1. This site uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn More.
Dismiss Notice
You must be a registered member in order to post messages and view/download attached files in this forum.
Click here to register.

Use of Normal Inspection in MIL-STD-105E

Discussion in 'Sampling, Standards and Inspection' started by isepals, Apr 29, 2020.

Tags:
  1. isepals

    isepals New Member

    Joined:
    Apr 29, 2020
    Messages:
    3
    Likes Received:
    0
    Trophy Points:
    1
    Hi there! I'm and undergrad currently studying quality engineering and I have a question regarding the use of MIL-STD-105E in industry.
    I was wondering what are the reasons for normal inspection being the only sampling plan used by some companies? I was thinking along the lines of ensuring that quality is retained while at the same time having a higher probability of acceptance according to the OC curve but I'm not sure.

    Secondly, I believe that such a practice is not advisable as the switching rules would follow the ideal OC step function and therefore is more preferred. Is this correct?

    Any opinion is greatly appreciated thanks!
     
    isepals, Apr 29, 2020
    #1
  2. Andy Nichols

    Andy Nichols Moderator Staff Member

    Joined:
    Jul 30, 2015
    Messages:
    5,399
    Likes Received:
    2,686
    Trophy Points:
    112
    Location:
    In the "Rust Belt"
    isepals: In all reality, very few people know HOW to use Mil 105. In my experience, they select an AQL they feel comfortable with, often because 100% inspection is too slow and costly and they want to get some comfort they have the process under control...
     
    Andy Nichols, Apr 30, 2020
    #2
  3. Bev D

    Bev D Moderator Staff Member

    Joined:
    Jul 30, 2015
    Messages:
    669
    Likes Received:
    726
    Trophy Points:
    92
    Location:
    Maine
    That was very kind and gentle of you Andy.

    I can’t do that today: Why do companies only use the Normal inspection plans? For the same reason they use the mil-std at all: Laziness. Pure and simple. Laziness of thought and laziness of action. Lazy Customers require it, lazy regulators accept it, lazy auditors look for it, lazy ‘quality managers’ reflex to it because that's what they’ve always done. Uniformity and conformity are safer and easier than thought.
    The standard is statistically and empirically flawed. But nobody cares because it’s what we’ve always done...

    That may sound harsh to some but I’m just tired of all of this old crap being accepted because too many people don’t have the intellectual energy to think, research, learn and change their behaviors. We settle for the easiest path forward and just repeat and defend the old dogma because it’s easier.
     
    Bev D, Apr 30, 2020
    #3
    Andy Nichols likes this.
  4. Andy Nichols

    Andy Nichols Moderator Staff Member

    Joined:
    Jul 30, 2015
    Messages:
    5,399
    Likes Received:
    2,686
    Trophy Points:
    112
    Location:
    In the "Rust Belt"
    Thanks, Bev! Yes, indeed. You are totally correct. To apply the requirements of a zero defect sampling scheme requires knowledge of the process and its output. IIRC, the use has to do a capability study (pretty much) and it takes time to understand, work with customers on what is acceptable and so on (I'm trying to keep away from the arcane jargon used). No-one does that. Indeed, to Bev's point about "lazy", I've seen a 0.4% AQL set (at receiving) items inspected (using variables data) and then the charts for attributes used. How lazy is THAT?
     
    Andy Nichols, Apr 30, 2020
    #4
  5. Mike S.

    Mike S. Member

    Joined:
    Apr 20, 2020
    Messages:
    37
    Likes Received:
    7
    Trophy Points:
    7
    While Bev may be technically correct, we could say the same things about a whole host of things, right? Why do people go to Starbucks for coffee? Why do people pay to have their oil changed? Why do people pay consultants?

    Some people are experts at statistics, like Bev, and it is therefore easy for her to generate custom sampling plans for most any situation, and generate statistics about the sampling plans, etc. But not every company has a Bev. In fact, very few do.

    I've used MIL-STD-105 / ANSI Z1.4 for 30 years. Most of the time as a customer requirement flowed down from customers ranging from Mom and Pops to many prime defense contractors. Most of my suppliers do the same for the same reasons. Almost every user (dozens, probably hundreds) used normal sampling only and avoided the switching due to complexity of getting the inspectors to understand the rules and keeping track of the number of lots, switching status, etc. The hassle wasn't worth the pain. And in the end, what matters most is not what sampling plan are you using, it is "does it work for you and the customer"? In all those years and users, maybe 1% used the switching rules, and the other 99% did not feel the hassle was worth the pain. YMMV.
     
    Mike S., Apr 30, 2020
    #5
  6. Andy Nichols

    Andy Nichols Moderator Staff Member

    Joined:
    Jul 30, 2015
    Messages:
    5,399
    Likes Received:
    2,686
    Trophy Points:
    112
    Location:
    In the "Rust Belt"
    I wouldn't agree at all with these "analogies". For many, coffee at Starbucks is better than a) can be made at home (without considerable investment) or is a quicker/convenient option. Nothing to do with laziness. People have their oil changed or use a consultant - like me - because they don't posses the resources/knowledge. Again, nothing to do with laziness.

    Most - generally speaking - don't know how to apply Mil-std 105. I HAD to learn because I inherited a legacy at RI where my predecessor set an arbitrary AQL to EVERYTHING at great cost to the company and I decided to do something about it. Learning what we were doing and then whether 105 was helping WAS HARD WORK. There aren't many classes where you can go to learn this, unlike Lead Auditor etc. Odd, that...
     
    Andy Nichols, Apr 30, 2020
    #6
  7. Mike S.

    Mike S. Member

    Joined:
    Apr 20, 2020
    Messages:
    37
    Likes Received:
    7
    Trophy Points:
    7
    So you can call someone "lazy" for not learning all the statistical knowledge necessary to create and defend the use of custom sampling plans on demand, but they are not lazy for not learning how to make a good cup of coffee or change their oil or implement their own ISO 9001 QMS?
     
    Mike S., Apr 30, 2020
    #7
  8. Andy Nichols

    Andy Nichols Moderator Staff Member

    Joined:
    Jul 30, 2015
    Messages:
    5,399
    Likes Received:
    2,686
    Trophy Points:
    112
    Location:
    In the "Rust Belt"
    Yes. Exactly right.
     
    Andy Nichols, Apr 30, 2020
    #8