in how often it is used, and it's relative importance on the CQE Exam.
Can you make more videos on practicing questions ?? topics like probability, Anova table , statistics like numerical questions please that will be really helpful.
Hi, wonderful video. But I have question, sampling is destruction, so if we in example No 3 inspect whole lot (100% inspection), isn't that destruction of whole lot, which means nothing to sell to the customer? Thank you
Thanks, I'm glad you like it! Yes, i'll be doing videos in the future on both the Binomial and Poisson probability models that will show you have to calculate these values normally!
That's a decision your organization has to make. And it all depends on the risk that you're willing to accept. The default is general inspection level 2, but if you can tolerate more risk (producers risk and consumers risk), then you can choose to go with general inspection level 1, or the special inspection levels. Also, think about your entire quality system. Do you have other elements in your quality system (downstream inspection), poka yoke, that would mitigate or offset the additional risk that comes with choosing a different inspection level.
Hey Deja! Great question - okay, so the ANSI standard tells you how many samples you have to take (lets say 50 for example). Then lets say you receive product on a pallet that has 16 boxes on it, the square root of n +1 tells you how many boxes that you should take samples from - SQRT(16) = 4, then we add +1 to get a total of 5 boxes to sample from. So in that instance you should take those 50 samples from 5 of those 16 boxes.
great question Manoj!! So the default inspection level is General Inspection Level 2 - Moving up or down should be risk based. If you can tolerate more risk (producers risk and consumers risk) you can move down to level 1, if you risk appetite is lower, you can increase your inspection to Level 3.
Hey Sai, That double side pointing arrow (Up and down) is actually 2 separate arrows. One arrow is in the top box pointing up (Example is F at AQL 1). The second arrow is in the bottom box pointing down (Example is G at AQL 1). (Those examples are for single, normal sampling) The up arrow means that you move up to the plan above that sample size code letter. The down arrow means that you move down tot he plan below that sample size code letter.
Here's an example on how you can use the double arrow: Say you have a lot of *495* parts, you want to apply General Inspection Level III, your AOL is 0.4%, that is, you can tolerate a maximum of *0.4%* defect within this lot (or maximum of 495* 0.4% ~ 2 bad parts, i.e., as long as this lot contains no more than 2 bad parts, you will accept it) According to the left Table, this lot size of *495* falls under Sample Size Code Letter "*J*". According to the right table, with 0.4% AOL, you can start with sample size of 32, if zeros are found, you can stop the sampling and accept the lot. If you find one defect, you will have to increase the total sample size to 125, or sample 125 - 32 = 93 more parts. If no more defect is found, you can accept the lot. If you find more defect, then reject the lot.
Hey Muhammad, that all depends on the testing that you conducted. If your testing altered the product in any way, you might need to discard it. But if your testing is non-destructive, and you maintain lot traceability with the items, you can consider putting them back into the lot. Also think about the mixup risk. If you're testing similar items, and there is a risk that when you put them back into the lot you accidentally mix them up, then you might want to just toss them. There's not a hard and fast rule for the samples, and it really comes down to your business and the risks.
@@CQEAcademy Thank you Sir for the explanation. Another question: One more thing, how about the retention sample, how many percent should we take retention sample (for finished product). Is it any guidelines or ISO can I refer?
Hey @@MuhammadFirdaus-yd5dw!! There is no hard and fast rule about sample retention. It really comes down to your own internal requirements, or the requirements of your customer.
Great job nailed it. Can you please let me know when to use single , double and multiple plans thanks We have 15 different test for same sample so do I have to devide 200 sample by 15 or 200 for each test ?
Hey Shakir! You have to take 200 samples total, but as long as none of the tests are destructive, then you can take the same 200 samples and test them for the 15 different attributes that you are inspecting for. If the tests are destructive, then you'd have to take 200 samples per attribute. Also in terms of single/double/multiple, that's a choice you get to make. If the process of taking additional samples is easy, then a double/multiple sampling plan is likely a good choice. But if taking samples is alot of work, then the benefits of a double sampling plan are reduced, and a single sampling plan is likely the best choice.
Hello, I have a question for you. Is it possible to apply testing standards (ANSI Z1.4 for example) to a full batch of production? I mean using all the population to be tested? I know it will require more time and resources but we know very little our third party (manufacturing company).
Hey Fernando, yes the ANSI Z1.4 standards are intended to be applied to completed batch production. I'm assuming here you want to test product produced by a 3rd party manufacturer. If you want to test product that you are producing, I'd recommend using a control chart, over ANSI sampling, to control and monitor your process. It'll give you a more advanced warning to issues occurring in the process - and not waiting until the entire batch has been produced.
The AQL depends on the price of the product, and the effect of a possible failure. There are tables that recommend possible AQL levels depending on these 2 attributes
Sir you videos are amazing helping me for preparing the exam . I Have a CQE book get it from ASQ latest version but the book has not thse sampling table . How I will get these tables? They will give me in the exam or I have to bring my own separately ?
Thanks for your reply. They said they will accept CQE Book . Are you sure they will accept sampling plan book too ? is there any other kind of book I can bring with me ? The
@@jatindersingh8796 Yes, the exam is open-book, and there are no limitations around what resources you can bring into the exam. The only thing you CANNOT bring into the exam are question/answers/solutions texts.
Hey Alfred! It sounds like you're elbow deep in preparing for the CQE Exam. If you have a study resource, it should contain those sampling tables. If you don't have a sampling resource, just email me at Andy@CQEAcademy.com and I can share the most important ones with you. -Andy
Hey there, great question! So, likely yes, these tables will most likely covers the question on the exam. However, there's a small chance that there are other tables that might be required. The best thing to do is to have the ANSI Z1.4 standard with you during the exam, or some other equivalent resource that has the tables.
Depending on sigma, some industries doesn't need 0 defective, for example company making pencils, if 1 defective no issue, but if it medical company it's big issue so zero defective important
Hey buddy, I am a mechanical engineer doing my internship at Supplier Quality department. I have to say that Sampling plans and incoming inspection was totally another world that what I learned at school. This video really helps me out to do a better job and understand the job of my peers.
Excellent! Interns: we are gonna get that engineering position, do not stop dreaming and working!!
I'm glad you enjoyed that Adrian, and I love that you're growing in your career. keep it up!
Andy, your explanations are so easy to follow and understand. Thank you!
Thanks so much James!!!
You covered all the points beautifully. You clearly explained up arrow, down arrow & 100% inspection note below the table with such clarity. Thank you for sharing the knowledge!!
Wow, thanks Vinayak!!!!
Andy, thank you for the clarification. You explained things very thoroughly and clearly. No extra unnecessary vocab .
Thanks so much Karla!!!!
This was great, broke down 5hrs of reading in 12mins….thank you.
Wow, thanks so much Alfredo! I get alot of positive comments, and I always appreciate when that feedback is specific like yours!!!
So thank you so much!
Hii sir thank you very much for your support. I have passed my CQE exam in first attempt with your help. Thank you very much again.
Hey Jatinder!!!! Congratulations, that's super exciting!!
I'm glad i was able to help you on your journey!!!
All I can say is excellent content and presentation.
Thank you so much!
Very good explanation. Thank you very much❤
You're welcome!
Hi sir, Your explanation makes me very understandable about the sampling process. So I can explain it to our co-workers. Thank You!
Nice explanation of single, double and multiple sampling plans.
Thanks David!!!
Best explanation ever
Thanks Sadia!
Wow you doing great job ... thank you , thank you very much
You're absolutely welcome!!!
I really like your posts Andy
Thanks!
That was seriously very helpful.
I'm glad you liked it!
You are the best. Your explanation is so easy to understand!
Thanks!!!!!
And I love the wholes Girls for Stem - we need more women in Stem, so thank you for that!!!
@@CQEAcademy thank you and I am getting ready to right my Exam! Thank you for all you doing
@@girlsforstemusa970 You're absolutely welcome! Let me know if you're struggling with anything, so I can help!
@@CQEAcademy sure!
Great explanation!
Thank you!
Great explanations! thanks
Thanks Erez!
Hey OEM QE here ; COuple of Questions
- How do we define the AQL ? Who defines it generally?
- How do we decide if we use single , double or multiple sampling plan ?
Hey Pranay!!!!
Okay, so the AQL associated with your sampling plan should be based on the criticality/risk associated with the feature that you're inspecting for.
High risk/high criticality features should have low AQLs, because we want the occurrence rate (defect rate) for that defect to be low. Low risk failure modes can tolerate a higher AQL.
But ultimately you determine the AQL. You are also the person who gets to decide between single/double/multiple sampling plans.
This basically comes down to the pros/cons of each plan. Double sampling has, on average, a lower average sample size, than a single plan. However it can be harder to administer (aka people can make more mistakes).
@@CQEAcademy Thanks for the quick reply. I own the suppliers so I give them AQL. What AQL should I assign to PC and SCs and what should be for all other measures.
Like when a supplier is proving capability, on what basis should i give them the AQL?
@@pranayrkr Hey Pranay, okay so again that has to come from you, and the defect rates that you are willing to accept from a risk perspective.
High risk quality attributes should have low AQLs, and then the supplier should design and control their process to ensure a capability level that is sufficient to achieve that AQL.
@@CQEAcademy Perfect Thanks
This is exquisite. Thank you very much.
You're welcome, I'm glad you like it!!
I liked it very much help full to me.
Thanks, glad you enjoyed it!
✌love your explanation
I am going to watch your all videos
Thanks for the awesome feedback, I'm glad you're loving it!!!
The video was easy to follow and understand it's a great help! But I am just confused in this part though where did the parameters from the multiple sampling came from? I mean its sample size, acceptance number and rejection number. How can we find it? Thank you
Your videos are very good. They are knowledgeable, clear and precise. I really like the slides that you use. Could you please make a video on MSA as well. Thank you so much!
Thanks Manuj! Yeah gauge R&R is on the to-do list for the future!
Good job! Thank you for sharing.
You're welcome, I'm glad you enjoyed it!
thank you sir
You're welcome!
This video's are so underrated
Great explanation!!! I just have one question regarding the double sampling... what does the black square mean on some fields? for example H with 0.25 AQL
Dear teacher, I really love your videos, may I ask you a question? What the meaning of ASL data? for example, when we choose ASL criteria 1, does it mean the CPK is 1? Looking forward for your answer, teacher.
Very nicely presented! I have question on AQL - Does the AQL % represent the actual % defects? So if an AQL =1% does this mean the acceptable % defect by the customer is 1%?
Hey There VK, that's a great question, and honestly it's a long answer.
Sooo, the ANSI Z1.4 standard talks alot about the definitions of AQL, and the common miss-interpretations of it.
Technically, AQL is commonly defined as the worst tolerable process average that is still considered acceptable.
So you're right to think about it as an on-going, average defect rate.
But I don't think it should be described as an "acceptable" defect rate. It's the worst defect rate that would be acceptable. You can set an AQL at 1.0, but you should desire a defect rate much lower than 1%. Does that make sense? Sorry for the long reply.
Very clear. Good job!
Hi sir, Thanks for your excellent explanation. And I have a question, did the Double sampling plans Table have a Tighten and reduce like a single sampling plans
You're welcome, and yes the double sampling plans also have tables for tightened/reduced inspection.
Hey Andy, if you have free time would you explain or upload videos explaining about acceptance sampling for Variable Sampling (ANSI/ASQ Z1.9) or this table is the same like ANSI Z1.4?
Thank you! btw I love your videos :)
Thanks so much Nur! I've added that idea to the list of video ideas. In all honesty though, Z1.9 is very minor (in relationship to Z1.4) both in how often it is used, and it's relative importance on the CQE Exam.
Can you make more videos on practicing questions ?? topics like probability, Anova table , statistics like numerical questions please that will be really helpful.
Hey Jatinder, yes, absolutely I will definitely continue to publish more great content in the future!
Great! We are waiting!!!!
Hi, wonderful video. But I have question, sampling is destruction, so if we in example No 3 inspect whole lot (100% inspection), isn't that destruction of whole lot, which means nothing to sell to the customer? Thank you
Not all sampling is destructive. In that example, you wouldn't use destructive testing, or you would use a different sampling plan.
Great presentation, appriciate your efforts. Can you please explain how to calculate acceptance and rejection no. manually??
Thanks, I'm glad you like it!
Yes, i'll be doing videos in the future on both the Binomial and Poisson probability models that will show you have to calculate these values normally!
@@CQEAcademy Thanks, Looking forward for your future videos.
@@CQEAcademy Hope you are doing well. Still awaiting for your video on manual calculation of acceptance and rejection number.
And dear teacher, why we usually choose the general inspection level II ? but not I or III?
Who define the "Special Inspections Levels" and the "General Inspection levels"? Thanks
That's a decision your organization has to make.
And it all depends on the risk that you're willing to accept. The default is general inspection level 2, but if you can tolerate more risk (producers risk and consumers risk), then you can choose to go with general inspection level 1, or the special inspection levels.
Also, think about your entire quality system. Do you have other elements in your quality system (downstream inspection), poka yoke, that would mitigate or offset the additional risk that comes with choosing a different inspection level.
whats the purpose of √n+1 if youre going to use tables to determine the sample amount?
Hey Deja!
Great question - okay, so the ANSI standard tells you how many samples you have to take (lets say 50 for example).
Then lets say you receive product on a pallet that has 16 boxes on it, the square root of n +1 tells you how many boxes that you should take samples from - SQRT(16) = 4, then we add +1 to get a total of 5 boxes to sample from.
So in that instance you should take those 50 samples from 5 of those 16 boxes.
one quick question? Why did you select general inspection level II only, which is "L"? why not level 1 or 3?
great question Manoj!! So the default inspection level is General Inspection Level 2 - Moving up or down should be risk based. If you can tolerate more risk (producers risk and consumers risk) you can move down to level 1, if you risk appetite is lower, you can increase your inspection to Level 3.
@@CQEAcademy Thank you so much for the reply
@@Manoj-je5oy You're welcome!
great
You're welcome!
Great Explanation Sir. What is the Exact Meaning of Double Side Pointing Arrow i.e; both top and bottom side pointing Arrow?
Hey Sai,
That double side pointing arrow (Up and down) is actually 2 separate arrows.
One arrow is in the top box pointing up (Example is F at AQL 1).
The second arrow is in the bottom box pointing down (Example is G at AQL 1).
(Those examples are for single, normal sampling)
The up arrow means that you move up to the plan above that sample size code letter.
The down arrow means that you move down tot he plan below that sample size code letter.
Here's an example on how you can use the double arrow:
Say you have a lot of *495* parts, you want to apply General Inspection Level III, your AOL is 0.4%, that is, you can tolerate a maximum of *0.4%* defect within this lot (or maximum of 495* 0.4% ~ 2 bad parts, i.e., as long as this lot contains no more than 2 bad parts, you will accept it)
According to the left Table, this lot size of *495* falls under Sample Size Code Letter "*J*".
According to the right table, with 0.4% AOL, you can start with sample size of 32, if zeros are found, you can stop the sampling and accept the lot.
If you find one defect, you will have to increase the total sample size to 125, or sample 125 - 32 = 93 more parts. If no more defect is found, you can accept the lot.
If you find more defect, then reject the lot.
safe and sound!
Thanks!
After samplings process, then what happened to the item that we sampling?
Should keep aside or can use back?
Thank you sir.
Hey Muhammad, that all depends on the testing that you conducted.
If your testing altered the product in any way, you might need to discard it.
But if your testing is non-destructive, and you maintain lot traceability with the items, you can consider putting them back into the lot.
Also think about the mixup risk. If you're testing similar items, and there is a risk that when you put them back into the lot you accidentally mix them up, then you might want to just toss them.
There's not a hard and fast rule for the samples, and it really comes down to your business and the risks.
@@CQEAcademy Thank you Sir for the explanation.
Another question:
One more thing, how about the retention sample, how many percent
should we take retention sample
(for finished product). Is it any guidelines or ISO can I refer?
Hey @@MuhammadFirdaus-yd5dw!! There is no hard and fast rule about sample retention. It really comes down to your own internal requirements, or the requirements of your customer.
@@greenbeltacademy Thank you Sir.
@@CQEAcademy AQL 1.0%, 2.5% and others.
Can we own decide for the percentage AQL?
How can I decide which AQL will I choose for different lot size?
Hey Eltot, the AQL you pick should be based on risk, and how many failures you can tolerate for the attributes that you're inspecting for.
Great job nailed it. Can you please let me know when to use single , double and multiple plans thanks
We have 15 different test for same sample so do I have to devide 200 sample by 15 or 200 for each test ?
Hey Shakir!
You have to take 200 samples total, but as long as none of the tests are destructive, then you can take the same 200 samples and test them for the 15 different attributes that you are inspecting for.
If the tests are destructive, then you'd have to take 200 samples per attribute.
Also in terms of single/double/multiple, that's a choice you get to make.
If the process of taking additional samples is easy, then a double/multiple sampling plan is likely a good choice.
But if taking samples is alot of work, then the benefits of a double sampling plan are reduced, and a single sampling plan is likely the best choice.
@@CQEAcademy wow that's quick response,
appreciate your help
Thanks a lot.
@@shakirsaiyed8733 You're welcome!
Hello, I have a question for you. Is it possible to apply testing standards (ANSI Z1.4 for example) to a full batch of production? I mean using all the population to be tested? I know it will require more time and resources but we know very little our third party (manufacturing company).
Hey Fernando, yes the ANSI Z1.4 standards are intended to be applied to completed batch production.
I'm assuming here you want to test product produced by a 3rd party manufacturer.
If you want to test product that you are producing, I'd recommend using a control chart, over ANSI sampling, to control and monitor your process. It'll give you a more advanced warning to issues occurring in the process - and not waiting until the entire batch has been produced.
How do you know to default to general inspection level II and not I or III?
That default is defined in the ANSI standard
Hi Andrew,
on what basis the AQL is chosen, I mean what is the criteria to choose AQL 1.0 as in your case?
thanks
The AQL depends on the price of the product, and the effect of a possible failure. There are tables that recommend possible AQL levels depending on these 2 attributes
What the square symbols indicating on the Double Sampling Plan table? (For example, Code H at AQL 0.25) thank you.
That Square symbol indicates that you should use the corresponding single sampling plan for that AQL/Code Letter (H at 0.25).
Sir you videos are amazing helping me for preparing the exam .
I Have a CQE book get it from ASQ latest version but the book has not thse sampling table .
How I will get these tables? They will give me in the exam or I have to bring my own separately ?
Hey Jatinder, you have to bring your own separately
Thanks for your reply.
They said they will accept CQE Book .
Are you sure they will accept sampling plan book too ?
is there any other kind of book I can bring with me ?
The
@@jatindersingh8796 Yes, the exam is open-book, and there are no limitations around what resources you can bring into the exam. The only thing you CANNOT bring into the exam are question/answers/solutions texts.
Thank you very much for your support.
Last question sir . What else useful book or tables I can bring in the exam. ?
Where can i get a copy of the sampling tables for the cqe exam 🤔
Hey Alfred!
It sounds like you're elbow deep in preparing for the CQE Exam.
If you have a study resource, it should contain those sampling tables.
If you don't have a sampling resource, just email me at Andy@CQEAcademy.com and I can share the most important ones with you.
-Andy
Are these the only tables needed for the exam?
Hey there, great question!
So, likely yes, these tables will most likely covers the question on the exam.
However, there's a small chance that there are other tables that might be required.
The best thing to do is to have the ANSI Z1.4 standard with you during the exam, or some other equivalent resource that has the tables.
Is it enough to have MIL-STD-105 if don't have ANSI/ASQ Z1.4 ....if I use both stds for solving same problem will it be same answer?
Andy your video are really knowledgeable and precise.
this concept is so confusing because how can some one will accept a lot even with 1 defect.
Depending on sigma, some industries doesn't need 0 defective, for example company making pencils, if 1 defective no issue, but if it medical company it's big issue so zero defective important