I need some help here.

My registrar told me of a quote from Juran (?) saying that in 100% inspection there is an 80% chance of finding the problem (if it exists). If three people inspect the item and they know that someone else has inspected it, the chance of finding the problem drops to 60%.

Does anybody know if these numbers are correct and where that quote came from, if it was indeed from Juran?

Contrary to what the intro to this forum says, I don't have a copy of Juran's Quality Handbook, mainly because my boss is a penny-pinching cheapskate :mad:, but that's another story.

Thanks!

RCW,

Dr. Juran noted that inspection was at best, 80% accurate. Others have guessed less, but that doesn't really matter. Think of it this was. One inspector is .8 in 1 chance likely to find a problem. The second inspector has the same chance. However, if you take .8 and multiply it by .8 (the second inspector), your outcome is .64. With the second inspector, your odds are reduced thus making 200% inspection 64% likely to find a defect.

Dr. Deming offered this in regards 200% inspection: Inspector 1 thinks inspector 2 will catch the defect while inspector 2 thinks inspector 1 will catch it. Net result: nobody inspecting.

I hope this helps.

Kevin

I don't have a copy of Juran's Quality Handbook, mainly because my boss is a penny-pinching cheapskate

RCW - can't garuntee anything I'ts been awhile since I looked in my library archives - but I may have a copy you could use.

If yer interested Shoot me an E-mail & I'll look around this evening.

RCW,

I'll sell you my book for $60.00!!!

100% inspection is as bogus as saying my gage does not have wear and shouldn't be calibrated.

If your boss is that cheap you are in trouble. Try this as a test:

1: Take 10 parts that you know are flawed;
2: Employ 3 random employees to measure the parts;
3: Tell the trio that there are 5 good and 5 bad parts;

(if you really want to get sneaky call it an R&R test)

4: Record all off their results and review the data.

Reply back to the forum after you present the results to your boss.

Originally posted by RCW

My registrar told me of a quote from Juran (?) saying that in 100% inspection there is an 80% chance of finding the problem (if it exists). If three people inspect the item and they know that someone else has inspected it, the chance of finding the problem drops to 60%.Kevin explained the guts, but remember: This is only true for human inspection. A 100% poka yoke (mechanical/electronic/electro-mechanical) 'inspection' is a different animal.

> I don't have a copy of Juran's Quality Handbook, mainly
> because my boss is a penny-pinching cheapskate

You may not like me saying this, but it's worth it to buy it for yourself. For me it's mainly a paperweight now, but if you looked at mine you would see it is well worn, marked up all over and has saved my butt several times. When I sat for my CQE over 10 years ago it was close to a bible for me. To me it is a tool. I wouldn't want one from an employer unless it was agreed it woud be mine to keep. I want it in MY library! :thedeal:

Originally posted by Al Dyer

I'll sell you my book for $60.00!!!What edition?

I stand corrected. Thanks Marc!;)

Browse ebay for Juran's hanbook. Not too long ago I found a fifth edition copy selling for an opening bid of $60 (oddly enough). There are many great deals there folks for books for a small fraction of what they originally sold for. In addition, you can shop amazon or borders used books. Please read the fine print!! Many copies are in terrific order, others have highlighting and marked pages. I bought a back up copy of Out of the Crisis for $10 including shipping. In all honesty, this is in better condition (still stiff when you open it) than my own copy at home which I paid $60 for.

Happy hunting!!

Kevin

Thanks for all the responses!

Here's a little background info on my original question:

1. Upper management and the sales dept. have always bragged to customers that we only use 100% inspection and no SPC is used here. (We are a small electronics firm that produces small quantities (1 - 10 pieces) of circuit boards, cables or wired panels, so it is hard to justify using SPC for the small quantites).

However, the big-wigs make it sound like 100% "visual" inspection catches everything, which it doesn't.

2. Recently, due to excessive workloads, product has been getting out of here with minor defects (i.e. missing labels, missing i.d. stamping). So the directive from upstairs is that more people have to check the product before it goes out the door. This brought to mind the quote my registrar told me about that adding additional "eyes" to inspection actual reduces the odds of finding errors. That's why I wanted something in writing to show the disbelievers.

3. I'm also planning to use Al's suggestion of the mock R&R study. While your example suggests measuring parts, I'm sure that it can also be applied to visual inspection (i.e. finding missing solder from connections).

4. As far as getting a copy of Juran's book, I'm trying to obtain a copy through an inter-library loan, to examine the book for merit. And no, I'm not insulted by the suggestion that I buy my own copy. It's just that (and I don't mean to step on anybody's toes) I'm an electronic technician and a computer science major who is masquerading as a Quality Manager. Through a bad decision of my own and a raw deal from the then president of the company, I ended up being "the quality department". It's because of this that I can't see dropping my personal money on buying the book.


Much thanks to those out there in this forum for all the help and input received for this lost soul. It warms my heart to find a place where I can find practical information instead of glossed over double-talk.

I sure don't want to argue with the anybody on this issue, especially Juran. IMHO I think the "100% inspection is only 80% effective" statement is a cop out used way too much to justify a poor inspection process.

The original study which prompted the statement was done on highly technical circuit boards and the results in such a case were understandable. You can't apply the same statement to an inspection line determining if refrigerators had been coated with red paint.

Obviously, the degree of effectiveness of any inspection is dependent on what you are inspecting for.

I am not a statistician but the logic that the chance of a second inspector finding a single defect would be lower than the first inspector doesn't seem to make sense. The chance of finding the defect would INCREASE for the entire process (using mechanical inspection to eliminate the "he thought - she thought" sceanario).

A process has 100 defects - Inspection 1 (80%) effective finds all but 20. Inspection 2 (80% effective) finds 16 of them. Inspection 3 (also 80% effective) would find 3.2 of the remaining 4.

Can the argument be made that depending on the sensetivity of the inspection, a lot inspected 3 times will catch almost all the defects (ruling out the human side issues mentioned by Kevin and Deming)?

The validity of inspection is dependent on both the criteria for inspection and the process (people/machine). The more complex the inspection criteria, the more intense the process has to be.

If you are inspecting the color of refrigerators, don't be afraid to use the inspectors in Kevin's example. If you are inspecting something else you had better be sure the people you use are dependable, conscientious employees. Short of that, get a good vision system.

When all else fails, remember - IF 100% INSPECTION IS ONLY 80% EFFECTIVE, SORT IT 3 TIMES !!!!!!! :biglaugh:

Dave

Marc,

Sorry Marc, it is only the fourth edition but has hardley been opened.

Originally posted by D.Scott
I sure don't want to argue with the anybody on this issue, especially Juran. IMHO I think the "100% inspection is only 80% effective" statement is a cop out used way too much to justify a poor inspection process.
I am not a statistician but the logic that the chance of a second inspector finding a single defect would be lower than the first inspector doesn't seem to make sense. The chance of finding the defect would INCREASE for the entire process (using mechanical inspection to eliminate the "he thought - she thought" sceanario).
Dave

My sentiments exactly! As an Inspector-Full Time for 14 years, and still am occasionally, I find the assumption that Inspectors are generally prone to not performing as they should, crap! I wonder if the great master Juran ever spent time doing this type of work. Maybe when this study was done, he was standing behind the Inspectors, no doubt looking down on them, with his clipboard pacing back and forth. :))
When I had an assignment, it made no difference to me if it was looked at prior to my inspection. Most serious Inspectors assume that there is something not right and will not let it go until they are completely satisfied that it meets all the requirements. We are a paranoid group. I have been following this thread and this 80-20 rule, another buzz word I might add, that is used for just about anything these days. Not wanting to ruffle anyone's feathers, nor offend the Juran aficionados, I was waiting to see a post that questioned what appears to be acceptance by the majority that Inspectors are lazy, or worst yet, incompetent. Thanks for a breath of fresh air. Just because it's printed and preached doesn't always make it so! :smokin:

I think Mr. Scott's point is a good one (and I expected energy to weigh in ;) God bless his conservative soul). I can't disagree. I know they teach this with the 'find all the F's in this paragraph dealie' (one version of which is posted here somewhere. Someone - I forget who - probably Eileen {she is a wonderful contributor here} - uploaded it) which does show - and I believe - inspectors do miss things from time to time. I'm less sure about the double inspection aspect of the theory.

It seems to me most important would be the inspection. As Mr. Scott pointed out it is hard to miss whether a refrigerator got the red paint or not. Another issue is how long and how many. If you have 0.5 minute to inspect a circuit card with 10 critical features to be inspected, and you have to do that for 2 hours streight, you start increasing the odds of inspectors missing a defect.

I have worked in several places that would only allow an inspector to 'work' for 2 hours and after a break would have to inspect a different part. These were in Mil houses doing circuit card inspections. The rational was that an inspector can 'get used to' the part and after a while not see some defects.

In another thread someone asked: "...how much receiving inspection should we use?..." I haven't read the whole thing yet (I saw it in my e-mail and have yet to get to that thread) - so I might be missing something - but it depends upon the part, it's criticality, etc. There is no single answer in a forum setting such as this unless a lot of info is posted for people to have a reasonable understanding of the situation as a whole. I mention this because I would never hesitate to use a visual inspection when it is appropriate - which is what you have to determine. If your NC-CA system shows parts are getting by a visual inspection, regardless of the "80-20 Rule", you investigate the problem. for example: Did this just start happening (i.e.: was visual inspection working before but is not now)?

You may have to give inspectors more time or more frequent breaks. It may be the best way is a poka yoke. But the bottom line is visual inspection is not worthless. You have to look at the inspection and the context it is in.

As a last comment, I have seen non-visual inspections fail. In one scenario there was a simple continuity check. When we looked at the situation, it was mostly a matter of boredom and part the human brain. After a while the inspector 'thought' s/he heard the buzzer when in fact it had not buzzed. The inspection was simple and it would be too expensive to break up the line and insert an auto test machine. We slowed down the number of inspections per person per hour and added an inspector and the problem dropped dramatically (not entirely, but to an 'acceptable' level). Final test was an automated test fixture which 'exercised' the CCA which precluded any earlier inspection failures from leaving the plant.

Originally posted by Kevin Mader
Dr. Deming offered this in regards 200% inspection: Inspector 1 thinks inspector 2 will catch the defect while inspector 2 thinks inspector 1 will catch it. Net result: nobody inspecting.
Kevin

Hi, Me Again:eek:

I agree with the premise that Inspectors do get tired. The degree of fatigue is proportional to the complexity. But the statement that an Inspector won’t inspect if he knows another Inspector has or will inspect the same thing, amazes me. This isn’t fatigue, boys. This is outright fraud.
Having shown my righteous indignation of suggesting that Inspectors may be not inspecting, there has to be reason for double inspection. I used to be a “double inspector” working on an aircraft assembly line. Two inspectors would begin inspecting flight control cables, turn buckles, control rods, etc., from opposite end of the aircraft. The lead seal was attached to all adjustable linkages that allowed for two Inspectors to each apply their personalized stamp. You would eventually pass each during the process. Does that mean I would stop inspecting because I saw his stamp on the seal? I would just stamp it? Give me a break! Then there was third Inspector who verified that all the seals contained a double stamp impression and that everything looked good. He/she also had overview authority and experience in this area. So, there is reason to double, triple check things. Is it fatigue? I agree. If it’s because they didn’t inspect it all, absolutely unbelievable:ko: :smokin:

3 inspectors inspecting the same thing = .8 x .8 x .8= .51 or 51%

I see that I may be swimming in the deep end alone here, but I'll take my chances.

p.s. energy, you would still jump in an save me if I went under, wouldn't you? :eek:

Kevin

Originally posted by Kevin Mader
p.s. energy, you would still jump in an save me if I went under, wouldn't you? :eek:
Kevin

Kevin,

Anything for a fellow Connecticut Quality Dude! You're just quoting what's published. :)) You don't believe everything you read. CT Yankees are the worst:rolleyes: :smokin:

Energy,

It might not be a case of fraud; it may be a case of too much faith in the other person. If I know you are a good inspector, I might not work as hard as if I had little faith. Is this fraud? Perhaps.

In another situation: A video duplication discovered they mislabeled a bunch of videotapes. Because of records, they knew exactly how many were nonconforming. They sent the inspectors out to the warehouse to find them. On the first pass they found all but a dozen. The second pass lowered the number of missing tapes to 5. The third pass found two more. The fourth and fifth passes could not find any. “There are none left. We know this.” The inspectors claimed. A new batch of inspectors was sent in to find the remaining 5 tapes (Not a typo). They returned saying they could only find three (which was all that remained). The first group of inspectors couldn’t find the mislabeled tapes because they thought they found them all. Consciously, or not, they quit inspecting.

Not the same thing, but the same result.

Dave B (the other Dave)

db

OUCH!! That smarts. I thought we were tight?:ko: :smokin:

Heres more food for thought:

Recently on the news, I heard that after recounting the votes, Gore did win Florida.

Gore wins, then loses on a complete count, then wins again (sort of) on a recount.


Sample results - 100% inspection results - 100% inspection results.

What result do you believe in?

Not the Dems!!! :)) :)) :smokin:

Energy, and Kevin

You know it almost sounds like we are saying exactly the same thing differently. My concluion is that in everything from videotapes to elections, we can't count on 100% counting.

Dave B (the other Dave):))

Good example, db.

I seem to come to grips with understanding 80-20 thing from the "outside/flipside" of the box way of thought.

energy - 80-20 has nothing to do with tired - or unqualified - or unmotivated inspectors - or bad inspection techniques.
its just plain more difficult to find bad ones when there are fewer to find. (unless of course we're talking the difference between black & white paint.) Even "top management folks" could find 5 elephants - amongst 25 donkeys. ;)

Discovering say 4 known bad parts out of 100 gives me 25:1 chance of finding them - If I find & remove 3 and do a second 100% inspection then my odds are 97:1 of finding the ones that are left.

Once again - I am not a statistician - BUT - multiplying the 80%s doesn't seem to be valid here. I agree you would multiply the probabilities to determine the probability of something happening a given number of times in succession. I don't believe this is the question for this situation.

In this instance, we are looking at an 80/100 chance of finding a defect on the first pass PLUS a 16/20 (80%) chance on the second and PLUS a 3.2/4 (80%) chance on the third. The chance of discovery increases with each addition to a 99.2/100 probability that the defect will be found. :confused:

Jim's example would only be valid if samples were scrapped randomly (ie. pick out one - chances are 1 in 25 that it is a bad one).

Of course, I could be wrong. Who is our stats whiz? :ko:

Dave

Just for the record, I checked my 4th ed of Jurans handbook and was unable to determine that he made the comment that 100% inspection is only 80% effective. What he did say about 100% inspection is outlined on page 18.25.
Maybe someone is confusing this with his development of the pareto chart.

The Pied Piper story is all too familiar to me! Let's keep it going. It made for a lot of interesting and thought provoking posts. Show me the money! :p :smokin:

Just read through this post this morning.

While I was reading it my wife piped up from the other room about the how the government is wanting to federalize the airport checking posts. She says, "CNN keeps reporting these incidences of people getting by".

I responded, "How many stories are you hearing about them finding things."

"None".

"Well then we don't know how effective the current system really is. Do you think that the Federal Government will be more effective?


Now I don't want to get political here, but this issue seems to fit right in with the discussion we are having here.
How effective is the current system?

Just some more fuel for the..... MMMM,,, I mean food for thought.

James

Inspection efficiency is made up of many factors, those contributing to the success of detection, the others contributing to noise. Juran’s estimation of 80% efficiency is based on the noise factor interfering with 100% detection in his personal experiences over the many years in industry. Think of Man, Machine/Equipment, Method, and Material. Now think of the possible noise contributing to the inefficiency of any one of these inputs. If Juran is right, then all the noise will reduce my ability to detect a defect by 20% regardless of the fraction defective in the lot. The lower the fraction defective, the lower my probability of detecting it. This is also an exponential function, not a linear regression.

James presents a good topic for discussion: the efficiency of private screening services at airports. Think about Man, Machine/Equipment, Method, and Material. Will the federal government do a better job in any one of these aggregates? Marc’s comments in correcting me are right on the money. What can be done to the process to poka yoke it (this will reduce the noise factor and increase detection)? What low cost/no cost modifications be done immediately to improve security? What more costly solutions can be instituted provided there is data to support the expenditure?

Suppose there is a 95% efficiency in the category Man, 95% in Machine/equipment, 95% efficiency in Method, and 95% efficiency in Material. What is the System’s overall efficiency (they are all interrelated)? Here is something posted by Don Winton (our Stats Wizard) on September 28, 1998 that explains System's Thinking:

“…Possibly. I prefer to think of this particular case as a lack of systems thinking. Each of the solid boosters had three joints. The specification read that each joint had to have a less than 2% probability of failure during any single firing. The testing verified this to be the case (satisfied the specified requirements). Now, let us look at the system requirement.

Each system contained two solid rocket boosters, each with three seals that had met the 98% probability of success. What is the probability of a system failure?

This is defined as (0.98)^6 = 0.886 or 88.6% So, for each system launch, there is a 88.6% probability of success!!!!!! By the time STS-56 was launched, it only had a 50% chance of success. That is a lack of system thinking personified!…”

Did this just muddy the water?

Kev

Kevin,

A good set of ear muffs with a 33RB Protection factor should take care of the noise factor:bigwave:

In the case of the seals, just like your underwear, you change them to keep the "protection" factor up. Also, saves the waistband:ko: :smokin: You know I'm just being a "prisoner" of my limited capacity to comprehend statistics. Forgive me!:)) :smokin:

YUP !! :ko:

Thanks for trying though.

Here is my scenario - The bad part gets past the first inspector, OK nobody's perfect. The part gets past the second inspector too, I get a little skeptical. The part gets past a third inspector as well? Somebody is not paying attention - period. We are going to spend a few quality moments together.

Then again, in our business it isn't too far from elephants in a pasture of donkeys. Anything critical, we use vision systems.

Dave :bigwave:

Am I a pessimist or does anyone else feel that putting the federal government in charge of security is going to cost us way more than revamping the private security practices? Even if the private companies brought wages into line, hired citizens that passed a background check etc., don't you think it would be cheaper for the American Taxpayer?

I don't think that having a federal employee doing inspections is going to be any more effective than a private employee (provided he is being fairly compensated.) Actually, having spent some time around govt. employees, I'd guess that with the rampant entitlement attitude, the govt. employee might be more prone to slack off knowing that it takes eons to oust someone who doesn't do their job.

Now, I don't want anyone who is a govt. employee to get offended, there are a lot of them out there that are good, concientious people. For them, I say good work and thanks. But I have met a lot through the years that just plain don't give a hoot about their job other than that retirement check they will be eligible for.

Sorry for bringing politics into it...I just think that the government gets too involved in too many things it has no knowledge about.

SteelMaiden,

I agree that industries need to govern themselves, but not all industries adhere to ethical, moral, and legal practices. I would say that government here should assess how well the airline industry deals with security and levy fines where they aren't making the grade. It would be far less costly to audit practices than it would be to use taxpayer dollars to support an airline security division.

Kevin

Who wanted federalization of airport security! Everybody knows who!;) These people do not care one iota about your security. Never did. It's all about about votes! As far as tying this into the competency of Inspectors, well, have at it!:p

Sifting through angry people's laundry does not equate to the topic of Inspection of a company's product. Stretch as you may, justify by similarities to make your point, workers at a bench with all the tools and plenty of time to do their job do not miss defects because of some pre-ordained sanctified premise. Remember, all you who never get prompted by spell check or some other fail safe device, "To err is human-To blame it on someone else is more human" or "Don't criticize us unless you walk a mile in our penny loafers". It is so easy to to explain from the altar why the populace has failed to measure up. How about this one? "He who lives in glass houses, shouldn't get stoned" Human error, no excuse for a CAR response, is real. You do it, I do it. What do you do? Quadruple Inspection? Are you making bottle caps or sending a man to Mars? Inspectors Rule!! Amen!:eek: :smokin:

Originally posted by SteelMaiden

Am I a pessimist or does anyone else feel that putting the federal government in charge of security is going to cost us way more than revamping the private security practices? Even if the private companies brought wages into line, hired citizens that passed a background check etc., don't you think it would be cheaper for the American Taxpayer?

I don't think that having a federal employee doing inspections is going to be any more effective than a private employee (provided he is being fairly compensated.) Actually, having spent some time around govt. employees, I'd guess that with the rampant entitlement attitude, the govt. employee might be more prone to slack off knowing that it takes eons to oust someone who doesn't do their job.

Now, I don't want anyone who is a govt. employee to get offended, there are a lot of them out there that are good, concientious people. For them, I say good work and thanks. But I have met a lot through the years that just plain don't give a hoot about their job other than that retirement check they will be eligible for.

Sorry for bringing politics into it...I just think that the government gets too involved in too many things it has no knowledge about.It's not that you're a pessimist. It's that you've bought the 'big lie' that private industry is more efficient that government. Look around. Look at Ford. Look at Enron. Look at the airlines (failing before 11 Sept).

Large companies / corporations are no more efficient than Government employees and government 'departments'. You say:

> Actually, having spent some time around govt. employees,
> I'd guess that with the rampant entitlement attitude, the
> govt. employee might be more prone to slack off knowing

Would you like me to start with the stories about the fraud, waste and abuse in big business? Ask yourself this... If tomorrow you lost your job but were quickly hired for a government position would you auto-magically become stupid and not care as people like to project onto government employees?

If you believe big business is more efficient, profitable or whatever, you haven't looked very far or worked in many companies.

> that it takes eons to oust someone who doesn't do their
> job.

Unions often serve the same functions. But before you start downing unions, make sure you know your history of labour relations and what companies did to employees (especially Ford, the coal industry and the garment industry) - especially during the 1920's and 1930's here in the US. Companies had their own private armies. People were killed / murdered. Some people really do believe in protection of workers. You may not like unions - I'm not particularly a union fan - but they do provide a balance you may not recognize.

Nor are unions more corrupt than politicos.

I will also point out that there are an awful lot of CEOs out there who are 'hard to get out' before they come close to ruining the company. Then - just like in the government (pension or whatever) the disposed CEO gets a Golden Parachute which most of us could retire on. Many Golden Parachutes are big enough that many of our kinfolks could retire with us.

> But I have met a lot through the years that just plain
> don't give a hoot about their job other than that
> retirement check they will be eligible for.

I see this every day in companies. Do you really believe this is exclusive to government? I could go on - but you get the idea.

The biggest difference is that government is supposed to be of the people, by the people and for the people (which we all know is not the case in this day and age - nor was it in the 1920's and early 1930's) while 'private industry' is interested in their profit (as opposed to the people and population). Take a look at how far GE is going to keep from having to clean up the toxic mess they made in the Hudson(?) and other rivers.

Unions came about precisely because 'private' industry left unchecked (the government not doing it's job) typically cares most about their profits and ends up abusing the employees.

> I just think that the government gets too involved in too
> many things it has no knowledge about.

Private industry never does this????<hr>In so far as the inspection aspect, I don't believe it will make much if any difference at all whether it's government or private run. This is a case of tools as well as understanding a problem that was not perceived to exist prior to 9/11. Prior to that, anyone could carry on a pocket knife, box cutter, etc. The legislation is to calm the populace - and not a reaction to a failure of the screening system. It didn't fail its intended purpose.

In 1973 the ALPA (Air Line Pilots Association) practically demanded fortified cockpits including reinforced doors. They even proposed small gun ports (like on armored cars). They said in 1973 - this is what to protect in so far as hijacking goes. Guess who deep sixed the idea as far fetched and stupid? Ummm, the government? Nope - the airline companies. Of course now.... almost 30 years later.... cockpits are being reinforced.

That said, the 'screening issue' is a smokescreen. There was no failure of the screening (inspection) system with respect to 9/11.

Originally posted by energy

Who wanted federalization of airport security! Everybody knows who!Get real. It is a shame for you to say something like this.

Let me get this straight. The Dems (your "...everybody knows who...") want votes. They get votes from the people (except in Florida). From your comment legislators should say "...screw the people and what they want - we'll do what we think is best for them. And cheaper is always better...." You know very well this is silly. It's pretty apparent that the people in this country will feel safer with government involved whether it is truely the case (security and safety will be 'improved') or not.

It is also apparent the 'Repubs' want votes as well. They went along with it when they could have stopped the bill. Seems to me the Repubs want votes every bit as badly as the Dems.

> These people do not care one iota about your security.
> Never did. It's all about about votes!

Let me see here... I want your vote but don't care about you as a person. Oh, what shall I do? Did it occur to you that at least some politicians do care about their constituents and what they want? Even some Dem politicos?

Apparently you and Rush believe the only people in the US who care about other people are the Repubs. That's a pretty shabby accusation. It's totally unfounded and you know it. It makes good spin, but it's BS.

How you cast the quest for votes in a negative light is telling - I thought that was the idea of our government --> a democracy where votes count and that votes represent the will of the people. If I follow your logic, then we should rid ourselves of this stupid system where votes are (according to you) a bad thing. Geezzz - politicians looking for votes. Imagine the stupidity of that - only the Dems would do that! Let's dump elections and go strictly by political appointment or by dictatorship. We can bring back torture and put the Bill of Rights and the US Constitution on hold while we're at it. We don't need no stinkin' elections with all those stupid votes.<hr>This said - back to inspection - the topic. Let's all drop the political aspect of this thread. There are plenty of political sites where you folks can throw political darts.

Outside of a little preface about the driving force to federalize the airport workers, it was about Inspection. I forgot how touchy I can be.
mea Culpa:( :smokin:

Do I multiply the Democrats by the number of votes or add the Republicans to my flight number to determine if my luggage will get lost?

You guys need to meet someday for lunch. When it happens, please invite me too. I'll pick up the check.

Dave

Originally posted by D.Scott
YUP !! :ko:

Thanks for trying though.

Here is my scenario - The bad part gets past the first inspector, OK nobody's perfect. The part gets past the second inspector too, I get a little skeptical. The part gets past a third inspector as well? Somebody is not paying attention - period. We are going to spend a few quality moments together.

Then again, in our business it isn't too far from elephants in a pasture of donkeys. Anything critical, we use vision systems.

Dave :bigwave:

Gotta chime in here... :bigwave:

If the part gets past a third inspector, I would strongly suggest looking at the tools the inspectors have to work with, or the process by which they have been trained to inspect parts.

If the process and the tools are adequate to detect the non-conformance - start looking around the environment for the cause of the problem.

Here's a little story about just such a thing. When my wife and I were living in Miami, FL - she decided that an Iguana would be a fun pet to have (BIG mistake - impulse purchase). We bought a cage and a heat rock and figured that Miami being tropical, the iguana would do just fine on our back porch. Well... a week later, we had a very very sick iguana and couldn't figure out what was wrong. We had the right temperature, we had the right water, food, etc., etc., etc. The vet suggested just sitting on the back porch and "absorbing" the environment... What was it that was causing the iguana to become ill? After about 10 minutes out there, we had it. Our apartment was on a main thoroughfare with fairly heavy traffic. We had gotten used to the traffic so we didn't pay any attention to it. However, the noise and car exhause was more than poor Iggy could handle. So, to make a long story short, $600+ dollars later, Iggy was fine and had taken up residence inside. :smokin:

I think spending a few quality moments just "absorbing" the environment where your people work would be time better spent.... ;) However, if it comes to light that the people are just goofing off... I have a few choice quality words to share with them... (Ooops... my Navy is starting to show again) :biglaugh:

Cheers!!!

Originally posted by D.Scott

You guys need to meet someday for lunch. When it happens, please invite me too. I'll pick up the check.
Dave Dave,

Actually, “us guys” have had an opportunity to talk for a substantial amount of time one Saturday afternoon. We discussed contentious moments and misinterpreted posts due to the e-mail method of communication. The “Smilies” have been a big help. Had I used these “smilies” in some of my posts, I’d have more times at bat. We’re both passionate in our beliefs and pretty much agree to disagree. The fact that I’m still here (I can be expunged with a flip of a switch) should give you an idea of the degree of tolerance. Thanks for invite to dinner, though. Maybe some beaver fat Armadillo with squirrel brains stuffing accompanied by a glass of road kill crow squeezings. It don’t get any better than that!

Admin: Sorry I let the cat out of the bag.

Originally posted by energy

Outside of a little preface about the driving force to federalize the airport workers, it was about Inspection. I forgot how touchy I can be.
mea Culpa:( :smokin: Yes - I AM a testy little **** from time to time, aren't I! Yeah - energy and I do go a bout now and again. We tolerate each other I think with respect to politics. We've spoken and I think we pretty well get along. We just disagree politically.

As an aside, I'm a former pilot - well, I still have my license but haven't flown for close to 20 years now. In the 1970's I in part worked my way through college doing charter flights and some mail runs now and again. I know airport security evolution to some degree from my experiences. Back in the 1970's there were security gates at most airports for us pilots. What we didn't like is that every airport had the same combination on the lock - to (supposedly) make it easier for us pilots. Airport security has always been more show than effective. That's why when I look at airport screening - I don't care who does it. It's more a matter of technology than diligent personnel. I personally feel safe on airplanes - even now. Statistically speaking, you're safer up there than you are driving or even rambling around your own house.

Originally posted by D.Scott

Do I multiply the Democrats by the number of votes or add the Republicans to my flight number to determine if my luggage will get lost?Neither - the airline companies pre-arrange that by their own methods... :lick:

First off, I think this thread does lend itself to this particular angle on feralizing aiport security workers. Exactly what are they suppose to do...INSPECT carry-ons and people. They have the tools to do their job, but because it is "privatized", the bottom line ($$$$$$$) is the highest priority. These people are there to provide for our well being and safety, much like the police officers and firemen of our towns and cities. My thoughts are that airport security should be turned over to the police departments of the respective cities. In Chicago, the police patrol the airport and have full access to all areas. Why not add some of Chicago's finest to O'hare and Midway? These men and women are trained for this type of work. Rotation would have to be heavy to prevent burn-out. Personally, I would rather see security handled locally than federally.



stepping off my soapbox...

CarolX

Hello All,

Just got back to the thread today.
I must apologize for setting us off on a political tangent.:eek:

My intent was more along the line of what tools are to be used; how can the system be improved etc.

I too come from an inspection background. But I accept the rule of 100% inspection being 80% effective as a "rule of thumb". Although it probably doesn't hold true in most cases anymore. I think that it was developed and used to kick old style management types into new ways of thinking about quality control. You know, prevention and SPC and training etc. instead of detection.

As far as airport screening goes, I'd have to weigh in on the private industry side. With Federal oversite.
Also I'd like to see maybe more canines used for explosives sniffing etc.

I see many references in the threads to wages. Yet the information I have seen on motivation indicates that wages are not the most significant factor in job motivation.
I feel no safer with a high paid screener than with a low paid screener.

Perhaps the screeners should be shown films of Lockerbee Scotland, and the Pentegon, and the Twin Towers etc. You know like we were shown in Drivers Ed Class.


James

:rolleyes:

Here is my take, all of you are wrong, and all of you are right (good political answer!) Here is what I have found. Who an individual’s employer is, and how much the individual gets paid has less to do with performance than the individual. A rock solid performer is a rock solid performer. A slacker is a slacker. We can banter all day long with horror stories on both sides, and all it will do is add to my argument. The same thing applies to any inspector. If I don’t have a reason to perform a “good” inspection, I probably won’t. One rock solid inspector can out perform a dozen slackers.

I just wish I knew which category I fit in.

Dave B (the other Dave)

> I must apologize for setting us off on a political
> tangent.

No big deal. Nothing for you to apologise for. All the regulars here know I tend to be easily set off by what I perceive to be political darts. I apologised to energy in a PM. All is well. On with the show. :thedeal:

mea culpa...

Busted again, yep I have to admit that since I graduated from college, I have been limited in the number of companies I have worked for. That would be one corporation, three divisions, and yes sometimes I base my views on that. I am extremely fortunate in that we are pretty much the industry leader and have a very high job satisfaction rating overall. So, I apologize for sounding anti-anything, but I'd still rather have either private security or local policing of our airports with some set of govt. standards.

I am willing to just say, we disagree and that's ok, it doesn't make either of us right or wrong. ;)

Have a good one

Yea, and I apologized right back. Some are little testy and some are a little taunting:)) :smokin:

I have been involved in one way or another with inspections for a big number of years. I can sure buy into the statement about Rocks vs Slackers. One good inspector is worth a lot.

The point Kevin was making though is still valid even with the greatest inspector in the world. There are outside conditions which make it "impossible" to be "perfect".

I remember a horrible discovery on one of the inspection belts we use. Parts are fed from a hopper onto a traveling belt past the inspectors. Rejects are removed and the good parts follow along the belt to a holding bin. There is a switch allowing the belt to be stopped when necessary. I had been part of an inspection group and had been sorting for a short time (about 10 - 15 minutes). I can't remember why, but the belt was stopped for a minute. To my surprise, the parts kept right on moving down the belt. No, the parts didn't really move, it was an optical illusion caused by staring at the previously moving parts.

This experience got me looking into more of the tricks the mind plays with a moving belt. It is not only the belt that causes illusions. As part of the same study I placed my keyring, loaded with keys in the hopper. They fed through, onto the belt and past the inspector into the bin. The inspectors brain was "programmed" for a number of defects but the mind didn't recognize my keys as one of them.

No, I didn't blame the inspectors. I had them inspecting in an environment which promoted "hypnotic" influences. We now run the parts onto the belt, stop while we inspect, then run into the bin.

I am sure others have even wilder stories.

Dave (not the other Dave)

> The point Kevin was making though is still valid even with
> the greatest inspector in the world. There are outside
> conditions which make it "impossible" to be "perfect".

I agree. To me, this is the problem with a lot of what goes on even with ISO 9001 (not to mention QS-9000 or Ford's QOS - There's a lot of rhetoric with these). There are a lot of theories, but reality gets in the way. Unfortunately the complexity is such that it is similar to weather forcasting - there are so many variables that we come up with predictions.

And then someone (like me) goes out and tells people of things like the 80-20 rule (or whatever) as if that is always true. Unfortunately one person in the audience may have, in their company (and in their mind), an inspection like the 'red refrigerator' example. That person may even think the 80-20 rule is BS. I mean, who can miss that? Meanwhile the next person is thinking of inspection within their company which is inspection of solder joints on printed circuit cards or trying to judge gloss on a grained surface (where gloss meters are notoriously unreliable).

These forums were started because of reality. I would add to Dave's statement that not even poka yoke is always 100% 'correct'. Sensors can fail. A software bug can show up - and maybe it's only once or twice a month.

I just want to comment on the probability portion of this thread. D.Scott actually start down the right path on page 3. The only scenario I worry about is if all three inspectors DON"T catch the problem. If the probability is 80% of catching it, then it must be 20% of NOT catching it.

The probability of all three inspectors NOT catching the problem is 0.2 x 0.2 x 0.2 = 0.008, which means 99.2% chance of someone catching the problem. Adding inspectors improves the probability of catching problems, not reduce the probability as originally stated by the auditor.

Seeing that I originally started this thread, I would like to put in the other portion of my two cents.

I do not believe that the 80-20 hypothesis holds up for every inspection. MOST of the time I would suspect that the accuracy of inspection within any company is above 80%.

However, I have seen within my own company, multiple people assigned the responsibility of verifying characteristics of a product in production. Please note that I did not say they were inspectors. They are people in test, inspection and shipping. Product has passed throught 3 different departments and gone out the door with discrepancies.

It has been upper management's opinion that "more people need to look at the product" before it goes out the door. There is never anything said regarding a) defining specific details to look at, b) assigning specific people to look at specific details or c) investigating what actually went on to determine where the problem exists.

Within there is a "bull-in-a-china-shop" attitude to hurry up and fix things before investigating the problem. Henceforth, just throw more people at the problem and it should go away.

Adding more "inspectors" does not always work. I've seen it done here and it sure doesn't (work).

Hence my original question was to determine the source of the 80-20 rule to present it to upper management, armed with the fact that while the 80-20 rule isn't always an absolute number. In the same manner, 100% inspection does not always yield 100% success, even with multiple inspection points.

That's the bottom line. Now I'll get down and put my soap box away.

I have a piece of paper somewhere that says I'm a statistician of some sort. Won't comment on the political issue.

Mark R and D. Scott are right on in their analysis.

I'm afraid Kevin committed what some wag has called the Error of the Third Kind, that is "solving the wrong problem" . :rolleyes:

If we wanted to know the probabilty that all three inspectors would find the defect it would be .8x.8x.8 = .512 - but as Mark and Dave suggest, what we really want is the probability that all three inpectors will miss the defect. That is .2 x.2x.2 = .008.

Try a tree diagram. It has eight branches. You will find one branch where all find the defect , one where all miss it, 3 where two find it and one does not and 3 where two miss it and one finds it. But the only one we really care about for the probability of letting it out the door is where all three miss it.
The tree diagram shows each inspection being performed regardless if the defect was found on the one before or two befoire etc. But that must be to allow all the probabilities to sum up to 1. That's the way this universe works mathematically! :eek:

Dave,

Word-problems have tripped me up from time to time: I guess I was fortunate my high school math teachers gave me partial credit for setting up the equation!:bonk:

I guess the real question comes down to the use of detection over process controls. Can the organization afford the use of mutli-inspectors/appraisers? Is sampling a more effective means? Can the process be error proofed?

Returning to RCW's problem, management seems to think that more people looking at it is the answer. To one degree, they are right. But to a larger degree, I think they are wrong. The end must justify the means, and adding cost with additional inspectors is probably not the answer. There are better means to this end.

Thanks for the correction and the use of the tree diagram in the explanation.

Regards,

Kevin

Many years ago when I was at Raytheon I did a study on inspection accuracy. As part of a twice a year recertification, solderers and solder inspectors were given a board and asked to identify any solder defects. The people doing the inspection were not told how many defects were present, if any. Typically in these tests, some 20 defects would have been created.

I checked the test results over a period of several years and dozens of employees. The actual average number of correctly identified defects was very close to the proverbial 80%, although some people were as low as 60% and some were at 100%.

Almost as disturbing as the misses was the high number of good solder connections identified as unacceptable. In a production situation these would have been sent to rework, causing extra expense and potentially impairing reliability.

These were all good and diligent people. It's not a question of motivation or attention span or antying other than human nature and variability.

Originally posted by John Finn

Many years ago when I was at Raytheon I did a study on inspection accuracyVery interesting study, John. It's nice to see that the 80-20 rule holds up to more than just casual scrutiny. I'm a strong proponent of the 80-20 rule. :D

You mentioned that you were distrubed by the number of good solder connections ID'd as unacceptable and that in production these would have been sent to rework. While I cannot comment on your particular study, I can offer my opinion about what you may have seen.

Personally, I don't think as many of the good solder connections would have ended up in rework as you may think. First off, these people were taking a test. There may be an underlying assumption that there will be mistakes on the board - regardless if you have told them that a board may have no mistakes. So, given that these people are under more pressure to perform (the test) and that they may not necessarily believe that they have gotten a board that has no defects, they will conjure defects where none exists out of fear of missing one. Again, mindset may be that it is better to put a good part into rework than let a bad part get out of the plant, depending upon the paradigm on the shop floor. Also, depending upon the potential ramifications of failing the test, someone may opt to err on the side of more caution and find an error where none exists...

What would have been very interesting to see is the following data plotted against each other:

1) Number of defective boards sent to rework 2 weeks prior to testing
2) Number of defective boards sent to rework 2 weeks after testing
3) Number of defective boards sent to rework 3 months after testing

Anyway - these are just my thoughts on an interesting study. Thanks for sharing it... :vfunny:

Cheers!!

It is better to have an A error and call good ones bad rather than a B error and call a bad one good. If you think the expense is high now, wait till you err in the other direction.

All our vision systems are set up to "find a good one" and let it through. The gate stays closed and rejects everything unless it is triggered by a good part. If something goes wrong in the process, everything gets rejected and re-inspected.

It is much cheaper for me to find the error than for my customer to find it.

I tell my line inspectors "When in doubt - throw it out".

Dave

Originally posted by D.Scott
I tell my line inspectors "When in doubt - throw it out".
Dave

What ever happened to my version.." When in doubt-stamp it out"? :vfunny:

We just finished our annual inventory. The last chore was to re-visit all the aisles and look for boxes with no Inventory tags on them. Some of the boxes are quite small, so you have to lift existing tags and look at every container. The first one who checked was "ME". :( Then a warehouse person and then the Purchasing Person who was running the show. Right after that we go and tear off half the tag, matching ends, for entry into a database. Well, there was a box with no tag on it. We located the tag in the "non issued tag-cannot find them" pile. Three of us looked without seeing it and the fourth found it while not responsible for looking for it. (Like a Customer after triple inspection). I know it's not like actually inspecting the part, but this was easier and we missed. As I had strong words for those who said inspectors miss 20% of what they look at, I have to munch down some more humble pie. Boy, am I getting fat.:smokin:

Originally posted by energy
As I had strong words for those who said inspectors miss 20% of what they look at, I have to munch down some more humble pie. Boy, am I getting fat.:smokin:

Tastes nasty, doesn't it... :bigwave:

The only thing worse than humble pie is crow... damn feathers get stuck in the throat every time...:vfunny: :ko: :biglaugh: :D

Have a great weekend!!!!

Energy,

I was a pretty good inspector, once upon a time. But with all the "and and" or "the the" typos in my procedures, event though they were read and reread several times, one would get the impression that I wasn't a very good good inspector after all.:eek:


Kev

Kevin,

That's why we have spellcheck:ko: :smokin:

Maybe an attitude check could be put in place for me!:bigwave:

Spellcheck? Whatt's thatt?:bonk:

Several years ago, I developed a training program. It was several days in length, and at the end of each day was a slide for "Questions and Anwsers". No one noticed it during spell check (perhaps that slide wasn't checked), three proofreaders missed the typo. The client missed the typo. It was only during the third running of the class that someone brought it our attention.

If we inspect the thought, and not the words, we might miss a lot.

Dave B (the other Dave)
:p

I agree with the general sentiment that when an inspector is not sure of the acceptability of an item, he should identify and segregate it from good production. However, Quality has a role to reduce or eliminate, if possible, the element of doubt. Throwing out potentially good product because of uncertainty also has a cost, both to one's organization as well as to society as a whole.

My first management job was with a a captive molding operation within a medical device manufacturer. Our division molded disposable plastic parts that were sent to another company which filled them with a reagent, sealed and packaged them, and delivered them to customers of our medical systems.

Our inspectors were throwing out all doubtful product, which when I was hired amounted to 40% of what was coming off our molding presses. The good parts were then sent to the other company whose inspectors again rejected what they thought was doubtful quality and returned those parts to us. The overall yield of good product throughout all this was less than 50%.

One of the main causes of rejection was in an optical surface of one of our parts. Working with our Engineering people, we were able to define the specific critical area within the optical window that had to be defect free. We were then able to make go-no go gages to inspect for this. Another area of rejection was flashing from the molding operation which could affect the amount of reagent dispensed by our equipment. Again working with our engineering people we wereable to define very precise specifications on this. We also were able to give the molding process engineers a realistic target to shoot for when optimizing the molding parameters.

There were many other things that we did, including standardization of our production equipment, instituting process control checks, improved maintenance, etc. Within one year, we had our process yields up to 95-98% good parts and our returns from the other company down to less than 0.2%.

Sorry to be so long-winded, but this is a subject that I feel strongly about.

My point is that unless the quality department does something positive (like help to make the situation better), they will never get the respect that they deserve and will always be in an adversarial relationship.

John - Agree wholeheartedly. However -

The inspection process, no matter how robust, is going to include a fine line between good and bad. This line must always exist in order to justify inspection in the first place. At some point in the process the best inspector in the world will meet a situation where one of these borderline decisions has to be made. IMHO the only acceptable decision at that point is to err on the side of the customer i.e. "throw it out".

If, as you say, cost/scrap/rework or whatever become an issue, it must certainly be addressed.

Dave

Dave,

Thank you for your response and for having the patience to read my lengthy diatribe. Certainly I agree with the premise of protecting the customer.

I guess my point is this. If you are producing product that is on the fine line of good - no good, then there is essentially no difference to the customer between product that you ship and product that you reject. Think of two students; one averages 69 for a term and gets a failing grade, the other averages 70 and passes. Is there really any difference between the two students? I don't think so. What the teacher and the student need to do is get the averages up to 75, 80, etc. so that it is clear that the student has actually learned enough to pass. The same applies to manufacturing: make the process so good that there is no doubt.

Lets not forget that the best form of assuring product quality is to make it right first time. Having to inspect is undesirable, but any method is better than none, assuming that the inspector does find some defective pieces.

I dont agree with the Deming quote that you should either 100% inspect or not inspect at all. He totally dismisses sampling plans as a waste of time. This is one thing I cannot agree with Deming on. He fails to appreciate different products and different markets. There is a time and a place (cost effective) for sampling inspection.

Does anyone agree ?

I heard on the radio this morning a school had the state name spelled wrong on basketball jerseys for six games before anyone noticed. If this is true, it adds fuel to the fire that we see what we want to see. :bonk: :vfunny:

Dave B (the other Dave)

Suppose that you choose a sampling scheme with an AQL of 1.0. What you are saying is that if my supplier has his process under control at a level that produces 1.0% or fewer defects, then I want a sampling plan that will accept those lots most of the time (typically at least 95%). I am comfortable with 1% nonconforming.

Now suppose that in fact the supplier has his process under control and is producing at 1.0% or better. If you do no inspection at all, you will receive product at the level that you want. Some of the lots will be more than 1.0% bad, some will be less than 1.0%, but the average will be 1.0% or better. If you sample inspect, you will occasionally reject good lots due to sampling error and incur all the costs associated with that, but your product will not be any better because of your inspection.

What if the supplier has his process under control, but is producing product at a defect rate of greater than 1.0%? While you might make some improvement overall, you will never get the nonconforming level down to less than 1.0% by sampling. The nature of AQL plans is that they are designed to protect the producer of product. So a 1.0 AQL plan will usually accept product at 2%, 3%, etc.

If the supplier does not have his process under control, then acceptance sampling plans really become a crap shoot. You might as well randomly accept or reject lots.

If your intent is to make sure that, say, there are no more than 1.0% defectives in any given lot, then you need to use something like an LTPD plan (lot tolerance percent defective). Unlike AQL plans, this protects the buyer, not the producer. An AQL plan says "If it's close, give the edge to the producer". An LTPD plan says "If it's close, give the edge to the buyer". However, LTPD plans require a much larger sample size and for many practical applications is essentially 100% inspection.

Originally posted by db

I heard on the radio this morning a school had the state name spelled wrong on basketball jerseys for six games before anyone noticed. If this is true, it adds fuel to the fire that we see what we want to see.I have an excellent, but old (like from 10 years ago or more) video here somewhere on paradigms which shows this to be true in several experiments. It's a part of how our brains take in information, including visual, and processes it. Often the brain causes a person to see what the person expects to see - not what is really there. It's one serious, valid (I believe) arguement against eye witness testamony in court trials.

For those of you into magic - you know exactly what was happening and understand the phenomena. This phenomena and distraction are the basis of most magic 'tricks'.

Marc,

At the risk of getting too esoteric or philosophical, everything that we see is really a creation of our brain. All that our eyes really "see" are light waves of various frequencies and amplitudes from some object, these get communicated electonically to our brain, which somehow interprets all this into a picture of the world. The picture that we see, however, is not out there, but totally in our heads. There is really no way of knowing what the world "really" looks like. No wonder that eye witnesses disagree so often.

> At the risk of getting too esoteric or philosophical

Not a problem.

Yes - I understand these issues and the physics involved. And one can use philosophy to explain certain things and to produce some arguements along the lines of what we 'see' is - well, we really can't say because we can't know what is real and what is not. Unfortunately, philosophy can take things to the extreme of "...do we even exist or are we some part of the imagination of someone else or something else..."

My major in college was biology with minors in chemistry and anthropology. My senior thesis was on neuron cell dendrite to dendrite and axon to axon connections - which in the 1970's was poorly understood. The context was always axon to dendrite. The purpose and interaction of dendrite to dendrite connections, for example, was just starting to be investigated. I have a relatively strong background in human physiology and biological processes to the molecular 'engine' level.

> The picture that we see, however, is not out there, but
> totally in our heads. There is really no way of knowing
> what the world "really" looks like.

For me what ties together the validity that what we see is pretty close to what is there is that two people can look at the same object or scene and each will describe the same basics. So even if you and I 'really' see (for example) the color (frequency) red differently, we will both describe it as 'red' because that's what we learned as youths. There are some interesting studies in colour blindness which also come to bear on issues like this.

Secondarily is tactile feedback of the nervous system to certain things. Stick your hand in a fire and I'll bet it won't be there very long. I do not believe this is solely a part of a system of learned response. If you interview someone who has been badly burned they will argue pretty strongly that the pain and tissue destruction is real and not a figment of someone elses imagination.

One thing I did learn that I think is really interesting is that the spinal cord its self learns. Responses to certain stimuli over time bypass the brain entirely. In some cases the brain is 'notified' of the stimuli-reaction 'event', but with some stimuli the brain doesn't get so much as an e-mail to let it know the stimuli-reaction 'event' took place - the routing is entirely limited to the spinal cord.

Bring on the esoteric, philosophical and/or technical talk any time. I'm ready!

Hi Marc,
Thank you for your encouraging response. You obviously know more about the science of this phenomenon than I do. (My undergraduate degree is in electrical engineering).

I, of course, believe that we have to behave based on the information that we receive from our senses. Otherwise, we wouldn't survive. There is also obviously a correspondence between what we see and "reality" or we couldn't function. The fact that we all appear to see things the same way, however, doesn't mean that we see it correctly (whatever that is), only that our brains, if functioning properly, work the same way.

I don't mean to imply anything mystical. I believe that reality exists, we are not dreaming, we are not somebody else's dream, etc. It's just the idea that we can function so effectively with what is essentially a series of pictures in our head that fascinates me.

I also have to confess that I/m trying to get enough posts to get out of the "shy" category. Thank you for the interesting conversation!

I don't know if you believe in evolution - I know many people do not - but I'm a believer. From humans to roaches to porposes to shrimp. All have evolved (some may use the term adapatation) some wonderful 'features'.

> I also have to confess that I'm trying to get enough posts
> to get out of the "shy" category.

Go to the UserCP and then click on 'User Profile'. In there you will see where it says: "Custom User Text". Put anything you want in the box that says "New Title". Then 'Submit' the changes. That should make the 'shy' title go away.

The beauty of this medium: self-directed evolution, from "shy' contributor to "Red Sox Fan", although some might call that "de-volution"! Thank you for the advice.

I don't think that it is possible to be a scientist/engineer without believing in evolution. Even understanding how the process works does not take away the wonder of it all.

Red Sox Fan is alright by me!! It's a tough job and someone has to do it!!(I do my part)

Another tidbit in regards the 'devolution' of this thread. Our senses give us feedback, data and information that our mind must process. While we can learn to keep our hands out of the fire by sticking it into a flame (movement), we can learn through other means and self reflection that sticking our hand into a fire will cause us pain (motivation).

Pop behaviorism is the belief that our worlds and behavior are shaped in our minds through external forces. B.F. Skinner, a late Bostonian, promoted this thinking. However, this is being challenged by another Massachusen, residing outside of Boston, named Alfie Kohn. His message is pretty powerful and his evidence staggering.

In regards M Greenaway's post, Deming's view on 100% inspection and his harshness was founded on his disappointment with the US reliance on the use of sampling plans (the same plans he helped to develop for use during the war) after WW2. These plans were created to help expedite the inspection process as well as reducing the expense of time and money to the producer. They were also created to improve detection of defective product/lots since inspectors and methods for 100% inspection where fatigue, gage error & maintenance, and manipulation errors were found to contribute to higher error rates.

50+ years later, we are still highly dependent on "detection." Many QMS standards are still very suggestive of inspection over process controls. While this fueled Deming's concern, he certainly acknowledged that inspection was inevitable in certain situations (he used the example of computer chip inspection for rating the chip's speed). He was quoted as saying, "Do 100% inspection, or do no inspection at all."

Deming suggests, "Routine 100 percent inspection to improve quality is the equivalent to planning for defects, acknowledgement that the process has not the capability required for the specifications. Inspection to improve quality is too late, ineffective, costly." OTC, page 28

What he wanted was for management to pursue quality through process improvement and preventive actions. Toyota and many other Japanese organizations use this method, much to the detriment of Western organizations and industries (Toyota requires certain workers to wear white work coats. Reason: when Western visitors ask, "Where are your inspectors?" they point out the folks in white. This came about because Western visitors did not accept that Toyota's high quality came without the assistance of teams of inspectors). Sadly, in my opinion, our transition is painfully slow. We have to rely on inspection because we do not develop systems and supplier relationships that will allow us to eliminate inspection. Reason: the time commitment to supplier and process development neccessary to supplant inspection is too long for short term profiteers!!

Regards,

Kevin (CT Sox Fan:bonk: :)

Kevin,

It's always a pleasure to make acquaintance with another sox fan. Could there be a connection between the frustration of being a sox fan and the frustrations of the quality life? By the way, I thought that everyone in CT was a Yankee fan!

Your comments got me to thinking a little about why the Japanese (and some American firms) can eliminate inspectors and still achieve high quality results. I don't believe that it's a matter of process or training or knowledge, but more of attitude and trust. Management trusts people to do the job right and the people have the attitude of responsibility and ownership that is necessary to make it happen. How do you develop that environment?

Interestingly, many times I have heard even line people express great dread when the idea of eliminating inspection was broached. Yet these same people would be appalled if you suggested that someone should inspect the results of their cooking or the tune-up that they just did on their car or the way they do any of the myriad tasks that they do in their daily lives.

I'd really be interested in hearing ideas on this subject.

We don't have any inspectors. Each operator inspects as a part of his/her job function. Management expects everyone to do the right thing the right way at the right time, and everyone is really conscientous about it. There is an obvious show of pride and ownership among our operators. I know that it doesn't always work that way in every company, but in all the divisions I've worked for here it works every time. I think I've said it before, but it is a culture kind of thing. If people are trusted and given the responsibility and authority, they usually meet the challange. I think that given the opportunity, most people will live up to, or down to what is expected of them if the expectations are realisic.

Originally posted by John Finn
Kevin,
It's always a pleasure to make acquaintance with another sox fan. Could there be a connection between the frustration of being a sox fan and the frustrations of the quality life? By the way, I thought that everyone in CT was a Yankee fan!


Only North of Kevin's town!:rolleyes: They beat up Red Sox fans where we are! :vfunny:

We also do not have Inspectors here. Just me. It takes an Inspector to properly fill out records. Most assemblers and general laborers could care less about filling out paperwork. As long as there is someone who can provide accurate records, you don't need Inspectors. Therein lies the rub. Long Live Inspectors! Inspectors Rule!:ko: :smokin:

Energy,

I'll be sure to stay away from Naugatuck! By the way, some of my best friends are inspectors.:truce:

Energy works in New Britain, once the home of the Red Sox triple A team. It may not be a coincidence that the Sox left with a heckler nearby!! Now I am concerned about my favorite hot dog vendor!!:eek: :biglaugh:

SteelMaiden wrote: If people are trusted and given the responsibility and authority, they usually meet the challenge.

I agree with this statement. However, senior management needs to guard against the delegation of duties/responsibilities that should not be. Too often duties are delegated and the consequences (negative) are left for the worker to bear.

John wrote: Management trusts people to do the job right and the people have the attitude of responsibility and ownership that is necessary to make it happen. How do you develop that environment? (in reference to the Japanese)

I believe the clear distinction between the Japanese workforce and the Western workforce is this: understanding what their work means. This comes from the education of the workforce and customer focus. Management trust comes from two way feedback and the elimination of fear in the workplace (reprisal for giving negative feedback). Here in the States, you only bring the boss the good news!! Where fear is present, you will find manipulated figures and distrust.

Well, just a bit to get things stirred up again…

Regards,

Kevin

Thanks for the information about the origins of effectiveness. The truth is a lot less simple! How difficult the inspection is decides how likely you are to find the problem. The example of "Is the fridge covered in paint?" is a good one. The likelihood of picking up an unpainted fridge when the primer is white is approx 100%. If the requirement is "Check the red paint coverage across the whole surface is a minimum of 0.25 mm thick and we're giving you a tape measure to check it" then the likelihood of finding a non compliance is somewhere near to 0%. Any other combination of checks and capability falls somewhere on or between these two markers.

The next point is that you cannnot add inspectors and bring down the probability. i.e 80% probability followed by 80% probability is not 64% probability of finding the problem. You have to take the opposite view. If the first inspector misses it what is the likelihood of the next inspector missing it 20% * 20% = 4%. Therefore the probability of two inspectors catching a fault becomes 96%. This assumes that the two inspections are not connected. i.e. the first inspection isn't cursory because "The next person will inspect it" or the 2nd inspection is cursory because "Well it's already been inspected."

I hope this has helped.

Dont know where that calculation of probability comes from but it sounds very flawed to me.

If the probability of missing a defect is 20%, then it is 20% for everyone. You could have ten inspectors checking the batch, each has a probability of missing the defect of 20% therefore the overall probability is still 20%.

Does one inspector missing a defect make it more likely that a following inspector will catch the defect - I think not, the calculation is flawed.

I think the poster is going for the probability of the combination rather than the individual probability of the second inspector.

MG is right - the probability of the single inspector is always 20% (simple event), but in the poster's example the second inspector is dependent on the first inspector missing the part (multiplicative law). To determine the probability of both events happening you have to multiply the probabilities (in this case 20% x 20%) which gives you the probability of both inspectors missing the defect (4%). The inverse would be the chance of discovery (96%).

When it is looked at that way, it makes more sense.

Dave

9 pages of discussion!! This thread is giving the Humor thread a run...

Kevin

Cant get my head around this one.

I can appreciate that if an inspector picks one piece from a batch and the probability of selecting a defective piece is say 20%, that if he selects another piece the probability of them both being defective is 20% X 20%, and the probability that either is defective is 20% + 20%. But I cant see how this simple theory relates to two inspectors that are unrelated by the process.

Its a bit like saying that the probability of me winning the lottery is 1/14000000, so the probability of someone else winning is 1/14000000 X 1/14000000.

It just dont make sense.

Martin -
We aren't asking what are the chances of winning the lottery. We are asking what are the chances of both winning succesive lotteries.

Make a simple tree diagram. The first inspector has an 80% chance of finding it and a 20% chance of missing it. Make a forked branch with one labeled 0.2 for the chance of missing it and the other branch label 0.8 for the chance of finding it. This is the first inspector. Now, make a second fork ON BOTH BRANCHES and label in the same way. (the second inspection would not be done in the real world on the 0.8 branch because the defect was already found, but we need it to count all the probabilities).

The branch where the defect was found by the first inspector has probabilities of 0.8 times 0.8 = 0.64 (multiplicative rule for independent events) for the case where BOTH inspectors find the defect and 0.8 times 0.2 = 0.16 where the first finds it and the second does not. On the other branch where the first inspector missed it , we have 0.2 times 0.8 = 0.16 and for the case where both miss it we have 0.2 times 0.2 = 0.04. Now adding them all up we get 0.64 + 0.16 +0.16 + 0.04 = 1 which confirms we have not missed any cases. ( that's why we had to "do" the second inspections even when we found it on the first).

The only case we are interested in is the one where both inspectors miss the defect. This is the 0.04 probabilility. So we can say there is 1-0.04 or 96% chance of the defect being found by two inspectors versus 80% chance of one finding it.

Each inspector has a 80% chance of finding it but both together in sequence have a 96% chance.

So your chance of winning the lottery or anyone elses in THAT lottery could be 1/140000000. The chances of you winning the lottery AND any other individual winning the next lottery (both winning in sequence) is 1/140000000 times 1/140000000.

But the chances of ANY body winning within any one drawing is the chances of each summed. So if 1000000 tickets are sold , the chances of one of them being the winner is 1/140. (assuming the numbers played are all different.) That's why people do win, but your individual chances are low.

Hope that helps.

I just fell out of the tree. :ko:

Dave

Originally posted by Dave Strouse
Martin -
We aren't asking what are the chances of winning the lottery. We are asking what are the chances of both winning succesive lotteries.

Make a simple tree diagram. The first inspector has an 80% chance of finding it and a 20% chance of missing it. Make a forked branch with one labeled 0.2 for the chance of missing it and the other branch label 0.8 for the chance of finding it. This is the first inspector. Now, make a second fork ON BOTH BRANCHES and label in the same way. (the second inspection would not be done in the real world on the 0.8 branch because the defect was already found, but we need it to count all the probabilities).

The branch where the defect was found by the first inspector has probabilities of 0.8 times 0.8 = 0.64 (multiplicative rule for independent events) for the case where BOTH inspectors find the defect and 0.8 times 0.2 = 0.16 where the first finds it and the second does not. On the other branch where the first inspector missed it , we have 0.2 times 0.8 = 0.16 and for the case where both miss it we have 0.2 times 0.2 = 0.04. Now adding them all up we get 0.64 + 0.16 +0.16 + 0.04 = 1 which confirms we have not missed any cases. ( that's why we had to "do" the second inspections even when we found it on the first).

The only case we are interested in is the one where both inspectors miss the defect. This is the 0.04 probabilility. So we can say there is 1-0.04 or 96% chance of the defect being found by two inspectors versus 80% chance of one finding it.

Each inspector has a 80% chance of finding it but both together in sequence have a 96% chance.

So your chance of winning the lottery or anyone elses in THAT lottery could be 1/140000000. The chances of you winning the lottery AND any other individual winning the next lottery (both winning in sequence) is 1/140000000 times 1/140000000.

But the chances of ANY body winning within any one drawing is the chances of each summed. So if 1000000 tickets are sold , the chances of one of them being the winner is 1/140. (assuming the numbers played are all different.) That's why people do win, but your individual chances are low.

Hope that helps.

I was just going to say that!:vfunny: :biglaugh: :bonk: :ko: :smokin:

Don't remeber where it cam from, but I always used inspecting 3 and one-half times gets 99% of the defects. May only be an old wives tale.

Originally posted by Rick Goodson
Don't remeber where it cam from, but I always used inspecting 3 and one-half times gets 99% of the defects. May only be an old wives tale.

So to get that 1/2 inspection does that mean someone is doing a half a##ed job, or can we hire one inspector that is half-blind?

I'm sorry, forgive me, I just couldn't help myself.:confused:

SteelMaiden,

I am not sure which is the best way. Maybe we could perform a DOE to determine which process (half-blind versus half a##ed) produces the best results.

As an aside, what part of NC? I use to work in Albemarle for a couple of years.

Albemarle region, at the edge of the Great Dismal Swamp.

Originally posted by Kevin Mader
RCW,

Dr. Juran noted that inspection was at best, 80% accurate. Others have guessed less, but that doesn't really matter. Think of it this was. One inspector is .8 in 1 chance likely to find a problem. The second inspector has the same chance. However, if you take .8 and multiply it by .8 (the second inspector), your outcome is .64. With the second inspector, your odds are reduced thus making 200% inspection 64% likely to find a defect.

Dr. Deming offered this in regards 200% inspection: Inspector 1 thinks inspector 2 will catch the defect while inspector 2 thinks inspector 1 will catch it. Net result: nobody inspecting.

I hope this helps.

Kevin

I think that a better way to do this calculation is to say that there is a 20% chance a defect will not be caught. Thus, if you add a second inspection, the chance that a defect will not be caught is 0.20 * 0.20, or 4%. Thus, double inspection is 96% effective, and triple inspection is 99.2% effective, statitically speaking. It doesn't make sense that inspecting more decreases your chance of catching a defect using normal statistical analysis.

However, the situation where an inspector knows that someone else is also inspecting the product may be another beast all together. It may be that when this situation occurs, 100% inspection is only 50% effective - no way to know unless you conduct your own experiment!

Originally posted by JMitchell

However, the situation where an inspector knows that someone else is also inspecting the product may be another beast all together. It may be that when this situation occurs, 100% inspection is only 50% effective - no way to know unless you conduct your own experiment!

Why don't we ask the Inspectors. I, for one, would never not inspect because someone else has already inspected. It's absurd. Make the same mistake? Possible. Particularly, if the drawing is unclear or subject to interpretation. That's happened. Supplier, Code Inspector, Source Inspector, Assemblers all missed a very important attribute in my previous life. They changed their drawing practices, not blame Inspectors. I've posted on this before and feel very strongly that no self respecting Inspector would do such a despicable thing such as not do his/her job. It does make a good subject for those professionals who have never sat at an Inspector's bench and performed such an important high pressure task. Inspectors Rule!:ko: :smokin:

A Blast from the Past!