iSixSigma

how to break through during dmAic?

Six Sigma – iSixSigma Forums Old Forums General how to break through during dmAic?

This topic contains 33 replies, has 12 voices, and was last updated by  Schultz 14 years, 8 months ago.

Viewing 34 posts - 1 through 34 (of 34 total)
  • Author
    Posts
  • #37685

    Schultz
    Participant

    hi,
    i ran a couple of projects in IT, using DMAIC. great process. somehow we always made it, but i observed a nasty pattern: during A we always got stuck. not that we had no ideas. there were many. but no good ones, none which take your breath away.
    i wonder how to do better. any ideas?

    0
    #111375

    Dayton
    Member

    Charlie,
     
    You got stuck in Analyze?   Was it an issue of not understanding which statistical treatment or tool application was best used in the specific circumstances?   Makes me wonder a bit about the depth of training for your BB and MBB.  If your application of Six Sigma has stressed the learning and placement of analytical tools in the toolbox and not how to select and appropriately apply tools you have missed the point.  Six Sigma’s analytical strength is in appropriately applied statistics – not just understanding basic statistics, any techno-monkey can do that. 
     
    You might have meant that you get stuck in the Improve phase because that’s where I have seen many Six Sigma projects bog down, even after effective Analysis.  Frequently the Improve phase is most difficult for entrenched teams to address because many times substantive improvements take out-of-the-box thinking and moving away from the “that’s the way we have always done it” comfort zone.   But again, your BB and/or MBB needs to be able to, once analytical data is in hand from the Analyze phase, effectively facilitate non-judgmental brainstorming to get as many ideas and concepts for improvement on the table/dry-erase board as possible before a hierarchical Pareto-based winnowing down of probable improvements has begun. 
     
    Sounds like you are missing broad based and open participation in the conceptualization of your potential improvements.   In responding to your note, I have assumed much and know very little of your problem.   If you want to be more explicit in problem description and provide examples there are many in the forum with a great deal of expertise and willingness to help.
     Vinny

    0
    #111377

    Darth
    Participant

    Vinny, it is not unusual to get stuck in Analyze if one falls into the classicial analysis paralysis syndrome whereby you beat the stuffing out of the data to discover the true root cause.  Not a matter of not knowing which tool to use but more of using them all whether they are appropriate or not.

    0
    #111378

    Dayton
    Member

    Darth,
    Agree fully.  Hence the “Six Sigma’s analytical strength is in appropriately applied statistics…”    Effective Analysis is an imperative, but the conceptualization challenge of Improve can be a derailer also.
    Vinny

    0
    #111379

    Schultz
    Participant

    oh vinny, you are absolutely right. who invented the tpyo? I – improve is where we fail so often.
    we tried a lot. brainstorming. we have crazy ideas, hardly ever useful. fun thing, of course, bur of little value. many times all those many ideas are just little cariations of the main idea, somebody raised. and all this fighting. my idea is best, yours is nothing. in all politeness the team members fight for their lovely idea.
    we tried synectics. think you are the problem. feel the solution. nice, but … we tried brainwriting, to settle all our creative forces a little, to have somem ore discipline. nuts.
    but there should be something out there, which uses all this creative spirit, and widens the scope, and gives focus. why cant we just harvest great solutions from great analysis? concepts, where you will just know they work. no fancy, crayz, highfly idea, which may or may not work.
    thanks, btw. good to know this is well known trouble.

    0
    #111383

    Marty
    Participant

    Charlie,
    More focus, broader scope, stronger solutions? Sounds like you are looking for TRIZ. Kindly check out https://www.isixsigma.com/me/triz/ please.
    That’s the good news for you: use focused TRIZ, instead of unfocused brainstorming. It’s free, it’s available on the Internet, or via consultants, if you like.
    The bad news for you is: it takes time. It takes time to learn and practice TRIZ, just as it took time to make six sigma work for you. 1 or 2 years will be required to master TRIZ on average.
    Why does it take so long? – When entering the Improvement stage in six sigma you have already clarified a thing which TRIZ calls “inventive situation”. In fact, when a team acts like you describe this is an almost certain indicator that you do face an “inventive situation”. The problem may be clear, the route to the exit, the strong concept, is far from being evident.
    TRIZ now helps you to reformulate your inventive situation by using a kind of “language of innovation”. Your specific inventive problem turns out to be a well known generic one, viewed from a TRIZ perspective. And the generic solutions are well known, too (TRIZ covers a really broad range of these). All you need to do is understanding the generic structure of your specific problem.
    Finally, you can translate each of these generic concepts into several specific solutions, which fit your situation 100 %. And like learning any foreign language, this takes time to do. You can say an inventive ‘hello’ easily, but it takes longer to ask for the way to your hotel in this foreign language. It takes even longer to express yourself in this “language of innovation”, TRIZ.
    Anyway, TRIZ should improve results from your Improvement phase considerably.
    I have an other good news for you: there is TRIZ software around. If you think about it, it’s even cheap. The one I like most is this one: the Innovation WorkBench (IWB).
    There is one website which talks about Idea Opportunities, created by the IWB: http://www.mprv.biz/iwb/details/iwb_efficiency.html is a good entry point. I think they even sell the IWB, but I have to check again. TRIZ itself uses graphical descriptions of the inventive situation extensively. In the IWB this approach has matured into the so called Problem Formulator.
    The graphs you find on that page gives you the relationship between complexity of your model and the number of strong solution concepts for your specific case. It’s breathtaking and it matches my own experience.
    I think even as a beginner you can come close here. This is because the IWB has a good help feature, giving you access to many examples or explanations of different TRIZ techniques. So, even a beginner can start right away, to come up with stronger concepts during Improve. Ok, I think you understand “I like it”.
    Hope this helps.

    0
    #111390

    Schultz
    Participant

    wow marty, didn’t know abiut triz befor. didn’t know about iwb. the level of novelty. yes I am after essential improvments, no craz stuff.
    but please tell me, how will this bring more focus into my teams? and why will they cime up with even more ideas than befor, and even better ones. and why will they stop fighting?

    0
    #111391

    AlexP
    Participant

    Hi,
    I’m having my thesis on lean six sigma, however i’m thinking of creating a general model (if possible) for the selection of appropriate tool in each phase of the DMAIC or DMADV cycle, using AHP (Analyitc Hierarchy Process). AHP has alread been used as a tool for the selection of six sigma(DMAIC) or DFSS. Do you think it worth trying it ?  Any thoughts and suggestions would be appreciated.
    Kind Regards,
    Alex

    0
    #111406

    Marc Thys
    Participant

    Charlie
    DMAIC is NOT about breakthrough, nor about the most brilliant ideas coming up. DMAIC is about basic improvement using logic and thorough analysis. It is not “sexy”, nor was it intended to be.
    In order to keep focus during Analyse, ask one question: “what is the data telling us?” and then follow up with “how can we explain this?”, then go to question one again (and so on, until you find something actionable). This way, the solutions are generated pretty much automatically, no “whiz-bang” solutions generation using TRIZ or whatever.
    DMAIC is not about finding the “best” solution – it is about finding a “better” way to do things and settling for that. That is challenge enough within the timeframe you’ve got (4-6 months).
    If it is “breakthrough” you are looking for, and the “best” solution, you should be doing DFSS / DMEDI, not DMAIC. But then again, you probably won’t do it within 6 months.
    My 2 cents.

    0
    #111437

    beverly daniels
    Participant

    actually it sounds like you aren’t analyzing at all.  just brainstorming ideas about the root cause or even a solution without knowing the root cause?
    free-wheeling brainstorming is not six sigma.  analyze is about using the statistical & non-statistical analysis tools to narrow the field of potential causes to the root cause…it is a data driven iterative processs.  the data drives the discussions – not personal opinion, which sounds like what you are running into….
    of course if you know the root cause and need innovative solutions, TRIZ is one approach.

    0
    #111445

    Dayton
    Member

    Marc,
     
    While I certainly respect your right to express it, I vehemently disagree with your statement: “DMAIC is not about finding the “best” solution – it is about finding a “better” way to do things and settling for that. That is challenge enough within the timeframe you’ve got (4-6 months).”  Six Sigma’s DMAIC is not meant to be merely an incremental bettering of the process, it is truly an analytical methodology that allows you to derive root cause and determine and implement the best-case solution to your problem.  If you are doing what you described you are settling for a fix because of perceived limited time or other resources versus finding and implementing your best fix alternative.   If you are not aiming for breakthrough solutions you are wasting your time and that of your organization.    If we let Six Sigma DMAIC expectations become just making the process a bit better each time we touch it, Six Sigma will become the expensive fad many hope it to become.   Certainly DFSS methods design and redesign using Six Sigma analytics but all order of magnitude breakthroughs are not relegated to a complete redesign.   I think that you have seriously missed the point.
     Vinny

    0
    #111447

    Mikel
    Member

    Vinny,
    A couple of things to think about –
    – a basic premise of Six Sigma projects since the beginning is to get at least a 50% improvement. This does not imply a “best” solution, but a much better understanding of the process than when the project began.
    – Larry Bossidy, when speaking of Six Sigma, says he is most impressed that you can be successful at a project and turn right around and do another project in the same area and be successful again. Again, this does not sound like a “best” solution.

    0
    #111453

    Marty
    Participant

    Vinny, Stan,
    That’s a good point to make: we shouldn’t be satisfied with mediocre understanding and we shouldn’t be satisfied with mediocre results. Of course, what is “best” isn’t always known. The definition of what is “best” can even change over time.
    I was amazed when finding an example a while ago, which appears promising to me. With a little modification I think one can achieve outstanding results while the understanding of the process and its limiting root causes is still imperfect. Which means in turn: once our understanding is partly correct.
    Please have a look at http://www.mprv.biz/iwb/practice/iwb_examples.html at the package design example. It was posted here last year.
    It starts with a brief problem statement, which Mr. Senoglu posted, in chapter one. Next, it gives a simple model of the measurement problem described, and of its underlying design problem. This is TRIZ analysis of the situation(s), performed from the IWB perspective.
    Chapter 3 gives generic directions for improvement of the measurement problem, together with many detailed elaborations – still in a generic form. Chapter 4 does the same for the design problem mentioned. Which most likely is the more important issue of both.
    The amount of detail is overwhelming, at least to me. All suggestions given are derived from the formulated problem, following innovation strategies. Which is just another way of saying “well proven strategies to turn difficult situations into break through concepts.”
    When you go through it, you will see the root cause problem addressed occasionally. And you’ll see many other routes to the exit, too. Chapter 5 gives a glimpse of the directed thinking process, which usually starts next.
    Unfortunately  the following steps are not shown, which are needed to transform those still generic concepts into very concrete solutions. And of course there can no longer be generic answers. But each generic concept (and there are many listed in the example) can be turned into 1 (or a few) very specific solutions. How exactly these look like depends on: the specific system and its specific available resources, which can be utilized.
    What I value most in this example is this: this IWB analysis and its directions towards possible solution concepts helps me to view the problem quickly from many perspectives. Even those I never dreamed of. Some considerations, of course, I would have made anyway. But many I wouldn’t have considered.
    As I said earlier: “I like the IWB”. And I think it is very useful within the Six Sigma system.

    0
    #111456

    Darth
    Participant

    Wow, reading all these long posts has given me IBS instead of IWB but some good points have been made.

    0
    #111460

    Dayton
    Member

    Stan et al, 
     
    Following the essence of Aristotelian logic, “We are what we repeatedly do. Excellence, then, is not an act, but a habit.”   If you set your sights on “improving” versus “solving” that’s what you get – improvements versus complete solutions.   Effective and successive project solutions in the same area don’t mean that you didn’t get the best solution with the variables and data available at the time.   As technology and process knowledge change analysis and solutions change and minute changes in starting variables and constants can cause dramatic end point analysis changes – that’s why complexity algorithms are not effective in predicting weather changes – much to Edward Lorenz’ consternation in the 1960’s.   I strongly object to the creation of a business analysis paradigm whose aim is to make incremental improvements – even though that effect is too frequently the outcome.   There is no reason that practitioners and proponents of a statistically based analytical process like Six Sigma which demands beginning by effectively defining the problem, measuring current state outcomes, rigorously analyzing data, developing best case improvements, implementing same, and controlling effect should be describing their process as one that makes incremental improvements – big high impact changes, order of magnitude changes need to be your stated goal.   Some of my projects have been just that, big high impact changes – new ways of doing things, some have not, most have not, but if your goal is to improve and you are successful that’s what you get – an improvement.    
     Vinny

    0
    #111463

    Darth
    Participant

    Vinny, all well and good in some hypothetical dream world but the reality speaks of a slightly different scenario.  First of all, most DMAIC projects focus on a limited set of customer defined CTQs.  Secondly, the Team is given a rather narrow time frame of 3-4 or maybe even generously 6 months to complete the project.  Resources for dramatic changes are not provided.  DMAIC is a very focused attempt on improving some aspect of a functioning process to make it better, 5% better? 50% better?  That’s specified in the Charter.  DMAIC generally focuses on functioning processes that are seeking to better meet some customer expectation, not to achieve the ultimate.  If the process is that broken or has reached its entitlement, then DMADV is better suited where you have the time, resources and motivation to go for the stars.  There is nothing wrong with continuous and incremental improvement.  Singles will win more games than homeruns, according to that great philosopher Casey Stengel.

    0
    #111466

    Dayton
    Member

    Darth,
     
    Utopian dream world?   No, you make it the reality it needs to be.   For DMAIC projects you say:
     
    Projects address a limited set of customer defined CTQ’s?
     
                Resources for dramatic changes are not provided?
     
    Project teams given 3 – 4 or more generously 6 months?
     
    Project charters target 5% or possibly up to 50% improvement?
     
    Projects target process betterment not the ultimate?
     
    If those are your accepted working constraints, then you are correct, incremental improvement is your expectation and probable maximum achievement.  I understand what you said, and I was trained with that mindset, but it does not have to be your reality.  Looking at DMAIC projects through a different lens gives you a different perspective and tact.  You help determine and sell the project need, you determine the project scope, you help determine the project objectives and target, you determine the project timeline and resource allocations, and then you make it happen.   The organizational environment needs allow the Six Sigma leader the wherewithal to directly participate in determining organizational objectives, imperatives and priorities and to lead and facilitate the creation, drive, tracking, reporting and closure of the project teams.   Some higher impact projects take longer and need additional resources – you can’t take someone else’s template of project application and apply it robotically.   My point is don’t just try to hit singles, some should be and can be homeruns.  Casey had nothing against homeruns.  I’m not offering an academic perspective, my perspective was gained making it work in a fortune 500 company. 
     Vinny

    0
    #111474

    Mikel
    Member

    Vinny, my boy, let me turn you on to a simple truth –
    I can come in tomorrow and dramatically improve any of your high impact projects. I will do it by using what you have learned. If you think you “solve” things, take a lesson from TPS and their “countermeasures”. To them it means the next step, the best thing I know today. That is all your projects are if they have reached their full potential.

    0
    #111476

    Peppe
    Participant

    Charlie,  take care of John nash theory  “.. everyone do what he, believe, is the best for himself and the group …” 
    Is your team in aligned with this ? Are you doing what, you believe, is the best for you and team ? Ask to yourself  : why this guy could help me?  
    Rgs, Peppe

    0
    #111479

    Dayton
    Member

    Stan,
     
    Oh wise leader, why is this such a chore to communicate?   Each time you work on and complete a project have to you not aimed at the best solution given the data and knowledge that you have at the time?   A big reason that you are (or the boys at TPS are) able to further optimize your original solution is that you have acquired additional insight, frequently as a result of working your original solution.   Is this semantics or are you intentionally sub-optimizing your original “fix”?  If you are sub-optimizing in an effort to meet an arbitrary 3 – 4 month window because that’s all the time that “they” will give you to complete a project based on “their” perception of project time frames you are not doing the right thing for your organization and in turn selling Six Sigma analytics short.    
     
    Vinny

    0
    #111482

    Mikel
    Member

    I have never suboptimized any solution. I have never settled for less than 50% improvement between existing results and entitlement.
    Go reread your original post, we are in agreement, but your original post said you always “solved” the problem. This is arrogant and unrealistic thinking. We just take the next step on a journey and if we are going to use valuable resources we establish a minimun gain to make it a useful business proposition.
     
     

    0
    #111495

    Dayton
    Member

    Stan,
     
    That’s because I am arrogant and occasionally unrealistic – but I mean well, never knowingly sub-optimize, and drive projects to my understanding (based on rigorously collected and tenaciously analyzed data in hand) of best-case fix.   New or better-understood data can drive subsequently heightened/increased optimization.   It’s a journey and in using valuable resources you must assure gain – not “minimum” acceptable gain.  You get what you target and little more.   Target higher to get higher, which changes the dynamics of project timing, resources required and return on project investment – acknowledge that in project charter, move forward and make it happen.
     
    Vinny

    0
    #111497

    Mikel
    Member

    I agree.

    0
    #111503

    Darth
    Participant

    Vinny, I agree that we should target high to achieve high.  But your previous post with all the capitalized you infers that most BB/MBBs do in fact have all that influence and ability to get it done.  Possibly your experiences have been better than mine, but I have seldom seen them have the ability to work with such an implied degree of control.  In fact, in a former employer, many teams were chartered to check off the box of having a project rather than any attempt to achieve significant change.  Many teams were chartered and then killed.  Many DFSS projects were started to avoid using statistics and to allow the certification of multiple GBs.  Were there homeruns, absolutely.  Was there lots of wasted time and effort, afraid so.  I think we all have agreed that aiming high should be the goal.  Regardless, further improvements can always be made as things change.

    0
    #111515

    Mike R. Hill
    Participant

    The ability of you or anyone to dramatically improve on ANY of someone elses work would depend on the definition of “improve”.  Had you issued that challenge to me (to improve my result), which I assume you would in a New York minute, I would soon prove you wrong with reference to

    the remaining profitability improvement available due to
    investment necessary to atain said “improvement”
    the capability of the competition
    the capability and affordability of current technology.
    Six Sigma is about maximizing profitability, long term.  A good DMAIC solution provides a return that recoops invested resources, not just reduced variation.
     

    0
    #111519

    Mikel
    Member

    What nonsense.

    0
    #111523

    Mike R. Hill
    Participant

    Flippancy is a hallmark of constipated predudice.  Please enlighten me.  I appear to be totally in the dark in your opinion.
    Without regard to technology, competitive advantage, or profitability, how can you improve every, or any, solution in the business world?
    Is DMAIC an obsessive pursuit of the next level of perfection, or is it about creating and sustaining a healthy business?

    0
    #111524

    Mikel
    Member

    A healthy business of course.
    Your original rearks are the hallmark of ignorance. I am prejusticed against ignorance when displayed proudly like you did. As of this morning, I am not constipated but thanks for asking.
    Your original post insinuated that your business cannot be improved without heavy investment in technology and equipment. This simply is not the case and if you believe you will get taken by people smarter than you.

    0
    #111526

    Mike R. Hill
    Participant

    Yes, I am confident in my work, but it is not me that makes me so.  It is my dedication to this methodology.  That is why I am curious about your claims of rapid and limitless “improvement”.
    Let’s cease with the name calling and bring this up a notch.  I don’t want to misunderstand you, but I AM having trouble at this point.  I’m glad you’re not constipated, but I was referring to your argument, not your intestines.
    I did not mention heavy investment.  I am pointing out that the lengths necessary to drive to higher and higher levels of Sigma are limited by the time and manhours necessary to find and implement change and controls available in the real world to sustain them.
    Maybe it’s beacuse we work in different industries. I’m a slave to the automakers.  You?  I sense that we agree on most things (except my arrogance).  Please explain your original comment about “walking into any process any improving it” as soon as the last improvement is completed (pardon the paraphrase).  That is what has me puzzled.
    Thanks,
    Mike

    0
    #111527

    Adam L Bowden
    Participant

    Hi Mike,
    Welcome to Isixsigma – you seem to have the right attitude to fit in well on this forum !
    Regards,
    Adam

    0
    #111559

    Marc Thys
    Participant

    Vinny, and all the others that replied to him,
    Here I post an “innocent” point of view on Monday – 48 hours later I see a full thread has developed. This is the power of forums!
    Well, Vinny, I still disagree with you, and for 2 main reasons.
    The first reason I already mentioned in my previous post – time limitations. There is only so much you can do in 6 months. Other people in this forum appear to agree.
    The second reason is tools. In your typical DMAIC toolbox, you will find quite a few “improvement” tools, but few “optimisation” tools. A case in point would be DOE, where a traditional DMAIC curriculum does not go beyond classic screening and refining designs – so typically no Taguchi or Response Surface Modelling. There is nothing preventing you from learning and applying these tools of course, but the complexity and, again, time needed, are an order of magnitude higher. 
    Finally, let’s not forget why I posted my point of view in the first place. Poor Charlie’s team was getting lost in A (well – probably mainly in I really). I have seen this happen in a number of cases, and most of the time it was because the team felt their solutions were not “brilliant” enough. Some of the more obvious solutions that often were a direct result of a good analysis were discarded because they were “too obvious”! That is stupid, of course.
    Other teams felt that they had not done “enough” analysis. So I ask you – how much is enough analysis? My answer would be  – when you have explained most of the problem by validated causes, in order to be able to reach your goals by eliminating the most important ones.
    Note my use of words – I did not say “explain all of the problem”, nor did I say “achieve perfection”, nor “eliminate all causes”.
    There are some very good reasons for limiting objectives, scope, time and resources in a DMAIC project. The main one is “bang for the buck”. The goal of one of the simplest tool in DMAIC is exactly that and it is a Pareto analysis (you have mentioned this tool yourself in one of your posts).
    Marc.

    0
    #111570

    Dayton
    Member

    Marc,
     
    In your typical DMAIC toolbox, you will find quite a few “improvement” tools, but few “optimisation” tools.   [hack, cough, patoowie…]   Tools are tools, use them to improve or optimize.   You and the other minimalists are too content in aiming low.   Draw the bowstring halfway because it takes longer than you are “given”?   Very unimpressive logic my friend.   
     Vinny

    0
    #111626

    cconrad
    Participant

    You can get stuck in Analyze if you failed to really identify and measure all your X’s along with the Y.  We ran into this situation where the Y was easy to measure, but the teams had to be pushed really hard to capture all the data around the Y.  You don’t yet know what X’s will become significant, so you have to go after all of them, and some are expensive and difficult to measure, and if the team has pereconcieved notions about a solution, they will hesitate at working hard to get X’s they don’t believe will help them.  Once all the X’s are measured, Analyze becomes the fun part of the project, revealing what really drives the Y.  It also has a lot to do with Define, too.  Setting up the problem, without pointing to any solution.
    Good luck

    0
    #111882

    Schultz
    Participant

    i tend to agree for analysis, which is focus on the right thing. after reading http://www.mprv.biz/iwb/practice/examples/package-design.pdf I tend to diasgree with what you write about importance of root-cause. there seem to be more routes to the exit.
    in that edample the poster asked to solve a measurment problem for the package design. this was a cause which blocked finding a better design. i can imagine activities, which try solvint this root cause by finding a better measeurment method. this can take long and can result in no design improvement at all. but design improvement is probably the more important goal of the poster. and the more pressing one.
    anyway, i learned triz is not only for inventing somethnig new. its strength is focused analysis and strong solutions /innovation/ seem to become a by product or somethig which is hard to avaid. these seem to become consequences of focused and guided thinking.
    i start to like the solution process, whih is shown on that site. i think the real power of solution oriented processes comes from focus on facts. this can be data driven focus, as in dmaic. it can be driven by understanding, as shown by the iwb tool. the many directions generated seem to be a good reality check. when they make sense, the iwb-model is probably capturing the important aspects of the problem. interesting. and the solutions are available at the same time. makes me think.

    0
Viewing 34 posts - 1 through 34 (of 34 total)

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