iSixSigma

Lean and/or v. Six Sigma

Six Sigma – iSixSigma Forums Old Forums General Lean and/or v. Six Sigma

Viewing 32 posts - 1 through 32 (of 32 total)
  • Author
    Posts
  • #33122

    Dr. Scott
    Participant

    I am finally going to ask something!
    What is the deal with so many people attaching Lean Manufacturing and Six Sigma at the hips in resent time?
    These once were two competing initiatives in the market place. Each has a set of effective tools, neither of which is completely comprehensive. Please dont misunderstand, I do not disagree with such a merger in the least.
    However, why “Lean” and Six Sigma? Why not TQM and Six Sigma? Or ISO and Six Sigma? Or 5S and Six Sigma? Or TPM and Six Sigma? Why not etc. (and there are many of them out there)?
    Just looking for the thoughts of others as to why Lean and Six Sigma are now the buzzwords.
    Thanks

    0
    #89101

    Reigle Stewart
    Participant

    Dr. Scott:You raise a very pertinent question. It would seem that
    many are struggling with this concept. From the looks of
    things, just as many are quick to “sell” it, but few seem to
    have a comprehensive understanding about its merger,
    merits, or methodology. What strikes me is the
    coincidence of two things. First, the lowered economy
    generally motivates management to push-and-pull for
    more, but doing it faster-for-less. Many say that lean
    manufacturing is invariably less intellectually intensive.
    This would suggest a shorter less demanding curriculum.By overlaying the goal of “six sigma,” it would suggest that
    one can achieve breakthrough performance, but at a
    faster rate, presumably cheaper, over a greater audience.
    The theory sounds good. But in the final analysis it is
    likely that this branding of Six Sigma has been done more
    for consultants than for corporations. Remember that Six
    Sigma initially met with great resistance because of its
    rigor. Over the last several years many have praised it for
    its rigor. But now that the economy has moved south, the
    “in thing” is “easy and quick.” In addition to this logic,
    recall that many organizations adopted the principles of
    lean manufacturing and, by in large, were quite
    successful.At about the same point in time many of these
    organizations were flirting with Six Sigma (because of
    successes such as General Electric). Not wanting to drop
    their public commitment to lean manufacturing, but yet
    seeing the merits of Six Sigma, the merger only seemed
    logical. But then again, so did the merger of boats,
    airplanes, and automobiles. History is littered with many
    inbred and crossbred failures.

    0
    #89163

    DaveG
    Participant

    Mr. Stewart,
    Your pedantic answer has no weight.  The short, correct answer is that Lean and Six Sigma are complementary approaches that eliminate the (7 or more) deadly wastes, listed under the tool generally associated with the waste:
    Lean:  Inventory, Waiting, Transportation, Motion, Processing
    SS:  Defects, Variation, Unwanted Products
    The ASQ Body of Knowledge for SS requires a familiarity with Lean.

    0
    #89167

    Dr. Scott
    Participant

    But Dave, why Lean AND Six Sigma? Can’t the tools of Six Sigma address the seven deadly wastes as well as (or better than) the tools of Lean can? Seems to me, that Lean only targets certain wastes while Six Sigma can target anything.
    Your thoughts?

    0
    #89170

    jagdish
    Participant

    Lean and Six Sigma compliment each other very well. Here is what I would say about Lean Six Sigma.
    Lean is about waste reduction and Speed . The problems that Lean addresses are all done using the tools in Six Sigma.
    Value Stream mapping – Enhanced version of SIPOC
    SMED – Six Sigma tools can be used to make this happen. Looking at data and see how you can reduce the changeover time. Look at variation and reduce it
    Reduce Downtime – Use the tools like Pareto, Cause effect, ANOVA etc. which are all embedded in Six Sigma
    Reduce Inventory – How do you know where your issues are ? Where is the inventory stack up ? Which factory is doing better ?
    All these are analysis of data that uses one of more of the Six Sigma tool set.
    Root cause Analysis, Action workouts in Lean – We can use all the tools of Six Sigma methodology – ANOVA, Regression, Casuse effect, FMEA, Risk analysis , Pareto etc.
    To summarize Lean tells us clearly where are the focus areas and what to do. Six Sigma tool set helps us in HOW to do that.
    They integrate very well and they better be because you don’t want to run Lean and Six Sigma into separate initiative only for people to come back confused, disgusted and the CEO to say that nothing adds value to me. Integration is the way. Same with TPM, TQM. All these can be integrated and the Six Sigma is the way to work around data and come to data based decision making. May it be Lean, TPM, TQM , QS or ISO 900-2000.
    Six Sigma tool set and the methedology is the answer to HOW do I satisfy the requirements of ISO, QS.
    How do I analyse the data that comes out from my TPM, Lean initiative.
    How do I stengthen and internalise my TQM practices- Six Sigma is the answer.
    I would like to hear feedback from other experts.
    Regards
    Jagdish

    0
    #89181

    reinaldo
    Participant

    Reduce Inventory – How do you know where your issues are ? Where is the inventory stack up ? Which factory is doing better ?The only way TO REDUCE INVENTORY is using Just in time and Kaban. There is not another way to reduce inventory.FOR MORE on SICK SIGMA, visit:
    http://www.lean-service.com/6-news-nov-01.asp#2

    0
    #89183

    jagdish
    Participant

    I agree with you in philosophy.
    When it comes to increasing inventory turns the answer is always NOT JIT and KANBAN. That is a no brainer.
    We need to keep in mind that
    1. Not every factory can implement JIT .
    2. Not every one can have their Suppliers near them and supply in request.
    3. Not every factory can implement Milk Run concept
    If we are talking about mass scale production, good forecast accuracies, good customers, all these works.
    When you are operating in an environment where
    1. You are grappling with inaccurate customer forecat
    2. Suppliers sitting in far away places
    3. Low volume, legacy manufacturing systems
    you still have to reduce inventory without JIT and Kanban. What do you do ?
    Data analysis ! Try to improve forecasting process ! Work on ABC pronciples ! Improve Supplier lead times ! Improve productivity !
    How do you do all these ? SIX SIGMA WAY
    Regards
    Jagdish

    0
    #89185

    Reinaldo Ramirez
    Participant

    One thing only in yoir post:
    Productivity cannot be improved without improve quality (Productivity is a consequence of the qaulity improvement). This is supported by the Dr. Deming’s Chain Reaction (1986).
    All other concepts are rights.

    0
    #89217

    DaveG
    Participant

    Dr. Scott,
    I don’t understand your argument.  Please describe the tools you are referring to.  I do agree that Lean does not address all wastes.

    0
    #89232

    newcomer
    Participant

    I think you will find that even the likes of Motorola and other big org’s are now realising the importance of incorporating the two initiatives together, and yes they do go hand in hand.  The problem is when does a project become lean or six sigma.
    My understanding to how they fit together is to understand the nature of the problem first i.e. is it a process performance or process variation problem. 
    (And yes I agree you cannot ultimately improve productivity without an improvement in quality, but what if you have not got a quality issue??? This does happen)
    Then you have to understand how the tools will give you the biggest impact with minimal effort.  If you approached every project using the DMAIC method you could be there all year, when all you needed to do was a SMED.
    I recently had a project on Material usage, my superiors said its a lean project because the process was over-processing (one of the 7 wastes) but i said you can’t fix using 5S, SMED, TPM or any of the other lean tools.  I provided evidence showing it was a variation problem therefore a Six Sigma project.  Nothing has changed within the process as such i.e. the setup and running is done in a standard way, therefore some other factor(s) has taken place
    My point is whatever the issue / project choose the correct tool to do the job.  In a previous role at another company I developed  smed training material which included Six Sigma tools, simply because the nature of the SMED projects required high attention to quality when we were trying to speedup the process. 
    I have always started with the VSM, this tool is ideal to firstly establish what your problem is and whether it is a lean or six sigma project.
    Be careful to assume Six sigma will tell you How it can be fixed everytime, sometimes Common sense gives you the answer – and at the end of it all that is what Lean is all about.Sorry for the essay, hope this is of help.

    0
    #89252

    Reigle Stewart
    Participant

    It is agreed that both initiatives can do different things.
    But why mix the two at the risk of dilution. Competing
    aims is often healthy. History has shown that “cherry
    picking” through the “trees of corporate initiatives” is most
    often “a bad thing” that never seems to work out very well.
    I have worked on several corporate “staffs” that have tried
    to just this sort of thing. The results were disastrous.
    Mixing only causes confusion in the executive mind. This
    is counter to the aims of “selling.” The discussion can go
    on forever.

    0
    #89254

    newcomer
    Participant

    I agree to some extent but it depends how the organisation looks at it.  Some companies perceive Six Sigma as a tool within the Umbrella of Lean Enterprise.  I have seen both failings and rewards for combining the two initiatives.

    0
    #89256

    Reigle Stewart
    Participant

    What you say has great intellectual appeal and makes
    “common” sense. But please remember that 4 sigma
    companies are often lead by 4 sigma graduates of 4
    sigma business schools that have (for many years)
    promoted 4 sigma concepts. They produce 4 sigma
    products using 4 sigma processes fed by 4 sigma
    suppliers that ultimately produce 4 sigma profits. Are we
    to believe that 4 sigma businesses intuitively know “what
    it takes” to be 6 sigma?
    Are we to believe this in light of the fact that they have
    never planned, deployed, implemented, or validated 6
    sigma? In my experience, most 4 organizations (and their
    management) only think they know how to proceed with 6
    sigma. Once they have been “shown the light,” they are
    usually a little more humble. Recall that the “one-eyed
    man is king in the land of the blind.”
    Currently, there is no long-term data to support the
    merger of Six Sigma and Lean. There are lots of opinions
    but no data. To date, Six Sigma has proven its worth
    many times over. So has Lean Manufacturing. Again, the
    cross between an automobile and a boat “sounds good in
    theory,” but history has shown otherwise.

    0
    #89259

    newcomer
    Participant

    I hold my hands up.  I have no constructive answer to this debate because there is no data to prove or disprove, i can only speak of my experiences, which i believe this discussion forum is about. 
    Thats it for me, i look forward to the next discussion topic.

    0
    #89319

    GirishTiwari
    Participant

    When you are operating in an environment where
    1. You are grappling with inaccurate customer forecat
    2. Suppliers sitting in far away places
    3. Low volume, legacy manufacturing systems
    you still have to reduce inventory without JIT and Kanban. What do you do ?
    Data analysis ! Try to improve forecasting process ! Work on ABC pronciples ! Improve Supplier lead times ! Improve productivity !
     
    We are doing Six Sigma as well as Lean.
    We are in the environment shown by you as produced above.
     
    Can you tell me how to use the past data and find out the Kanban quantity?
    We decide the kanban based on the lead time the vendor will take. However the consumption varies and also the lead time in actual case, so we are having lot of hold ups.
     
    Please reply fast.
     
    Regards
    Girish Tiwari

    0
    #89766

    Mike Carnell
    Participant

    Newcomer,
    I haven’t had time to spend in the discussion group lately but I had a friend send me your comments about Lean and Six Sigma.
    Your Management Team screwed up when they treated the two programs as separate initiatives. Typically they are set up with separate goals and end up in a head to head competition on who gets credit for what dollars. The net effect is that you will end up double counting a lot of dollars. From the business side nobody cares who fixes what and what little box every project fits into. Both are waste reduction programs and you can work cycle time projects using Six Sigma tools – it is done all the time particularly in transaction type environments.
    If you have your BB trained in Lean concepts and you don’t put some kind of artificial constraint (such as classification of projects) on them they will do nicely choosing which tools are appropriate and just fixing things. I think it was Theodore Roosevelt that said something about that – tell them what you want done and let them do it – something pretty basic like that. I think that comes under that other obscure concept called empowerment. Maybe we just shut off that concept and tell them – you only get a hammer – fix it all with the hammer. That sounds like an enlightened approach. As Billybob might say “Who’d of thunk?”
    As faras the history of success? Start with “Lean Thinking” by Womack. The guys that came up with TPS used two techniques – kaizen and kaikaku. We have been kaizened to death so even 5 year old children recognize it as incremental improvement. You don’t hear much about kaikaku which is what they did when they needed radical change. What does that sound like? The names were changed to protect the inocent. Only the facts mame.
    Motorola issued a card we all carried (were required to have it and the CEO would ask to see if you were carrying it) and it listed all the initiatives we were using to accomplish the vision of Total Customer Satisfaction (yes the vision was not Six Sigma – it was one of the things we were using to accomplish the Vision). It had about 5 things listed on it – one was Six Sigma and another was Cycle Time Reduction. Basically a fore runner to the term Lean.
    We used it at Motorola’s Government Electronics Group to get the FMU-139 Bomb Fuse program (it is the program Mario Perez-Wilson used for the background for his book “Six Sigma” and we also had another obscure name Adi Bhote – guess whose son) up to speed. That was the one that won, not just one but two CEO Quality Awards and the Navy Salty Dog award (a quality award). Larry Bartlet was brought in from Motorola, Seguin, Texas to help us with high volume production since there were no volume programs at GEG in those days. We used a guy named Bill Fechter to help us with the Lean side and you can still find Bill around the SS world. He is always worth talking to since back in those early days he shared a desk immediately beside another SS noteable and may have some interesting opinions about members of the SS world these days.
    Motorola, Seguin had a very successful Cycle Time program with their EEC IV program (Ford Electronic Engine Controller). Tony Franckowiak (who is now with Columia 300 – the bowling ball manufacturer – and is a frequent speaker for AME) had run a good program – good enough that in the original release of Schoenburger’s book “Japanese Manufacturing Techniques” – Seguin made one of the Honor Roll lists in the book. That is also the same facility that is highlighted in “The Hunters and Hunted” (the section with Lee Craft) where when we introduced the Universal Production Line we did it all with Lean concepts. I digress. The point was Lean type concepts were used with Six Sigma techniques at Motorola and used successfully – a long time ago.
    When I went to Columbia with Tony Franckowiak in 92 we took both Lean and SS concepts with us. They had just had a reduction in force by 25% just before we got there. In 1 year we increased production by 33% without additional headcount by using focused cells and process improvement. That was the move that helped several of us understand that the Lean and SS tools are not industry dependent – we went from electronics to bowling balls and the concepts never changed.
    Allied Automotive had a group called MPE working beside us in 1995-1996 during the Six Sigma deployment. I will let you figure out who owns MPE and what he just bought. Jim Lambert was the VP at automotive who was in charge of the SS deployment. He understood Lean extremely well and was a true advocate of SS. He was also smart enough not to strap that “competeing metrics monkey” on our backs. My co-author Barbara Wheat came out of that program. Nobody deploys Lean as well as she does. She successfully deployed  a mix at places like Ingersol Rand. She is currently running the SS and Lean programs at Tenneco and doing well as usual.
    GE deployed Demand Flow Technology before we did SS. As a matter of fact they also had the Change Acceleration Program (CAP) and WorkOut. All three of those programs made the work at GE go much quicker. I had Power Gen and they had done a lot of work with QFD which was also a huge help also.
    Integration is the only thing that makes any sense. There is no single answer for any business. It takes Six Sigma level management to understand that and understand how to do more than one thing at a time successfully – lets see there is my model of a SS leader thay can’t chew gum and walk because they can only focus on one thing at a time. That doesn’t fit my definition of a SS Leader. Remember Deming said “there is no such thing as instant pudding” when a guy asked him for the formula for quality improvement. Now all of a sudden we have it? As they like to say here in the Hill Country (Texas) “That dog won’t hunt.”
    Stop backing off so fast. Just because your a newcomer doesn’t make you wrong. It is the newcomer that gets out of the infamous “box” faster because they don’t know where the walls are yet. Six Sigma requires you ask new questions (remember what asking old ones and expecting different answers will define you as) – it seems you can ask those questions as long as it doesn’t run up hill of the current opinion? That dog won’t hunt either. Keep pushing the envelope (I am assuming you don’t work for the Postal Service).
    Good luck.
     

    0
    #89778

    Murray
    Participant

    Mr.Carnell,
    You and Mr. Stewart give completely different answers. He says integration will dilute, you say integration is the only way. From the story he tells, he has better experience. Which advice is right?

    0
    #89786

    DaveG
    Participant

    Allen,
    I suspect Carnell is a practicioner and Stewart, whom I acknowledge based on posts to be very accomplished in SS, is a theoretician.  Mr. Stewart is a sesquipedelian (a person who uses words like “sesquipedelian”) and seems more concerned about the nuances of deploying SS than the endgame:  CONTROLLING PROCESSES.  See my post which IMHO reduces this argument.
    https://www.isixsigma.com/forum/showmessage.asp?messageID=31829
    I support neither mixing nor differentiation, but applying the correct tools.  I have observed that many SS people place a disproportionate emphasis on statistics.  Statistics do not control processes;  nature controls processes, and leadership is key to harnessing nature – by allowing people the freedom to achieve profound process knowledge instead of short-term gains (like the abominable practice of geometrically increasing output and defects as a cyclical production deadline approaches).

    0
    #89830

    Mike Carnell
    Participant

    Allen,
    A some point in your life you are going to have to make a decision based on what makes sense when you gun it through your experience prism. The stuff I referenced is verifyable and I gave you the references to do so. Your call.
    Just stick with the Deming thing. There is no such thing as instant pudding. It san be found in “Quality Productivity and the Competittive position. Now all of a sudden we have “instant pudding” (dMAIC). Your call again.
    I ran across an advertisement (I don’t remember the product) but it has the lady that wrote “A Beautiful Mind.” There is a black board behid her that is covered in equations. In the middle of the board in large letters is says THINK. I do the things when I am working that make sense regardless of what people believe.
    On the 139 program I ran some c charts that most didn’t understand from a classical approach ,but it worked. That is the bottom line.
    Good luck.

    0
    #89848

    Arthur
    Participant

    WOW, nice song and dance.
    Just to piss everyone off.  There is no difference to Lean, SS, TPM, TQM etc.  They are all tools to improve poor output / performance. They just address different aspects of the problem.
    While Lean mostly address the flow of manufacturing, or flow of processes. (waste reduction through the process flow), SS addresses the variation of the process (waste reduction through the variation within the flow).
    You can think of it as, Lean takes the low hanging fruit, and the hard to reach fruit.  SS takes the rotten fruit that you can’t see but you run into it from time to time anyways.
    (Just another dog trying to piss on someone elses tree).
    as

    0
    #89849

    Murray
    Participant

    Measuring or containing a foot and a half; as, a sesquipedalian pygmy; — sometimes humorously applied to long words.Measuring or containing a foot and a half; as, a sesquipedalian pygmy; — sometimes humorously applied to long words.
    DaveG,
    I like your conslusion, Mr. Stewart seems to be long on words and short on independent thought.
    Please Reigle tell us more of what Dr. Harry says and does.
    I assume the word is applied to how far Mr. Stewart can think for himself.
    I don’t always agree with you or Mr. Carnell, but I think you have nailed this one.

    0
    #89888

    somo
    Member

    Been reading the comments on Six Sigma vs. Lean Manufacturing – they both work because they fundamentally reduce the variation of the process to which they are applied (if used properly…). While this is a direct goal of Six Sigma, it is less clearly understood in Lean Manufacturing. (I would rather use the terminology Toyota Production System (TPS) because some of the approaches of ‘lean manufacturing’ are a weak alternative to the TPS – more about this later.)
    As a quick point, previously there was mention of the importance of kanbans to lean manufacturing – kanbans are a corrective action to buffer variability in the process — without variability there is no need for kanbans. A kanban itself is a minimally variable inventory control method – it has a predictable (minimal variation) location, identification, min/max level, and predictable response whenever these vary.   Interestingly, “Factory Physics” (Hopp & Spearman), provides the concept of CONWIP (constant WIP) as an alternative to kanbans – again stressing constancy or predictability. Similarly significant is Spear and Bowen’s comments from “Decoding the DNA of the Toyota Production System” (available as reprint from Harvard Business Review Publishing, #99509) page 104, basically, “Toyota does not consider any of the tools or practices – such as kanbans or andon cords, which so many outsiders have observed and copied – as fundamental to the Toyota Production System. Toyota uses them merely as temporary responses to specific problems that will serve until a better approach is found or conditions change. They’re referred to as “countermeasures”, rather than “solutions,” because that would imply a permanent resolution to a problem.”
    For any supply chain, there are three sources of variability: supplier variability, manufacturing variability and demand variability. TPS addresses variability from each of these sources with buffers and countermeasures, including, kanbans, standard work, leveling, TAKT, etc. In the end, these countermeasures result in a less variable and extremely competitive supply chain – exactly the same goal as Six Sigma.
    As a follow-up point, it is also proposed (by others) that the TPS is really two connected processes: a manufacturing process and an improvement process. Spear & Bowen provide an excellent overview of the improvement process, (pg. 98) creating, “. . . a community of scientists. Whenever Toyota defines a specification, it is establishing sets of hypotheses that can then be tested. In other words, it is following a scientific method. To make any changes, Toyota uses a rigorous problem-solving process that requires a detailed assessment of the current state of affairs and a plan for improvement that is, in effect, an experimental test of the proposed changes.”
    In short, the proposition is that TPS is a real time, ongoing experiment – because of the predictability of the System, variation is easily identified and reacted to. If the kanban is not replenished appropriately, it is easily visible to take corrective action. If an andon is initiated, correction is taken. TPS creates a process where significant variation (e.g., similar to variation of ‘special causes’) is easily identified and readily reacted to . . . the same concept as the control chart.  The TPS process is an optimization process — reduce variation (install countermeasures), shift the process (reduce inventory, increase production . . .) to continually increase productivity.
    Back to Dr. Scott’s original question of 8/21, Six Sigma, Lean Manufacturing and Toyota Production System have more similarities than differences – it may be totally inappropriate to separate them from each other in terms of their goals (ignoring the popular emphasis of ‘waste reduction’ in lean mfg.). As a set of tools, they may also eventually be fads, just like TQM, but the concept on which they are based is not going to go away. . .ever . . it’s all about reducing variation.

    0
    #89892

    Reigle Stewart
    Participant

    Somo:I must full commend your thinking. It is well
    grounded and has many points of merit.Reigle Stewart

    0
    #89906

    Mikel
    Member

    Wow! Points of merit from Reigle. You must be proud.
    Reigle – there is more useful information in this one post than in all of your Dr. Harry worship. You get points of demerit.
    The only thing I disagree with somo on is the saying that TPS and SS are similar. They are different in one very significant way. TPS is the way everybody is trained and expected to behave. You don’t get to play if you are not in the system. SS is elitetist – some are trained and receive special designation. They are outside the system. Most management are not well trained or disciplined. TPS is a system. SS is a fad.

    0
    #89923

    Mike Carnell
    Participant

    Somo,
    Nice response.
    I agree with you on your comments. I think we may be in violent agreement on the use of Kanbans. There is always a bottleneck or constraint operation. As a matter of fact they can be done in a Pareto just like many other things. We frequently use the bottleneck as a SS project and frequently the top 3 bottlenecks. Basically it is the idea of continuous improvement – if you are capacity constrained every improvement is money if you are not capacity constrained it will increase material velocity which in turn will increase inventory turns.
    As of this time I have never had an operation that did not require some type of buffering. As well balanced as we had the FMU-139 line we still had some operations (such as test cycles and encapsulation curing) where were were constrained by the physical properties of the material we used or the amount of work (testing) that needed to be done or was contractually required.
    As far as using Kanbans to buffer for process variability. One of the issues you see happening with Lean implementations is people start with Kanbans before they have addressed any process variability. The Kanbans frequently do not work as they believe they should – not because the Kanbans don’t work but because of the inputs and outputs of the Kanban (it is difficult to have tape on the floor, table, etc. not work). Unfortunately there is the typical conclusion that Lean doesn’t apply to them because they are different.
    Thanks.

    0
    #89928

    Trev
    Member

    Mike,
    Good to see you return to the forum. We missed you here!
    Trev

    0
    #89930

    Mike Carnell
    Participant

    Trev,
    Thank you. Just a little down time.
    Regards,
    Mike

    0
    #89942

    somo
    Member

    It was a rather quick post and possibly could provide another explanation….
     
    The variability in every process is ‘buffered’ by some combination of: time, capacity/resources, or inventory.  A good example is the fast food restaurant at 11:15 a.m. and 12:15 p.m.  The restaurant would like to maintain some defined level of service (on-time delivery & lead times, ignoring food quality at this time…) to their customers – and they would like to provide this service under conditions of  variable demand.  To meet the demand variability of lunch time, they have extra cooks ($) and possibly some finished or WIP inventory ($) in place.  If you arrive at 11:15, you probably will experience good on-time delivery & short lead times. 
     
    At 12:15, with higher demand, the extra inventory and capacity (possibly fully utilized at this time) may still enable the restaurant to provide short lead times and on-time deliveries.  At 12:20, a bus load of students arrives, increasing the demand above normal variation – inventory and capacity may not be able to meet the demand, so the students need to wait for their food — time. 
     
    This is characterized by every process; an ambulance service can’t keep it’s patients waiting, so they maintain extra capacity (drivers, ambulances) to meet variable demand (emergencies); a grocery store maintains extra inventory, again to meet short lead times and reliability under variable demand; an organ donor provider service has extremely variable supply, so the patients have to wait; on and on . . .
     
    This buffer combination holds true in the mfg. plant.  To meet variable demand, factories manage inventories, lead times, miss/hold promise dates, & add/shed capacity.  Add variability in the supply chain (suppliers – missed raw material deliveries, uneven quality, variable lead times . . . .and in the factory – product mix, absenteeism, variable work rates, variable inventory levels, variable cycle times, unplanned downtime, etc.) along with the demand variability causes a need for even more buffering. . . . more time, more capacity and more inventory . . . to maintain some acceptable level of service.  The reduced variability in TPS reduces the level of buffers needed — basically the competitive advantage of TPS – faster/shorter times, reduced inventories, and fewer resources.
     
    The existence of a bottleneck does not change the need for a buffer . . . .my comment was that kanban is a method of providing minimally variable inventory to the operation.  The kanban level must be sufficient to cover bottleneck production rate (and the requisite variation associated with that rate) in relation to replenishment rate of the kanban (considering again variation associated with the supply). If the kanban level is set too low, then the operation will starve/shutdown . . .  a unpredictable or variable event.  If the upper kanban limit is exceeded, again process variability is introduced.  If it were possible to eliminate all of this variability (impossible to imagine), then bottleneck demand would be 100% predictable, along with 100% supply predictability, and my comments were that some kanbans (min/max levels, etc.) are not needed – but some other inventory method may be adequate (I don’t know what that would be……).  So I do agree (no violent disagreement .. . .) with your comments concerning the use of kanbans – and the need for having sufficient inventory, or need to reduce process variability to make it successful. 
     Your constraint comments are along the lines of Goldratt’s TOC and, again I don’t think that TOC is fundamentally out-of step with the goal of reducing process variability.  Goldratt provides a great story of process dependencies and variability (using dice and matchsticks) in ‘The Goal’.  In that example, there needs to be a minimal level of inventory between ‘operations’ to ensure that no process (including a bottleneck) runs out of work.  This experiment could be run again using the same process average but lower process variability ((normal die average = 3.5, variability of 1 to 6) vs. (‘new’ die average = 3.5, variability of 2 to 5 )) with the result that less inventory is needed to maintain the same level of process output.  Possibly ‘constraint’ theory may indicate that variability does not have equal effect throughout the process – I haven’t really thought about this very much…… ((Some quick thoughts . . . .most of constraint theory deals with average production levels, and relationships among the averages. . . . the lowest average indicating a bottleneck . . . if the average levels between two operations are ‘close’ , then extreme variation in the production levels could cause a moving bottleneck……???))

    0
    #89955

    Mike Carnell
    Participant

    Somo,
    We do seem to be in agreement about buffers.
    Your comment about TOC using averages seems to be correct. I’m not trying fire up the TOC people but when you watch people running numbers on Lean or TOC they seem to focus on the average with very little attention to the variability. I can see where that could cause the perception of a moving bottleneck. We have had a lot of luck mixing the basic statistics with Lean/TOC and getting a predictable result. In most situations when we get started the bottleneck is so far out of whack with the rest of the operation something would have to float a long way to get a different result. When we drive the Pareto of bottlenecks it seems to avoid any further overlap.
    I appreciate your thoughts. Thanks.
    Good luck.

    0
    #90016

    somo
    Member

    Thanks.
     
    Sorry about the font……guess cut and paste doesn’t work for the post.

    0
    #90029

    Mike Carnell
    Participant

    Somo,
    I was wondering what the font was all about.
    Regards,
    Mike

    0
    #90030

    Ron
    Member

    You highlighted the root cause in your opening statement.  The two toolkits are linked at the hip.  I guess if we had smarter consultants they might have combined them from the beginning.
    It amazes me how many people think that if they are quality engineers they know Six Sgima and if they are Manufacturing engineers or Industrial enginers they know Lean.
    Neither could be farther from the truth.
    Practically you cannot lean out a process if the variation within the process is to excessive.  Six Sgima tools can aide in the removal of variaition but do not have any tools for improving process flow.
    The marriage of these two toolkits is where all continuous improvement professionals should be focusing their efforts.
     

    0
Viewing 32 posts - 1 through 32 (of 32 total)

The forum ‘General’ is closed to new topics and replies.