|
|
|
05/30/2012 09:22 AM
|
|
|
Hi,
Please let me know difference between Lean and six sigma and their relevance to TQM.
Thanks in advance.
rgds
Parag Kulkarni
|
|
|
|
|
05/30/2012 03:10 PM
|
|
|
Parag, here is one answer. According to Michael George in his book, Lean Six Sigma:
"Six Sigma alone cannot dramatically improve process speed. That's why you need Lean."
Lean alone cannot bring a process under statistical control. That's why you need Six Sigma. Achieving a fast improvement rate in process speed, cost, and quality requires the combination of Six Sigma and Lean".
Why do we need top control variation? Because everything we do can be considered to be a process or part of a process. Every process can be characterized by its average performance and its variation. Large variation swings around the average can result in defects.
Six Sigma seeks to improve the quality of process outputs by identifying and removing the causes of defects and by bringing variability in the processes under statistical control and keeping it there. Lean alone doesn't always keep it there.
In case you are not clear on the pervasiveness of variation, try this experiment. Write your name six times while trying to make each one exactly the same as the first one. You have written you name many thousands of times in your lifetime. You are the expert in writing your name. And yet, when you look at the six names you will see that no two are exactly alike due to the inherent variation in the name writing process. Six Sigma attempts to measure and control variation, Lean does not.
Sam Tomas
|
|
|
|
|
05/31/2012 04:57 PM
|
|
|
Thanks Sam
rgds
parag
|
|
|
|
|
05/31/2012 04:57 PM
|
|
|
Sorry Sam, I disagree that lean does not attempt to control variation.
Lean is business. You want to meet customer demands, with the least amount of cost, highest amount of quality and still make a profit.
Six Sigma, statistics, lean tools, are methods to achieve those goals. They all can fall under a proper business model. You will know you are doing them right, when you see the bottom line results.
|
|
|
|
|
05/31/2012 04:59 PM
|
|
|
Parag
Sam has a good answer to your question.
Here is another response based on personal experience gained at Toyota's start up plant in Georgetown KY and then managing the original Six Sigma program at AlliedSignal/Honeywell Automotive.
TQM is more of a concept or philosophy than a defined process. W. Edwards Deming, Joeseph Juran, and A.V. Feigenbaum did the original development and the U.S. Navy did a lot of work on TQM in the 1980s. We used TQM at AlliedSignal under CEO Larry Bossidy to begin to get everyone on the same page. In my opinon the biggest impact was that everyone worldwide had the same approach to meetings: Agenda, time keeper, scribe, facitator, minutes etc. Meetings were dramatically shortened and became effective. This helped set culture.
Next came Six Sigma with highly defined implementation methodology. It attacked operational problems mostly in quality that saved many millions of dollars by finding root causes and implementing permanent countermeasures. Success of the program was due to CEO Larry Bossidy's leadership. He attended all of the graduation events in N.A. and Europe and received quarterly progress updates on all manufacturing locations. Without that leadership it would have become another expensive flavor of the month effort.
After two full years of Six Sigma we integrated TPS (Toyota Lean) into the process and rolled out the approach to AlliedSignal Automotive in 1996. The Corporate rollout began in 1998. The integration expanded the scope of the improvement process and generated more savings through variation and waste reduction. This worked well in Automotive since we had many managers trained in Lean and then had them go through Six Sigma traning.
Six Sigma relies heavily upon the Seven Tools of Quality which includes statistical tools and analysis.
In the startup days at Toyota Georgetown we spent the first two years with one objective: Achieve the highest quality possible. Then we worked on increasing output and in year five began productivity improvements. Toyota knew that the highest quality systems drive the lowest cost and yields high customer satisfaction. For the first six years we did not rely upon SPC but spend time on true variation reduction and step by step quality confirmation.
Ron Turkett
|
|
|
|
|
06/05/2012 09:01 AM
|
|
|
John, here is another explanation of the differences between Lean and Six Sigma.
http://www.sixsigmaonline.org/...logies-explained.html
Sam Tomas
|
|
|
|
|
06/06/2012 11:40 AM
|
|
|
Sam
I know the difference.
My opinion is based on my definition www.leaneconomics.org, and through my own experiences.
I can not agree that lean does not search to control variation. What is poke-yoke. 100% error proofing. Six Sigma is only 99.97%. Lean offers much more up front.
Lean is business - Six sigma is a business improvement tool.
When a problem (variation) is not solved simply and easily thru common sense and immediate action, then you can turn to a six sigma project to try and find the root cause.
Six Sigma tries to separate itself from lean, while to me it falls under the lean umbrella.
The link was to a Six Sigma website. Of course they have their opinion and motivation. I for one have never bought into it. Cost me my job once as well. Oh well.
John
|
|
|
|
|
06/06/2012 11:41 AM
|
|
|
Parag,
Please do not go away with the idea that Lean is concerned with waste but not variation. In fact lean thinking includes both under the mantra of reducing Mura, Muri, Muda which translates into reducing Unevenness or Variation (Mura), Overburden or Unreasonableness (Muri) and Waste (Muda).
In fact Jim Womack once wrote a blog saying that with hindsight he believes that these concepts should be implemented in the order of Mura, Muri and then Muda.
In a lot of lean implementations the reduction in Muda is implemented while the reduction of Mura and Muri is largely overlooked.
In answer to your question the key differences I see between Lean and Six Sigma are scope and application.
Scope:
Lean is a broad philosophy using a multitude of tools to provide customers with what they want, when they want it while maximising quality and minimising cost. Therefore TQM is an element/subset of Lean.
Six Sigma is the application of statistical tools to minimise variation in the output of a process. Therefore Six Sigma is an element/subset of TQM.
Application
Lean needs to be applied by each and every member of an enterprise and holds them accountable for the standardisation and continuous improvement of the work that they do.
Six Sigma is the applied by specialists - Black/Green Belts - and can only be used where sufficient data can be collated to make the statistical analysis meaningful.
|
|
|
|
|
06/06/2012 11:42 AM
|
|
|
Sam
The reference you make comes up short of a good comparison. I realize it is someone's opinion based on limited knowledge.
As stated:
"Both the Lean and the Six Sigma Methodologies look at the same aspects of profit maximization but differently, Six Sigma focuses on the customer and end product, while the Lean version focuses on waste and production methods".
The goals of Lean are highest quality, lowest cost and shortest lead time. This looks like customer focus to me. Lead time includes Concept - to - Customer. The highest quality system also drives the lowest total cost. This provides the customer with high value products at competitive prices while generating profits to secure the future.
In 1996 in the Automotive Sector at AlliedSignal we fintegrated Six Sigma with Lean implementation primarily as a robust problem solving portion of the Quality Pillar. DFSS was included as the link between Product Engineering and Production Engineering and Manufacturing. This helped resolve the problem of overly tight tolerances on product specs.
Ron Turkett
|
|
|
|
|
06/07/2012 11:34 AM
|
|
|
Ron, I have a feeling that trying to describe the differences between how Six Sigma and Lean each plays a part in reducing variation is beginning to go the way coming up with a definition of Lean has been going - essentially in all directions since many people have divergent opinions about the differences and "never the twain shall meet". In other words, the two things, Six Sigma and Lean, are viewed as being so different as to have no opportunity to unite.
But into the breech we again go to attempt to make sense of the two terms.
Total Quality Management (TQM) is a management approach that focuses on a company as a system with an emphasis on teams, processes, statistics, continuous improvement, and delivering products and services that meet or exceed customer expectations. Six Sigma however, is far more than the latest quality trend. It's a disciplined extension of TQM.
For many firms the costs of errors, defects, and waste in the way they operate are not only huge, but are often hidden. When you don't know what it is that you don't know, it's going to cost you. The Six Sigma approach to managing is all about helping to identify what you don't know as well as emphasizing what you should know, and taking action to correct the situations.
The source of defects is almost always linked to variation in some form, such as material specs. Six sigma expands the scope of variation to also include time, for example, variability in production time. The thesis of six sigma is that variation is not acceptable because high variation means customers will not receive what they want. Six sigma's culture is customer-centric. The knowledge of what customers value most is the start of process improvement.
There is also another problem that Six Sigma addresses that Lean does not. Is it possible to build a product in which each component is a six sigma component and yet have the product exhibit yields of only 90%, 80%, 70%, 60% etc? What we are talking about is called rolled-throughput yield or the probability that all product, process and material characteristics will simultaneously conform to their respective standards. Six Sigma addresses this while Lean does not.
As Lord Kelvin had indicated, "When you can measure what you are speaking about and express it in numbers, you know something about it; but when you cannot express it in numbers, your knowledge is of a meager and unsatisfactory kind." Six Sigma is a numbers approach, Lean is not to the same degree.
The 6 sigma concept indicates there are strong relationships between product "defects" and such factors as Reliability, Product yields, Cycle times, Inventories and Schedules.
The higher the sigma value, the more reliable the process being monitored and the higher the improvement in all areas.
Very simply, one of the goals of Six Sigma is to improve cost, quality, and schedule performance via process improvement and reduction of process variations. It's approach however is different from Lean's approach.
Sam Tomas
Edited: 06/07/2012 at 11:27 AM by Lean Moderator
|
|
|
|
|
06/11/2012 11:01 AM
|
|
|
Sam
As usual we have some areas of agreement and some areas of disagreement.
For example:
"There is also another problem that Six Sigma addresses that Lean does not. Is it possible to build a product in which each component is a six sigma component and yet have the product exhibit yields of only 90%, 80%, 70%, 60% etc? What we are talking about is called rolled-throughput yield or the probability that all product, process and material characteristics will simultaneously conform to their respective standards. Six Sigma addresses this while Lean does not." I disagree. If each component is at Six Sigma level you cannot have the yields you describe. Also, the step by step quality verification process used in Lean effectively recognizes the rolled-throughput-yield impact on total process yield. Results are frequently recorded on the shop floor by team members so management can be aware of the real time results.
Maybe I am a purist in definitions but I rely upon my experience in Lean and Six Sigma beginning with employment at Toyota with its great teachers and managing the first major Six Sigma implementation after Motorola at AlliedSignal Automotive. That was followed by lean and Six Sigma implementation at G.M. Delphi, teaching both Lean and Six Sigma at the University of Michigan and then consulting in both areas with major companies in the U.S., Europe and other countries. The Shingo Prize requirements are very substantial and represent a full implementation of Lean. That is a much better source than relying upon trade or association dictionaries or academic papers. I have worked with many other experienced ex Toyota employees and Six Sigma implementers - academic and industry, and we tend to share the same ideas and definitions.
Classroom theory by people who read books leaves a lot to be desired. Thirty years of hands on learning with its successes and failures with great teachers helps a lot. Full understanding sometimes is difficult if you have not been there.
You can call a Kentucky Thoroughbred a mule, donkey or cow depending upon your personal opinion but it is still a race horse. Each to his own preference.
Ron Turkett
|
|
|
|
|
06/11/2012 11:02 AM
|
|
|
Hi Parag
Here is my personal take on this debate.
The first basic difference between Lean and Six Sigma. Six Sigma is more focussed on quality improvement and bring processes into a state of stability. It has more tools that can help focus early quality improvement in an organization. Lean on the other hand does have some useful tools early in an effort, such as 5S and TPM, but you cannot create a very good VSM of an unstable operation. On the other hand later on once you reach the six sigma level only Lean pushes you to improve beyond that. Remember there are industries where 99.97% quality is considered poor (would you buy pills from a drug company with 99.97% quality?).
Lean focuses on waste elimination and reduction with its ultimate goal is to eliminate anything a customer is unwilling to pay for; defects, scrap, unneeded work, record keeping, etc.; that includes quality inspection because no customer actually wants to pay for it, if it can be built in to the product or service.
As to their focus Lean is more end consumer focussed as it is the final consumer that can only determine value, while six sigma focuses on the customer who is the next handler of the product or service. Lean in fact will push an organization to eliminate a customer and move a step closer to the consumer, you eliminate waste when you cut layers from a distribution and production chain.
Now TQM despite its name only pushes quality to the six sigma level, its was designed to meet the needs of manufacturers that produced products for whom defects are rarely threatening the life of the consumer. Remember that when you start down the road of relying on statistics to control systems it get easy to use them in other areas including deciding that paying out lawsuits is cheaper than issuing recalls, it has been done. Lean on the other hand would push for the recall regardless, because consumer do not value their lives being in jeopardy and in fact are willing to pay for the security.
Hope that helps.
Robert Drescher
ELSE Inc.
|
|
|
|
|
07/05/2012 01:32 PM
|
|
|
Ron, look at the attached chart. As you can see in the six sigma coluum, as the part count in a particular product increases, even though each part may be a six sigma part, the probability of building the product without requiring repairs goes down as the number of parts used in the product goes up..
In case the chart didn't come through, here are some of the numbers for a product having all six sigma parts.
# parts: Probability of building products without requiring repairs.
100 99.966%
500 99.830%
1,000 99.661%
3,000 98.985%
17,000 94.384%
38,000 87.880%
70,000 78.820%
150,000 60,000%
If you are building a Boeing 747 aircraft with all six sigma parts, for example, It.will clearly have problems.
How realistic is it to assume that ALL the parts in a product are six sigma quality? Here is what can happen for a product having only 100 parts at various sigma levels .
Part quality: Probality of the product needing NO repairs.
+/- 3 Sigma 0.10%
+/- 4 Sigma 53.69%
+/- 5 Sigma 97.70%
+/- 6 Sigma 99.966%
Then of course, the picture gets cloudy if some of the parts in a product are 4 sigma, some 5 sigma, and some 6 sigma.
That's through-put yield.
Sam Tomas
|
|
|
|
|
07/06/2012 10:43 AM
|
|
|
Sam
Having managed the six sigma program at AlliedSignal and taught the subject at the Univesity of MI I am quite aware of the statistics. I reviewed the RTY statistics on 44 plant sites monthly and worked to drive improvement.
My point was that too many companies put non conforming tags on products and then repair after assembly. Working on prevention saves a lot of cost and improves quality and lowers warranty. Few products are as complicated as a 747 and even a 747 does not need six sigma performance on everything.
Ron Turkett
|
|
|
|
|
07/06/2012 12:34 PM
|
|
|
Ron, I think I've strayed away from trying to answer the original question submitted by Parag Kulkarni, which was,
"Please let me know the difference between Lean and six sigma and their relevance to TQM."
I recognize that RTY is the probability of an entire process producing zero defects and that RTY is an important metric to use where the process has excessive rework. But that doesn't answer Parag's question. The answer to Parag's question can be determined by answering another question. In deciding to reduce process defects, what are the characteristics of the defects that helps one decide whether to use Lean tools and techniques, six sigma tools and techniques, or both?
I'll stick my neck out on this next point but while Lean can probably improve processes from three sigma to four sigma and probably in many cases from four sigma to five sigma, I question whether Lean can be used to improve a process to a 6 sigma level. I would think you would have to use a six sigma approach.
BTW, I agree with you that working on prevention is the better way to go whether you use a Lean approach, a Six Sigma approach, or both.
Sam Tomas
|
|
|
|
|
07/09/2012 10:58 AM
|
|
|
Sam
During the time I worked for Toyota Six Sigma was never used and the Camry produced in Georgetown KY had the highest quality and lowest warranty of any U.S. produced car. What I saw was the full use of Lean (TPS) to achieve the high quality goals. Some of the managers who also came from the traditional Big 3 companies made comments that the quality standards were unreasonably high. Careful observation demonstrated that Toyota believed that the highest quality system drove the lowest total cost and made happy customers.
Another important input was Product Engineering's focus on design for manufacturing. The product engineers I talked to during new programs (many through interpreters) believe they had two customers: the final product customer and the team member making the product. Beside the effect on quality that philosophy had a positive effect on reliability as well and that also affected total cost.
I believed that my extensive operations and quality assurance experience was solid at my former auto company but sometimes in Georgetown, KY I felt like I was at the bottom of the learning curve. Since the opportunity to work in the entire value stream was provided I spent lots of time trying to grasp the depth of the Toyota approach and had many discussions with Fujio Cho, then site President and now Toyota's chairman. Several times he resorted to his simplified definition of TPS: Applied Industrial Engineering and common sense. Too many, including most consultants, don't understand the last item.
Ron Turkett
|
|
|
|
|
07/09/2012 02:18 PM
|
|
|
Thank you Ron for your comment about applying common sense. It frustrates me to see people go out of their way to complicate things. At least I can influence those that I work with.
|
|
|
|
|
07/09/2012 02:18 PM
|
|
|
Thank you Ron. You have been presenting some very interesting observations on the application and non-application of Six Sigma, which I appreciate.
I especially like Fujio Cho's description of Toyota's approach to quality as being "Applied Indusrial Engineeing and common sense". where Indusrial Engineering can be described as follows (from Wikipedia):
"Industrial engineering is a branch of engineering dealing with the optimization of complex processes or systems. It is concerned with the development, improvement, implementation and evaluation of integrated systems of people, money, knowledge, information, equipment, energy, materials, analysis and synthesis, as well as the mathematical, physical and social sciences together with the principles and methods of engineering design to specify, predict, and evaluate the results to be obtained from such systems or processes. Its underlying concepts overlap considerably with certain business-oriented disciplines such as operations management, but the engineering side tends to emphasize extensive mathematical proficiency and usage of quantitative methods."
Sam Tomas
|
|
|
|
|
07/10/2012 09:28 AM
|
|
|
Ron, I think you missed the two most important parts of Fujio's quote and zero'd in on the least interesting.
Lean is APPLIED industrial engineering and COMMON SENSE.
Industrial engineering is a branch of engineering that you can get a degree in and be considered an expert. Having that degree might help you to understand lean but it will get you maybe 5% of the way there. It is not until after you APPLY the theories in the real world and never forget the COMMON SENSE that isn't all that common (in fact it's down right extinct anywhere besides the gemba), that you will be anywhere in understanding lean.
Industrial engineering would include all of the minutia of Six Sigma. In fact most degreed industrial engineers these days walk out of college with green belts and sometimes black belts. But much like my mechanical engineering degree, when someone goes out and APPLIES their industrial engineering degree they realize that reality is actually much simpler and yet much more difficult to deal with. The most annoying lesson every mechanical engineer gets in any design class toward the end of their learning is the lesson on Factor of Safety. After spending all of that time learning how to get a precise and accurate figure on the right size of bolt to use, they tell you to put a factor of safety of 2 on it anyway and then pick the bolt that fits. All of that precision gets thrown out the door because there is no such thing as a perfect bolt or a perfect machine. Industrial engineering (and I think six sigma for the most part) spends way too much time doing analysis that has very little practical use because there is no perfect process and there are no perfect people.
When you apply industrial engineering with a dash of common sense, you realize that all of the complexity and over analysis of statistics can usually (there is a place albeit a small one) be ignored for the expedience of an oversized bolt. What a lean approach would emphasize is not finding the perfect solution to all of the variation that is out but instead finding a solution to some of the variation out there, trying it quickly to determine if it works, and then adjust from there.
Another thing that will probably distract you from the rest of my argument but...: I completely disagree with your assessment of lean getting you to 4 or 5 sigma's but not getting you to 6. I actually think that the opposite is much more likely to be true. In general, the tools of six sigma are VERY effective (overly effective based upon my earlier argument) at analyzing common cause variation. When you get down below the 100 to 200 ppm defect range, my experience is that you are not generally dealing with common cause variation at all. If you have a process capable of not making a defect 10,000 times in a row and then suddenly, it makes one, you are rarely dealing with the tails of a normal curve. You are usually dealing with an abnormality in the process in which something changed for some reason. Lean forces you to develop systems that escalate abnormalities and solve problems in one to one flow as they happen. Six Sigma asks you to take large samples to understand the trends of abnormality and attempt to solve one problem (that is really many problems). The lean approach of attacking abnormality as it happens in real time can get you to an actual zero defect state.
One last point on Applied with Common Sense. I have been around a lot of VERY smart people who loved to be the expert and had all kinds of belts. I happen to work for a company that has an abundance of bonafied rocket scientists and we LOVE data. Six Sigma is a holy term here. There is no questioning it. We have more data about many of our processes than anyone could ever need (literally). With all of that data and all of that brain power dedicated to analyzing it, i have never come across a solution to a process variation problem that the operators of the process did not already know. Identifying the proper design of the product to meet functional requirements is one thing but when it comes to actually reducing variation in an existing process with clear outputs, the associate will always be able to tell you how to reduce variation before the black belt can. They know their process, the only question is wether you trust them enough to make the leap or if you need someone with a pretty belt to tell you.
|
|
|
|
|
07/10/2012 12:31 PM
|
|
|
I have not missed the depth of understanding or Mr. Cho's quote. Having extensive operations experience on the shop floor and spending alot of time with Mr. Cho on the shop floor reviewing operations and increasing my depth of understanding of the Toyota approach I continue to reflect and increase learning so it can be passed on to others.
I fully agree that the best source of information is from those who do the work. A close friend who was one of the best Professors of Engineering I have have ever met always said if you can't speak in the language of the operator you are blowing smoke.
Although my training was BSME and MSIOE with course work in a PhD competed the most valuable learning was always on the shop floor. Toyota has a company focus on the value adders from the Chairman on down and by all support operations including sales and marketing.
We always tried to improve beyond 10 defects per million which is still short of Six Sigma. And the targets were always set on features that impacted vehicle performance, customer satisfaction and lowest total cost and had lesser focus on areas that did not need that quality level.
In managing the Six Sigma implementation at AlliedSignal Automotive I found that advanced statistical problem solving was needed in only a few percent of the problems. The old 5 why approach using team member knowledge was always effecient. Many still try to use the 5 whos approach or kill the effort with too much data gathering and analysis without a dose of common sense or knowing who to ask first - those doing the work.
Ron
|
|
|
|
|
>
Community >
Forums
Send to a Friend
