|
This thread is carried over and continued in the Current Elsmar Cove Forums 
|
|
|
This thread is carried over and continued in the Current Elsmar Cove Forums
|
The New Elsmar Cove Forums
|
 The New Elsmar Cove Forums
 Statistical Techniques and 6 Sigma
 Six Sigma
|
 UBBFriend: Email This Page to Someone! |
next newest topic | next oldest topic |
| Author | Topic: Six Sigma |
|
Sam Forum Contributor Posts: 244 |
posted 28 January 2000 04:52 PM
   
An article in the recent (winter) issue of "Automotive Excellence" concerning Six sigma states that 3 sigma produces 66,807 defects per million. I'm confused, I always thought 3 sigma was approximately 2700 PPM. Is there a different way of selecting these figures, or was I just wrong all this time?
IP: Logged |
|
Don Winton Forum Contributor Posts: 498 |
posted 31 January 2000 07:09 PM
It would appear the author was referring to +/- 1.5 sigma, not +/- 3 sigma. Of course, I could be wrong. IP: Logged |
|
James Cupello Lurker (<10 Posts) Posts: 4 |
posted 02 May 2000 02:19 PM
Breyfogle's 1999 Excellent Book entitled IMPLEMENTING SIX SIGMA notes the following in Table S. 2,700 defects per million is the +/- 3 sigma rate for a centered process (i.e.-no 1.5 sigma shift). IP: Logged |
|
Marc Smith Cheech Wizard Posts: 4119 |
posted 05 May 2000 08:35 PM
From: (Jaws) Newsgroups: misc.industry.quality Subject: Re: SIX Sigma fad or fact Date: Fri, 05 May 2000 04:52:41 GMT I was on board at Motorola when they conceived the 6 sigma concept(s) and actually wrote the very 1st ever six sigma software. I believe the most useful part is the concept of design margins. For years the 'industry' look at histograms and said basically, if you were inside of 3 sigma limits, then your parts were good. Of course if PART A is at -2.5 sigma and PART B is at +2.5 sigma, there is a large difference when you start mating the parts together. To quote a Motorola hand out from about 1987 ... 'The performance of a product is determined by how much margin exists between the design requirement of its characteristics (and those of its parts/steps), and the actual value of those characteristics. These characteristics are produced by processes in the factory, and at the supplier. Each process attempts to reproduce its characteristics identically from unit to unit, but within each process some variation occurs. For more processes, such as those which use real time feedback to control outcome, the variation is quite small, and for others it may be quite large. A variation of the process is measured in Std. Dev, (sigma) from the Mean. The normal variation, defined as process width, is +/-3 Sigma about the mean. Approximately 2700 parts per million parts/steps will fall outside the normal variation of +/- 3 Sigma. This, by itself, does not appear disconcerting. However, when we build a product containing 1200 parts/steps, we can expect 3.24 defects per unit (1200 x .0027), on average. This would result in a rolled yield of less than 4%, which means fewer than 4 units out of every 100 would go through the entire manufacturing process without a defect. Thus, we can see that for a product to be built virtually defect-free, it must be designed to accept characteristics which are significantly more than +/- 3 sigma away from the mean. It can be shown that a design which can accept TWICE THE NORMAL VARIATION of the process, or +/- 6 sigma, can be expected to have no more than 3.4 parts per million defective for each characteristic, even if the process mean were to shift by as much as +/- 1.5 sigma only 0.0041 defects per unit (1200 x 0.0000034). This would mean that 996 units out of 1000 would go through the entire manufacturing process without a defect. To quantify this, Capability Index (Cp) is used. ...' This is indeed just the tip of the proverbial iceberg, and when I worked at Motorola, we had many, many classes that dealt with a larger concept. So keep in mind that I have been brain washed, (in a good way) to think this way. When you add the above to the other parts of 'Six Sigma' it does reduce cost and in prove product. I've seen it in action. Getting from +/- 3 sigma to +/- 6 sigma is a whole other ball game. It can't be done over night. Furthermore, one needs to use all the Six Sigma concepts together to make it work. There must be a reason why GE and now Dupont, I believe, have adopted these concepts. For what its worth, that's my $0.02 (US) Jim Winings IP: Logged |
|
Marc Smith Cheech Wizard Posts: 4119 |
posted 09 June 2000 04:54 AM
See ***DEAD LINK REMOVED*** [This message has been edited by Marc Smith (edited 09 June 2000).] IP: Logged |
|
John C Forum Contributor Posts: 134 |
posted 10 June 2000 09:36 AM
Jim. Thanks for that explanation. I attended a presentation by Motorola in the Apple Computer plant in Cork, Ireland, nearly 20 years ago, and have not dealt with it since, so I'd forgotten the point of it. Looking at discussions here, I have always felt sceptical but couldn't nail down the reason for my doubts. I felt generally, that 6 sigma applies where it applies and that 3 sigma is great where it applies. Now, take the cases you mentioned; A product with 1200 steps delivers only 4% good at 3 sigma. No argument - 6 sigma is much better. But is it totally appropriate? How much pain can the process take? Would 7 sigma be more appropriate? Now take the product with only one step. We can get better than 3% fallout at 3 sigma. Is it appropriate to tool up for 6 sigma? Maybe it is but, in most cases, I think it isn't. So, I would take each case according to it's merits and the guys that can make that decision are the guys that design the tool, make the product and use it, pooling their information to come to the decision. The idea of 'management' deciding that we'll have an 'across the board' 6 sigma revolution, is what seems to come over, just like a Baldridge hype project. Do they apply 6 sigma to the camshaft dimensions and the same to the hubcaps? When do the apply 4, 5 and 7 sigma? As appropriate? never? Surely, sometimes they are more appropriate? Is 6 sigma totally inappropriate for most assemby operations because it is easy to live with lower tolerances, and far cheaper? Is it possible to mould plastic to meet 6 sigma requirements? Would you, if you could? There is something I'm missing? The question is, is it just me? rgds, John C [This message has been edited by John C (edited 10 June 2000).] [This message has been edited by John C (edited 10 June 2000).] IP: Logged |
|
Jim Winings unregistered |
posted 01 July 2000 05:11 AM
John ·Yes and No! Six Sigma is much more than measurements. But, measurements are my fort?. If suppliers for the proverbial guys with Six Sigma requirements don't built parts to Six Sigma specification, then Six Sigma could never be implemented. However, when I worked at Motorola and would help suppliers, I found out that a lot of times they couldn't meet it due to 'low tech' measurement devices. I.E. dial calipers trying to measure .00005. It wasn't cost effective for them to spend 'lots-o-dollars' on better measurement equipment. For example· Maximum Specification .055 Minimum Specification. .045 Nom .050 Their data would be· .05 . 05 .05 .05 .05 Now depending on if youâre a purest or not, you don't have a process. I.E. there is NO absolute zero! But if when you make your measurement with your dial calipers you add .000005 or so to the measurement, it looks and calculates much better. So that is the first thing to remember when making measurements for Six Sigma. I mean whom is going to argue about .000005 when the Specification. is +/- .001. The second thing that I found out was sometimes meeting Six Sigma was just as simple as changing a Maximum or Minimum Specification. For example, if to meet Six Sigma your specification had to be .045 and .055 but your specification was .047 and .052, and your parts were all like .049 to .051 just opening up the specification to what you can produce can meet it. I.E. where ever +/- 4.5 or 6 sigma falls. Regardless of what you specification is on paper, this is what you are capable producing. And that is the bottom line. What can you produce consistently? That is part of the idea, is consistency. Does this make sense? Jim Winings [This message has been edited by Jim Winings (edited 01 July 2000).] IP: Logged |
|
John C Forum Contributor Posts: 134 |
posted 01 July 2000 07:28 AM
Jim, Well I'm not sure that I quite see what you're getting at but it doesn't seem to make sense. First;'there's more to 6 Sigma than measurements'. Well I see from other discussions that to a lot of people it's just a slogan for another hype program with numerical goals which will frighten people. A sort of anti-Deming exercise. So let's forget that aspect. Regards the measurements; This is not my field (maybe that's what keeps my mind clear on it?), so I'm surmising: I'm not sure if you are telling me that people have ways of getting around it and you approve, or that you don't approve. For example, if the calipers can't measure to spec then it's ok to massage the data a little to make it look right. Let's assume that you agree with me. This is my take on the issues you have raised; There's no sense in this world having 6 sigma as your goal if the product is not designed to 6 sigma. All dimensions have a nominal and + and - tolerances. The tolerances can be such that the parts always work together, sometimes work together or that the great majority work together. What we are looking at is tightening the tols so that all within 6 sigma will work together. (based on the known capability of the process to be used!) If your tolerances can't produce this, then you will never have 6 sigma whatever you do in the process unless you ignore the tols and set up your own arbirary ones. This is possible but it's not a very sensible way of going about the business especially when, as you say, knowing they are well within the tols, people break the arbitrary rules and move outside or, change the tols to meet what the system is capable of producing consistently. This is not 6 sigma, nor is it good practise. It would explain how, in one of the other discussions, someone noted how a company achieved and maintained 6 sigma and yet the customers were no happier. In plain terms, it's bluff. Back to the managers; I'm always knocking managers but, they are responsible and paid to take responsibility. Managers will allow this sort of 'Kings New Clothes' stuff to go on so that the company can claim 6 sigma and the share price will rise. At the end of the day, it doesn't matter a damn how it is done, if the share price rises - we can fix all those problems next quarter "the main thing is to be still here next quarter!". But the short term gains lead to long term losses. Win the local battle this year and lose the industry to the Far East over the next decade. Managers should understand their processes, take responsibility, provide clear aims. the right aims, meet them openly and work then to set and achieve even better aims. Once we start muddying the waters with personality trips then we lose our way. Good engineers seek the basic cause of problems and the basic cause can only be bad management. Bad things don't happen under good management - by definition. Deming damns this 6 sigma sort of stuff. By the way, what did he actually say? I never heard but I'm sure he must have had some strong comment on the business. You are right in saying there is more to this business than measurements. If you are one of the thousands of good and sincere people who do a good job and have to make the system work despite the bad leadership, then OK. Do what you have to do. It's a tough life making your customer happy when your supplier is incapable of seeing the spec but, in the absense of someone above seeing the light, then you have to keep on keeping on. I did this for years until I saw that all roads eventually came back to the same place - then I started trying to resolve the problems at base. Someone out there knows what Deming said about all this. Come in please. rgds, John C [This message has been edited by John C (edited 01 July 2000).] IP: Logged |
|
Jim Winings Forum Contributor Posts: 13 |
posted 02 July 2000 07:07 AM
>>Well I'm not sure that I quite see what you're getting at but it doesn't seem to make sense ... OK probably me. I'm in the middle of trying to get another release of software out the 'door' so I may have not explained things as well as I should have! (sorry) >>Well I see from other discussions that to a lot of people it's just a slogan ... >>I'm not sure if you are telling me that people have ways of getting around it and you approve ... No, that's not it. Just as with most things in life, or at least this is what I have found, there are more than one way to solve a problem. There are at least two ways to get to Six Sigma. And there is a lot evolved with it. When I worked at Motorola, they kept stressing that 'Motorola was NOT the expert in making their suppliers parts, the suppliers were'. This is indeed part of the Six Sigma concept. Becoming partners with your suppliers and/or customers. There is much more in depth discussion that should be covered here, but I don't want to eat up all of Marc's server space, and besides, I should be working on getting my late software out. You can tighten the process and conform to Six Sigma, or, if the final design tolerance can stand it, you can open up a specification. The actual specification is NOT relevant. What is, is the 'Design Margin'. So it doesn't really matter what the specification is as long as it meets the Six Sigma Design Margin. These are topics covered in several Six Sigma classes. To include 'Design For Manufacturability' and 'Understanding +/- 6 Sigma'. So if one hasn't had the full gambit of classes there are huge gaps in the concepts so it won't make much sense. For example, my software was design with a 'slider' that is adjustable between 3 and 6 sigma in .5 sigma steps. This allows one to tweak the process or specification and move the slide to the next' 'milestone'. But in association with that is the distribution chart. This is NOT your standard histogram. It plots your actual data and draws a perfect bell shape curve around it. Remember that this is all based on a 'NORMAL' distribution. Then it draws another bell shape curve around the Specification Nominal. So all you have to do is look at where the tails of the curves lay in relation to the selected design margin via the slider. This is what ones process is capable of producing at that point in time. It also tells you where your specifications should be for the selected design margin. >>Deming damns this 6 sigma sort of stuff ... Well, he has damned other things before as well. (Jeeze, I thought he died) Well ya learn something new everyday! >>All dimensions have a nominal and + and - tolerances. The tolerances can be such that the parts always work together, sometimes work together or that the great majority work together. What we are looking at is tightening the tols so that all within 6 sigma will work together. (based on the known capability of the process to be used!)... Exactly! >>This is possible but it's not a very sensible way of going about the business especially when, as you say, knowing they are well within the tols, people break the arbitrary rules ... And when they do, there is a 1.5 sigma shift to allow for that? >>This is not 6 sigma, nor is it good practise. It would explain how, in one of the other discussions, someone noted how a company achieved and maintained 6 sigma and yet the customers were no happier. In plain terms, it's bluff ... Well that is your opinion, and of course you are entitled to it. There are many things that can happen to any concept. However, since, I have no knowledge of what company, etc. so this is just here say and I can't commit on it. >>those problems next quarter "the main thing is to be still here next quarter!". But the short term gains lead to long term losses. Win the local battle this year and lose the industry to the Far East over the next decade ... The reason Motorola went to Six Sigma was because they were losing market shares to Japan. Six Sigma helped turn that around. But in general I can agree with your statement, I just don't know how true it holds with the Fortune 1000 companies. And the 'big guys' should have liaisons in place to monitor their supplies to insure that doesn't happen. >>Once we start muddying the waters with personality trips ... It doesn't matter if you work for a Fortune 1000 company or for a company with 4 people. There is, and I can assume will always be politics. I hate politics, so I agree with you, but neither you nor I can ever change that or the fact that people in quality circles will always have different opinions on what is the 'best' quality concept. Personally, I haven't seen any 'NEW' concepts since Deming's concepts in about 1945. To me, they are all just the same old things repackaged. ISOxxxxx for example. Nothing new here. I was performing most parts of it back in 1982 or so. This may not have cleared things up much, but I can't cover the thousands of hours of training and experience that 'Black Belts' have and that is what it may take to make it comprehensible. I'm not 'PRO or CON' on any of the concepts in quality. I just play with what the big guys dictate. And right now that seems to be Six Sigma. Jim Winings [This message has been edited by Jim Winings (edited 02 July 2000).] IP: Logged |
|
Jim Winings Forum Contributor Posts: 13 |
posted 02 July 2000 07:13 AM
Just a 'side bar' you might think is funny!!! Because the 'X' key is next to the 'C' key, I keep having typos that say ... Sic Sigma hehehehehehehehehehehe ------------------ IP: Logged |
|
Steven Truchon Forum Contributor Posts: 89 |
posted 12 July 2000 04:48 PM
Excellent! Another forum to tap into. Our company has been given a requirement by a major customer to "join" them in the 6 sigma quest. They want their X% price reduction and this is how they feel we should do it. I am told that I will be on the "team" which is okay but all I know about 6 sigma is a few buzz words, a basic concept, and knowing that now my 18 hour day will truly be full! (haha. Anyone have a recommendation for a book, video, cd, or even "6 sigma for dummies" perhaps? -Fumbling in Florida IP: Logged |
|
Andy Bassett Forum Contributor Posts: 274 |
posted 13 July 2000 02:51 AM
Wow Ive hit the wrong button on my computer and found my way into a dark and dusty dungeon on this website where i have never ventured before (and now i know why). Do you people really beleive that you can improve quality (and in the case of the last post, acheive cost reduction) via statistics. Do you not think it might be more useful to invest time developing the mindset or cost consciousness of the people producing the goods?. Regards (Just feeling awkward today) ------------------ IP: Logged |
|
Steven Truchon Forum Contributor Posts: 89 |
posted 13 July 2000 08:15 AM
To Andy, Statistics do not do anything except reveal results based on the data input. Do I expect to improve quality due to stats? Yes, if I answer the results from careful input with a correction in an improvement direction which may well include your suggestion regarding personnel, among other things. -Always feeling awkward (hey, Im on earth!) IP: Logged |
|
Jim Biz Forum Wizard Posts: 275 |
posted 13 July 2000 09:34 AM
Steve - Andy Oh boy something I competley agree with - Stats don't "do" anything by themselves - (my opinion only) - stats are simply a tool that will give us an information "hint" about the decision to be made. 1)Where we should look to improve 2)Where we should not have a reason to be concerned. And YES there is a "cost factor" difference between +/- 3 sigma and +/-4 or +/-12 The "tighter" one tries to hold widgets within increased sigma cpk or PPk boundries the higher the cost. As I see it that aspect is dificult to avoid & needs to be "justified" by actual observance of defects. After all even if a histogram predicts there are 6 or 6000 "probable defects" out there - how many in reality do I actually see when the "probable" defects are interacting with 4 or 5 other assembly variables that each have their own "probable" rate? - If I match a "probably small shaft with a "probably large" bushing - the function of the part may in fact be no different than the design calls for anyway.. IP: Logged |
|
John C Forum Contributor Posts: 134 |
posted 13 July 2000 11:29 AM
Andy, Statistics are very useful for improving quality and achieving cost reductions, in some cases more so than in others. Sometimes it is more useful to develop people. In practise, the right amount of both, plus a lot of other good stuff is the way to go. rgds, John C IP: Logged |
|
Jim Winings Forum Contributor Posts: 13 |
posted 13 July 2000 01:06 PM
Statistics is the cornerstone to improving quality. It gives you a visual picture, not to be confused with the non-visual picture, of your process capabilities. What you do or don't do with the information is the key to if it works. It also helps if everyone that touches a piece part, looks at it, not inspect it actually, but look for the obvious. It cost much less in both $$$$ and PR to find defects and defectives in house. ------------------ IP: Logged |
|
mike1245 (aka mike525) unregistered |
posted 19 July 2000 01:59 PM
Several months back (maybe longer) I asked in one of these forums "what was six sigma?" Since that time, I've read a number of articles, plus Mikel Harry's book on sig sigma (his claim is he was one of the original players at motorola when six sigma was initiated). Personally, I think it is a great concept. What Harry has done, in a nutshell (IMHO), is give companies what they've always wanted, a method to quantify (and justify) cost savings by implementing a quality initiative. Harry's most basic premise is that costs increase and functionality dcreases the farther one deviates from the nominal (huh, sounds like Taguchi loss function to me). But it makes sense, directly affects the bottom line, and BOTH supplier AND customer benefit. The cost of quality has always been a sore subject with me, and while Harry doesn't state it explicitly (again IMHO), he does imply that the cost of quality is the cost of doing business. I have reservations about the 1.5 shift, and Harry really doesn't bring anything "new" to table as far as individual quality tools, but he's done a great job of packaging and selling the concept. Companies like GE, Allied Signal, and Polaroid, to name a few, have bought into six sigma, with some rather impressive results. IP: Logged |
|
Kevin Mader Forum Wizard Posts: 575 |
posted 19 July 2000 10:32 PM
John, I just read through this thread and in your July 1 you stated: Deming damns this 6 sigma sort of stuff. By the way, what did he actually say? I never heard but I'm sure he must have had some strong comment on the business. The first thing that popped into my mind was that he said precise optimization was not desirable. It would be too costly. I really enjoyed that post of yours, by the way. Mike 1245 posted: Several months back (maybe longer) I asked in one of these forums "what was six sigma?" Since that time, I've read a number of articles, plus Mikel Harry's book on sig sigma (his claim is he was one of the original players at motorola when six sigma was initiated). I felt at the time of your post you may have been fishing for some answers based on the large debate between Quality Progress (ASQ) and its members. Maybe the timing was coincidental. Dr. Mikel Harry may be a pretty smart guy, but his arrogance and lack of respect for the men who founded statistical thinking and management philosophy shrouds his accomplishments (if truly any) in a dark, stinky cloud (in fact, his whole program is based on their works). His subsequent articles have the same arrogant tone, IMHO. Needless to say, I am not a fan of the cowboy. Harry's plan lacks intrinsic motivation, focuses on numerical goal that are abitrarily set, and promotes short-term thinking. I believe that an organization is not in the business to make money as a primary objective (unless it is a fund raising organization). This belief was promoted and shared by Deming, Juran, and Sarasohn to name a few. Organizations should be in business to create jobs, and more jobs in fulfilling customers' needs, supporting with respect to the communittee, the enviroment, and its employees. Stockholders come somewhere after that. Harry's plan is about fulfiling the short-term investor at any cost. The reason that this is so popular is because of Wall Street's need to make instant riches. There is no instant pudding, WED. The promotion of Harry's plans by the CEOs of these organization only lend to the misunderstanding of the program's apparent success, IMHO. It isn't that easy, and if it were, we'd all be rich. JC's post from July 1 was right on the money, even from the perspective of a person acknowledging limited statistical knowledge (John, you do a pretty good job at explaining for such a novice). Six Sigma is more than statistics, it is a management methodology (philosophy). There is nothing wrong with Six Sigma thinking, from the perspective that it is used to promote reduction of variation and improve quality/productivity. However, Harry's slant on it and the Black Belt program at $35,000 a crack may show significant short term gains, but lacks long term thinking. However, it is making him a rich man! The organizations you listed have been successful for many years, even before Dr. Harry came along. Many factors can be attributed to their current success, and Harry Dent believes the success of many of the Fortune 500 organizations is based on spending waves (cyclical change) and structural change. I believe he is right on this. I would recommend Dent's books Our Power to Predict and The Great Boom Ahead. Regards, Kevin Back to the group.... IP: Logged |
|
Jim Winings Forum Contributor Posts: 13 |
posted 20 July 2000 07:06 AM
Ok, I'm confused ... Kevin Mader said ... Then where does it get the moneys to expand technology? These thoughts may have worked in the 40's, 50's 60's and even 70's, but, as technology grows in BIG leaps and bounds, it takes a lot of dollars, pounds, etc. to hire talent to stay on the cutting edge. Not to mention the cost of test equipment, etc. If large companies can't save money, which making money is an alternitive to saving money, then where labor is cheaper will always win out. Becaise the money to purchase test equipment, etc. must come from somewhere. This in turn leads to lower wages and a poorer quality of life, and even a potential for issues like slave labor. This is NOT the same world that Deming was in 40 to 60 years ago. Deming was correct, without a doubt in the 40's when he took SPC to Japan. But business is much more complex today. I'm just not sure that Kevin's statement from above is applicable in today's 'New Economy' with world wide markets. ------------------ IP: Logged |
|
Mike1245 Lurker (<10 Posts) Posts: 7 |
posted 21 July 2000 08:51 AM
Kevin: Just one simple question - if a company's primary goal isn't to make money, then what is it? Seems to me that that's the primary goal to every business I know about. If you think a company's goal is working towards some higher, esoteric purpose, I'd like to know which one's. Bottom line, publicly held companies need to show fiduciary responsibility to the stockholders and other investors, and privately held companies are in it for the buck.
IP: Logged |
|
Steven Truchon Forum Contributor Posts: 89 |
posted 28 July 2000 03:14 PM
I submit to any "non-profit entrepeneurs" that one ask any shareholder or investor what is the primary function of a business? I have never once in my brief stay here on planet earth heard anyone state an intention to start a business because they believed so highly in a means of managing quality or so that they might make "something" with zero defects. Excuse my 'tude but this was one that I had to have a moment with!  --- i know nothing and i can prove it ! IP: Logged |
|
johno unregistered |
posted 28 July 2000 04:44 PM
In occupied Japan in 1946 MacArthur needed radios for the Japanese but they were having problems building them. A former engineer of Crosley Corp., Homer Sarasohn, was dispatched to Japan, and with Charles Protzman of Western Electric they ended up putting together a course for Japanese managers. Some of the pupils in the first class were Akio Morita and Masaru Ibuka, who founded Sony, Hanzou Omi of Fujitsu, Masaharu Matsushita of Matsushita Electric, and others. A motto was on the first page of the course material, from Newport News Shipbuilding; "We shall build good ships here; at a profit if we can, at a loss if we must, but always good ships." IP: Logged |
|
Sam Forum Contributor Posts: 244 |
posted 28 July 2000 05:08 PM
If one starts a business for the "primary" purpose of making money; they will fail (or go to jail for counterfeiting). There are several things that have to be addressed to make money;Price,cost,efficiency,ROI,delivery,quality, wants/needs of the customer, customer satisfaction and ability to produce. All of these topics and more are covered under the guise of a good quality system. Making money is the by-product of making something that someone wants or needs. Of course we all know that we can make money by supplying inferior product at inflated rates, but for how long? IP: Logged |
|
AJPaton Forum Contributor Posts: 73 |
posted 31 July 2000 10:04 AM
quote: Jim, you're arguing for Kevin, not against him. The money that the company makes must be put back into the R&D department. So the objective was to continue supporting R&D, not just to make money. IMHO If the customers don't want the product, the company folds. If the company can't get fair compensation, they file for bankruptcy soon enough. Just my $0.02 worth. AJP [This message has been edited by AJPaton (edited 31 July 2000).] IP: Logged |
|
jwmgmt unregistered |
posted 31 July 2000 10:21 AM
Hello all, I am currently reading Harry's book on 6 Sigma as part of research I am doing for my own book. I was not familiar with 6 Sigma prior to reading his book and now after getting into the subject and reading your posts as well as other articles, I see a "house divided" with regards to the merits of this "new" methodology. I myself find two, IMHO, major flaws in the philosophy of 6 Sigma. 1). The concept of Defects Per Million Opportunities (DPMO) can be manipulated by an unscrupulous individual to "pad" the numbers and improve the Sigma rating without actually improving any processes. Example: This year I report to my shareholders that we are implementing 6 Sigma for our main product line. I report that the product has 4 parts or processes that 6 Sigma can apply to, and that currently we are operating at the 3.30-3.25 Sigma range or 37,500 DPMO. I wait 12 months and then report to my shareholders that we are now at 3.85 Sigma or 9,375 DPMO and my stock shares rise in value. How did I accomplish this without improving quality? Simple. I divided each step into 4 steps, thereby increasing the number of "opportunities" for defect from 4 to 16. Now according to Harry's formula provided in his book, I have, indeed increased my Sigma without increasing my quality. This may not be applicable in the real world and I would hope it isn't but the point I'm trying to make is that the numbers can be "fudged" to promote a virtual increase in quality without an actual increase. Maybe an improvement on the 6 Sigma push would be to lock controls in place to prevent this from happening. 2). The basic tenet of 6 Sigma is "If you build it, they will come." In other words, if your company implements 6 Sigma and subsequently builds a superior product, external factors, (such as the economy, competition, changing customer desires, etc.) shouldn't affect the sales of your company's product. A perfect example of the flaw of this argument would be Apple computers in the 80's. It was a superior product, in both design and function, but the wave of public interest went in the PC direction. Look at the New Coke fiasco. All their taste tests showed it to be a superior product, yet the public didn't go for it. This definitely affected the bottom lines of both these companies. With 6 Sigma, the only variables that can be controlled are the ones that can be measured and the only variables that can be measured are internal to the organization. I'm not sure of a way around this one. External factors MUST be considered when implementing quality initiatives. What will it matter if you build the best product nobody wants? These are some things that have gone through my mind as I research this material and read the opinions of others on the subject. I also have a problem with the 1.5 Sigma "shift". There seems to be something kind of fishy about a quality enhancement methodology requiring the removal of variances in production while not removing the variances from itself. IP: Logged |
|
Don Winton Forum Contributor Posts: 498 |
posted 31 July 2000 12:14 PM
Jim Winings said:
Mike1245 said:
Steven Truchon said:
I put these together so I could summarize my feelings on this. Everyone makes a valid point and I found the input interesting, but a small point should also be made (IMHO). The primary purpose of an organization should be to STAY in business, not to make money. A company that stays in business provides much more than a company that merely makes money. However, if making money in the primary means of staying in business, so be it! And if some so-called quality initiative is the means to make money, thus staying in business, so be it. But, make no mistake, if a company's primary goal is to make money, rest assured they will not be a company long.
That should be the goal of an organization. And you are right about Sarasohn. I have read that course over and over.
As you can tell from above, I agree. Jwmgmt: I agree with your post. Superior quality, thus a superior product, do not mean diddly if no one buys them. So I have issues with those quality 'programs' that do not examine all aspects of the business. Just the ramblings of an Old Warrior. IP: Logged |
|
ahuffman unregistered |
posted 10 August 2000 04:23 PM
I have been studying this site and it makes me want to throw up. When we will learn that no system is the cure all. It seems to be assumed by most of the participants here that as long as you make the best quality part imaginable, you are assured to be in business. I am in automotive and I can't tell you quality is nothing. Price is everything. Now the customers will throw you out for quality, but they will throw you out faster for price. It does not matter how good your quality is, if a competitor comes in with a better price, you are out. It may not be fair, but that is the way it is. So you have to find a happy medium between quality and cost. Mercedes found this out years ago when they could charge whatever they wanted because they "hand made" much of their cars. Then came Infiniti, and Lexus, quality cars made in production mode and guess what, now you can buy a Mercedes for 30k. Until we quit sitting around thinking up new concepts and work on things that truly affect the bottom line, we will continue to grab on to each new and improved concept that comes down the pike. IP: Logged |
|
skyc Lurker (<10 Posts) Posts: 6 |
posted 11 August 2000 09:19 PM
dear ahuffman, I'm not sure if it's etiquette - I apologise in advance if not - however, you might consider reading and replying to my post on deja-misc-industry-quality "is quality progressing" to further your argument. thanks Personally i have started to participate in these forums because it's obvious a lot of knowledgeable and experienced people are evident. There is a need for application of quality techniques and you'll see people talking about these in most posts ; much of what is discussed leads to cost reduction in the form of waste removal. So, I'm not convinced of your line of argument that is implying that to be taken seriously or contribute one needs to concentrate on the bottom-line. Are you saying quality techniques are not doing this? You cite an example of the price and quality of cars and indicate that the former is more important to the customer. I'd disagree, consumers don't see it this way, they chose on both equally and believe they're getting it but are often duped. In business, with suppliers and customers we are so fascinated by the bottom-line that it has become the only model because it's simple. But it comes with preconceived ideas - e.g. quality suffers because it's so temptingly linked to cost reduction and not to differentiation. Doing a good job goes in-hand with producing desireability ; and this is why quality professionals create something new or something extra special and don't focus narrowly on an accountant's balance sheet that doesn't consider the business potential, implications on reputation and what deming I believe called the unknowable costs. you think? [This message has been edited by skyc (edited 11 August 2000).] IP: Logged |
|
johno unregistered |
posted 13 August 2000 03:40 PM
On the quality vs cost debate it's perhaps more useful to consider quality as the 'great enabler', the foundation, as without it low cost is a moot point for most, and without it a good design is usually not realized. In a volume environment quality enables the pursuit of lower cost and the consistency also enables better designs to be utilized at lower cost. Note that volume can be either high product volume, or low product volume and high volume of parts and operations. Among engineers this seems to be the 'how many angels can dance on the head of a pin' discussion :^). IP: Logged |
|
jwmgmt unregistered |
posted 15 August 2000 10:52 AM
ahuffman, Your argument is justifiable if you change the word "price" to "value". I for one have read Harry's book on Six Sigma and found flaws in it and also after reading Philip Crosby's "Quality Without Tears", see a lot of information being copied by Harry to add support to his theory. You are correct in saying that quality alone is not what the customer cares about, but value is. I say again, you can build the best designed product in the world but if nobody wants it, what good is it? Something else that should be noted here is that Six Sigma reduces cost. If production remains constant or increases then with reduced costs, profit increases. Profit, not REVENUE. Again, this does not add value to the product, which is what the customer cares about but it does make the stockholders feel warm and fuzzy. IP: Logged |
|
Marc Smith Cheech Wizard Posts: 4119 |
posted 15 August 2000 06:12 PM
quote:I don't think many here are in any way saying any 1 system is 'the' system. I guess I read these and other posts as discussions of different systems one or more of which may be utilized by any company. Typically no company embraces 6 sigma alone, for example. Companies mix things up. One client I had was into 6 sigma, was completing a 5-S program, was implementing QS9000 and was undergoing an significant internal managem,ent restructuring. Different threads and topics here in the Cove forums typically are focused, however.
quote:I bet Ford will have less consideration of the price it pays for tires with respect to the 'quality' of the product. I hear this arguement often and I always say the same: GM wants a light switch and 2 companies make the switch. Company A sells for 25 cents and company B for 39 cents. Company A's switch has a MTBF of 1000 hours and company B's of 10000. Use predictions are 8000 hours. Considering MTBF to be part of the quality of the product, which one would you, as a GM buyer, purchase? If you want to argue that MTBF is a design consideration, I will point out that the design aspect is surely part of your overall quality system - not just manufacturing. I've seen a lot of companies go for that 25 cent switch and pay dearly later. Maybe the recent Firestone 'event' was a matter of specifying or buying cheaper materials. We may never *really* know. Whether one admits it or not, decision making in buying is never based only upon price. The quality of the product you are purchasing is considered in one way or another. IP: Logged |
|
Carlos_Elizondo Lurker (<10 Posts) Posts: 2 |
posted 22 August 2000 05:52 PM
Hi everyone!!! My name is Carlos Elizondo, Im a 22 years old student from Tampico, Tamaulipas, M?xico. Im almost to finish my Career, Chemical and Systems Engineering, My thesis theme is: "Development of a guide for the implementation of Six Sigma" Any suggestions about the methodology for gathering information?? Where can I find more info?? Thank you, IP: Logged |
|
Jim Winings Forum Contributor Posts: 13 |
posted 23 August 2000 07:11 AM
Sorry, I've been away getting the final version released, but I'm back!!! Seems to me this Six Sigma topic is a little skewed and should be in Six Sigma Want To Fight topic or something, but, let me see if I can get it re-centered? (Theres a quality joke in there somewhere, but ya have to DIG for it) (GRIN) AJPaton said ... quote: In order to do R&D a company must make money to pay for benefits to hire the talent to do the R&D. Perhaps we are all on the same page, it just depends, as we all know, on how one measure's it?
quote: Harry taught my 6 Sigma class at Motorola right after that book was 1st published
quote: So can any method. Lets face it, as with all statistics, it what and how you measure it. depending on this, you can make a bad part look good or a good part look bad. It's all in the measurement!
quote: Personally I think that the 'Coke' deal was all a genius marketing deal, and it worked. I think that in the long term, Coke got just what they thought they would get.
quote: According to my Juran Quality and Planning and Analysis book, the 'normal' shift to allow is 1 Sigma, I don't see how an extra .5 sigma, could be viewed as other than common sense. Of course I have had the Six Sigma courses and see the proverbial picture as Motorola wanted me to see it. Maybe that's it?
quote: Agreed, but a company can not stay in business without making money. From the business knowledge I have, a company needs to typically charge 75% more for a widgit to make money. Does that mean that at, lets say 25% profit margin to break even, they should sell their widgit for 25% of cost? In my Six Sigma classes, they stresses getting a Return on net assets and a return on the net investment. This was a very BIG part of the Six Sigma Methodology. johno said ...
quote: Yes, unfortunately, 'Quality' is still viewed in a lot of proverbial circles as the 'Necessary Evil'
quote: You are right. This is why Six Sigma requires Customer Suveys. To insure that this does not happen. And I have seen it happen when I was at Motorola. Nothing like spending, well millions at the time, to develop a product that nobody wants and ergo will not purchase.
quote: Correct, but I am starting to think that a lot of people think of Six Sigma as only the measurement part of the methodology. And the scope is much much larger than that.
quote: I live about 30 miles from the Firestone plant in Decatur IL. I've only been in the area for about 4 years now, but, let me tell you some things about it.
Well that should stir some excrement! IP: Logged |
|
Mike1245 Lurker (<10 Posts) Posts: 7 |
posted 24 August 2000 10:02 AM
Wow - just got through perusing this whole forum. Just a few comments - 1. Don Winton. Like your paper (on 6 sigma that I downloaded from the Cove (Marc - you put up some excellent information - so do you Don at your website). Did you write this specifically for the AF? 2. Jim Winnings. Try living in Indiana! And, the way I see it, you can measure something all day long, but there is a sh**load of truth in Taguchi's Loss Function, and that's why the Japanese beat the p*ss out of us everytime - they don't concentrate on the spread (6 sigma) - they shoot for the target (or nominal value). 3. Marc Smith. Bingo! There is NO ONE SYSTEM. If picking one's nose ensured quality, customer satisfaction, and profit, guess what everyone would be doing? The bottom line is - have an open mind. I didn't buy into 6 sigma at first, and now I would be the first to tell a company to go for it.
IP: Logged |
|
Carlos_Elizondo Lurker (<10 Posts) Posts: 2 |
posted 10 September 2000 03:44 AM
Hi everyone!!! I have a doubt... Thank you very much?? Thank you very much, Carlos Elizondo. IP: Logged |
|
Marc Smith Cheech Wizard Posts: 4119 |
posted 07 June 2001 06:31 PM
Folks: FYI - I'm not sure if Don is back with us. [This message has been edited by Marc Smith (edited 07 June 2001).] IP: Logged |
All times are Eastern Standard Time (USA) | next newest topic | next oldest topic |
  | Hop to: |
Your Input Into These Forums Is Appreciated! Thanks!
Main Site Search 
