iSixSigma

Bob J

Forum Replies Created

Forum Replies Created

Viewing 100 posts - 1 through 100 (of 476 total)
  • Author
    Posts
  • #142897

    Bob J
    Participant

    Hi Andrew!
    Never heard back from Wilson so I assume all went well…  You might want to ask him directly to see if he has any details for you…;- )
    Best regards,
    Bob J

    0
    #132913

    Bob J
    Participant

    Interesting perspective….;- )
    The key is to understand the expected role of the MBB in your business and define the necessary skills based on that understanding…  For our business, we expect a MBB to be able to effectively train at a BB level and to provide mentoring support.  To do this, you need the following skills:

    Demonstrated proficiency in the tools/techniques
    Ability to effectively teach the tools and techniques
    Ability to effectively communicate at all levels of the business
    Ability to effectively mentor Black Belts and Champions
    Much of these skills cannot be taught in a two week course but are better grown over time within the business or selected from those within the BB community who have an appropriate background (trainer, teacher, senior manager etc) .
    In summary, I would be very wary of those certification programs that promise (usually for a great deal of money) an effective MBB certification in two weeks…
    My opinion is that the best programs for MBB certification are organic to the business and driven by the business expectations of the role the MBB is to play….
    Best Regards,
    Bob J

    0
    #132895

    Bob J
    Participant

    Rolando,
    You are certainly welcome….;- )
    Have a great day and good luck on the project!
    Bob J

    0
    #132862

    Bob J
    Participant

    Rolando,
    Ahhhh, I see…..  That helps….
    If I understand correctly, you are trying to improve the number of leads that result in successful sales…. 
    Given that, the first question you need to ask yourself is whether the sample (one month) is large enough and truely representative…  Is this a new telemarketing campaign or an ongoing effort? 
    Assuming the sample is ok then the next hurdle is to determine whether the results for California are statistically significant…  I’d use a chi squared contingency table to test this…
    Assuming the results from California are statistically significant, the next challenge will be to discover what factors make California leads different.  The key thing here will be to look at how the leads are initially identified and what demographics are available to you for the next level of analysis…  The path forward will be to learn what factors result in the better results in California and use that information to improve the yield for the rest of the country…
    Hope this helps…
    Best Regards,
    Bob J

    0
    #132859

    Bob J
    Participant

    Rolando,
    Not much from what you have shared…
    What you are describing is a categorization of a static set of data…  Trend analysis will help you determine how this data significantly changes over time… 
    Also, not sure what TPV is….
    What are you trying to improve?
    Best Regards,
    Bob J
     

    0
    #132854

    Bob J
    Participant

    John,
    No sure what you mean by system application…  We typically draw these by hand on a layout when looking to simplify flow…  
    Best Regards,
    Bob J

    0
    #132853

    Bob J
    Participant

    Are they going up or going down to a significant degree over time….
    Best Regards,
    Bob J

    0
    #132852

    Bob J
    Participant

    Agree completely with Horiculture’s response….  The role of the BB is to drive improvement projects to a successful conclusion…  The role of the MBB is to train Black Belts and to provide appropriate coaching/mentoring within the business.  This last is for both the business leadership on the care/feeding etc for the program as well as the technical project support for the belts…
    To be successful in this role you should be able to effectively teach as well as work with all levels of management within the program… 
    This is a very different set of skills from the typical BB role…
    Best Regards,
    Bob J

    0
    #63864

    Bob J
    Participant

    Jiten,A couple of quick thoughts….A lot of current thinking in software development involves similar sizing issues, typically for level of effort estimates…  Research there might give you some insight…Check out Albrecht function point and McCabe Cyclomatic complexity in google…  Both are fairly easy to do and far more precise than LOC (and avoids all those issues with how do you count a line of code, ie commented?, frameworks?, etc)Hope this helps…Best Regards,Bob J 

    0
    #117192

    Bob J
    Participant

    Jiten,
    The software forum might be able to get you more responses….
    A couple of quick thoughts….
    A lot of current thinking in software development involves similar sizing issues, typically for level of effort estimates…  Research there might give you some insight…
    Check out Albrecht function point and McCabe Cyclomatic complexity in google…  Both are fairly easy to do and far more precise than LOC (and avoids all those issues with how do you count a line of code, ie commented?, frameworks?, etc)
    Hope this helps…
    Best Regards,
    Bob J
     

    0
    #116996

    Bob J
    Participant

    BIG k,
    No can do…  It’s very sensitive stuff for us ….
    Sorry…
    Bob J

    0
    #116989

    Bob J
    Participant

    Tabletop,
    A lot depends on the way that you have set up your survey…. 
    A few common ones (for us anyway)….
    If you are using a likert (ordinal) scale response (1=really bad, 5 = really good etc) for you can use use a regression (ordinal logistic) to explore linkage between questions…  Particularly interesting if you use a structure where you ask for an overall sat response and then have a number of detailed second/third tier questions….
    Pareto is good for verbatim summaries…
    Exploring differences in subgroup responses (region vs region, market vs market, tech vs purchasing manager, etc) is a good area for Chi Square.  Boxplots are also useful if you want to graphically display side by side subgroup comparisons…
    Trending evaluation is a good application for a control chart…. 
    A lot of other tools can be applicable depending on what hypothesis you wanted answered when you set up the survey..   It’s that old Covey 7 habits “begin with the end in mind” thing…  It helps to structure your survey tightly around what business questions you want answered as well as how you expect to analyze and present (graphically) the results….
    Hope this helps…
    Best Regards,
    Bob J 

    0
    #58413

    Bob J
    Participant

    Jonathan,Interesting project but how do you know that providing customers with better information is going to increase your sales volume…?  This sounds like someone is trying to jump to an answer without any supporting data and that path is heavily mined….  This is also way beyond the scope a single BB project…  See if you can rescope it to either improve your quotation hit rate (landed vs bid) or improvement of your lead generation process.  Even these could be too broad a scope but you will have a better chance to increase sales volume with projects at this level over what you were given originally…My 2/c…Best Regards,Bob J  

    0
    #116798

    Bob J
    Participant

    Paul,
    Glad it helped…  Check out some of the stuff by Watts as well…  While targeted at development life cycle maturity, it provides a neat concept that can be adapted to assessing the maturity of any key business process…  You would just have to restrucure it into the context of the key business process areas that your survey work identifies as critical…
    One last thought….  Be sure you build into your survey an “overall satisfaction” question with a likert or similar response…  It will give you a chance to do some pretty cool stuff with regressions against the lower tier questions…  Much better than asking the customer/employees to rate what they think is important… 
    Good luck on the project!
    Best Regards,
    Bob J

    0
    #116796

    Bob J
    Participant

    Hi Paul,
    Sounds like an interesting project…  A couple of thoughts…
    Survey is a good start but keep in mind that you are probably targeting the survey(s) at current customers/employees…  This will be a relatively limited response.  It would be better if you could expand the scope to include customers that have stopped doing business with you as well as potential customers.  While much harder to obtain, the results can provide an interesting insight into why new customers would want to do business with you as well as why you lost customers in the past…  Key considerations if you want to keep current customers and attract new ones..
    While targeted at the software/systems development industry, I still consider Watts Humphrey’s work (CMM/CMMI/PSP/TSP) to still be one of the best models out there in basic structure…  The overall assessment process rates the maturity of major areas of the business on a 1 to 5 scale which runs from 1=Ad hoc to 5=Optimized..  It is currently heavily biased towards software/systems development life cycle stuff but the basic concept could be adapted to what you are trying to do…  Lots of stuff can be found in this area with a Google…. 
    Hope this helps….
    Best Regards,
    Bob J

    0
    #116666

    Bob J
    Participant

    Don,
    The short answer would be in the middle somewhere….   A lot of how you define “somewhere” will depend on the specific metric details… 
    For example, if the metric is critical to the customer and you are considered best in class, you should go for the value represented by as large a percentage of the customer responses as you feel you can achieve…   If you are not best in class then you should shoot for at least the best in class value…
    If the metric is not critical but a clear differentiator in your market you may want to go with a median value of responses…
    You may also want to consider using the Pert Weighted Average (= (Optimistic + 4*Average + Pessimistic)/6) techique developed to estimate duration for a single activity….  The approach would work as well in this context…
    Hope this helps…
    Best Regards,
    Bob J
     
     

    0
    #116624

    Bob J
    Participant

    CertGuy,
    What do you mean by “legally defensible”? 
    As long as you do reasonable due diligence to ensure that you aren’t plagerizing someone else’s work I can’t imagine have a problem unless you are looking at something beyond what I would normally expect…
    Best Regards,
    Bob J

    0
    #116619

    Bob J
    Participant

    Jonathan,
    Interesting project but how do you know that providing customers with better information is going to increase your sales volume…?  This sounds like someone is trying to jump to an answer without any supporting data and that path is heavily mined…. 
    This is also way beyond the scope a single BB project… 
    See if you can rescope it to either improve your quotation hit rate (landed vs bid) or improvement of your lead generation process. 
    Even these could be too broad a scope but you will have a better chance to increase sales volume with projects at this level over what you were given originally…
    My 2/c…
    Best Regards,
    Bob J
     
     

    0
    #116471

    Bob J
    Participant

    Kirk,
    No problem…  Seems like you have a good working understanding of how best to approach product reliability tracking and assessments…. 
    Good stuff…
    Best Regards,
    Bob J

    0
    #116465

    Bob J
    Participant

    Kirk,
    Are you a reliability guy?
    Best Regards,
    Bob J

    0
    #116432

    Bob J
    Participant

    Eve,
    If you are counting defectives, you can use p-Za/2*SQRT(p(1-p)/n) < P < p+Za/2*SQRT(p(1-p)/n)…
    You will need to convert to DPM afterwards but that shouldn’t be a problem…;-)
    Hope this helps…
    Best Regards,
    Bob J

    0
    #116429

    Bob J
    Participant

    RG,
    If you have a date code you might be able to evaluate the Time To Failure trending by looking at how long the product was in the field before it was returned…   
    Just a thought….
    Best Regards,
    Bob J 

    0
    #116339

    Bob J
    Participant

    No Maintenance Man Here,
    From a narrow perspective you should use Availability (vice MTBF or MTBR) which factors in MTTR…  OEE is even better…
    Hope this helps….
    Best Regards,
    Bob J

    0
    #116060

    Bob J
    Participant

    Vinny,
    No problem…  Our development guys use this as a way to show that a new (or revised) product design meets the predicted reliability targets… 
    Of course we also have a serious derating program so the actual field performance averages almost 4 times the predicted values….
    Good stuff…;-)
    Best Regards,
    Bob J

    0
    #116041

    Bob J
    Participant

    Steven,
    Sounds like predictive MTBF (Mil-HDBK-217) is the way to go…   As Vinny pointed out, it’s pretty easy to do…  Take the EBOM and map it against the failure rates (FITS=Failures in Time) in Relex…  For example, most resistors and caps are 5 FITS, diodes are 20 FITS etc…  The active silicon stuff (RAM, Arrays etc) are typically 100-300 FITS… 
    After you’ve mapped the EBOM, add up the total FITS…  This is the total predicted failure rate per 1000000 hours.  The model was originally developed for the Military so there are scaling factors that can be applied for especially rigorous environments…  Since I assume that you are working with commercial grade products, the scaling doesn’t apply. 
    Your predicted MTBF is one over your total FITS…
    Good luck….  If you have any other questions and want to take this offline you can reach me at [email protected].
    Best Regards,
    Bob J

    0
    #116022

    Bob J
    Participant

    Steven,
    A couple of thoughts on what I think you are trying to do….  If you have a product that you are reasonably sure will come back to you when it fails and you have a large installed base, you might be able to do some rough MTBF calculations…
    Before I go too far, a couple of quick questions….  How familiar are you with Weibull and do you have minitab…
    Best Regards,
    Bob J
     

    0
    #116021

    Bob J
    Participant

    Steven,
    For most applications we use excel or Minitab…  Relex is good for reliability prediction work but may be significant overkill for what you are trying to do…
    Best Regards,
    Bob J

    0
    #116019

    Bob J
    Participant

    Steven,
    DPM in the traditional sense does not convert to MTBF…  DPM is defects/defectives per million opportunities….  MTBF is  failures in time at a point where you have a constant failure rate (not infant mortality or wearout ends of the bathtub curve)….
    Give me a better sense on what you are trying to do and I’ll see if I can help….
    Best Regards,
    Bob J

    0
    #115864

    Bob J
    Participant

    Peppe,
    I understand…  Each of the major projects here follow DMAIC (even the lean stuff)…  The difference (for us) in in the MAIC with the Define making sure that we are focused on the “big hitter” opportunities…  Some of these end up being a mix of tools (Lean/SS) and some end up being almost pure SS…  We’ve cross trained everyone so (though skills vary) each one has a reasonably complete tool set and a “who do you call?” list within the network if they need something a little more specialized…
    We don’t require specific tools to be used at each of the steps but rather use a logic that says (for example):
    Measure;  Does a measurement system exist that accurately measures the project opportunity ($, time, defects, etc) and has a baseline been established.
    In this example, we sometimes use an R&R and sometimes not…  Depends on the metric and what we know about the measurement system…
    Keep in mind that this is what we have done…  As we used to say down south, “There are many ways to skin that cat!”….;-)
    Hope this helps….
    Best Regards,
    Bob J

    0
    #115857

    Bob J
    Participant

    Hi Jay,
    Lay out your materials and get a good nights sleep… 
    Tomorrow will be a long day….;-)
    Best Regards,
    Bob J

    0
    #115856

    Bob J
    Participant

    Peppe,
    Agreed for throughput and flow opportunities…  I’ve had other projects that are pure SS…. 
    For example, we had a significant customer issue with one of our products…  This customer experienced a number of process trips that accounted for roughly $10M in lost production due to false positive inputs from this particular group of products…  The “defective” products came back and we were unable to duplicate the problem in our labs…    We did have several clues that we were able to build a hypothesis on…  The majority of the trips occured during the monsoon season so we suspected humidity was a factor…  This failure mode was unique to this site and the site was situated on the coast leading us to suspect salt was a factor…  In addition, the problem “went away” when it was returned to us, leading us to suspect the environment at site…
    The product has two barriers that are designed to prevent moisture from getting into electronics (our hypothesis).  These are potting and conformal coating on the board…  We devised a doe where we varied the potting and varied the conformal coating and put the products in a salt fog chamber.  The output characteristic was how long they survived before going belly up….
    The results showed that the problem was associated with an interaction between the conformal coating on the board and the potting..  The solution ended up being to remove the conformal coating…  We built test units and sent to site…  To make a long story short they worked great…  No trips…
    We ended up revising our product so the conformal coating was removed from the design…  Cost came down…  Our vendor was happy (conformal coating is so much fun..;-)…  Our customer was happy…
    Couldn’t have done it without the DOE….  The answer was not one that we would have reached in a million years by intuition… 
    Bob J

    0
    #115844

    Bob J
    Participant

    Peppe,
    No problem….;-)
    Its all about what you need to accomplish and what the best tool is to accomplish it…  It’s all about how to most efficiently identify the areas of improvement, make the necessary changes, evaluate and then move on…  I agree that Lean requires less training and gives great results for process flow, capacity and throughput time issues….  The problem is that some problems don’t fit nicely into this box and need a little different approach…. 
    For example, one of my original certification projects was associated with a new product that was having horrible scrap and throughput issues…  With a series of kaizens we were able to take the average throughput time from 3.5 weeks to 2 weeks, effectively doubling the capacity of the line…  One of the biggest remaining parts of the process (time wise) was a temperature conditioning that was critical to our final yield…  Three DOEs enabled us to reduce this from 5 days to 3 hours with an associated yield improvement from 80 to 96%.  Couldn’t have done this without the right tool…. 
    I think arguing one tool set over another is like arguing whether the brain or the heart is more important to your body…  Both are necessary if you want to be truely effective…
    My 2/c
    Bob J
    It all depends on

    0
    #115765

    Bob J
    Participant

    Peppe,
    My mistake…  I thought you wanted logic and context…  Without that any answer is pretty much meaningless…
    Best Regards,
    Bob J
     

    0
    #115757

    Bob J
    Participant

    Peppe,
    I think it’s a pretty narrow view….  From my experience they’re talking apples and oranges…
    Most Kaizens etc I’ve been involved in are pretty low on tools and high on the intuitive analysis….  What looks like muda often is muda and so on…  Getting the right team to stop and think etc can generate some great breakthroughs…
    Six Sigma in my company is more focused on the stuff that doesn’t lend itself much to the intuitive side.  Many of the projects that I have been involved in are targeted at processes with factors that are anything but intuitive…  If we have one that looks intuitive it goes in the JDI bucket and doesn’t (usually) end up on the docket…
    This doesn’t mean that we don’t have “tool zombies” but that is something that is on the radar and (post certification) is not much of a problem…  Results (and successful projects) are the only thing that really matters and most of our BBs use only those tools necessary to successfully resolve the project opportunity…  
    My 2/c
    Best Regards,
    Bob J

    0
    #115619

    Bob J
    Participant

    BBC,
    Focusing on forecasting is BS…  Any forecasting tool is wrong and the error is a function of time… 
    Order fulfillment time is what is critical….  Map your process and work with your vendors to take out the muda and improve flow…  Reduce your inventory accordingly…
    You’ll get a lot more actual results in far less time…
    My 2/c
    Bob J
     

    0
    #115530

    Bob J
    Participant

    Miranax,
    Probably half of our BB classes are engineers and the catapults are always a hit ;-)
    Best Regards,
    Bob J

    0
    #115529

    Bob J
    Participant

    Roshan,
    To amplify a bit on Mike’s comments….  What is the business context or put another way, what do you expect the RFID implementation to improve/fix in the business…  Why does the business care?  Answering this question usually puts you on a good path to discover a good primary metric…
    Hope this helps…
    Best Regards,
    Bob J

    0
    #115528

    Bob J
    Participant

    Ron,
    You are correct….  Small samples sizes tend to “fail to reject” just as *very large* sample sizes tend to reject the null hypothesis… 
    Best Regards,
    Bob J

    0
    #115231

    Bob J
    Participant

    Luk,
    Darth is right… 
    Find out at what point the wheels fall off your process/product and set your specification accordingly…;-)
    Best Regards,
    Bob J 

    0
    #115230

    Bob J
    Participant

    Hanna,
    This might help….
    http://www.qualityadvisor.com/library/capability/capability-menu.htm
    Best Regards,
    Bob J

    0
    #115224

    Bob J
    Participant

    Joshua,
    Based on the data and your analysis model the best estimate is what is listed under the “Standard Percentile” column…  Since this is an estimate based on a finite sample the CI gives you the probable bounds on the estimate… 
    Bottom line is to use the estimate…
    The change in the CI (as well as the Error) for the 99 percentile value could be associated with a poor fit between the model and the data…   You might want to check a few different distributions to see if you can get a better fit…
    Hope this helps….
    Best Regards,
    Bob J  

    0
    #115082

    Bob J
    Participant

    PS,
    A negative Z value indicates that you have more than 50% out of specification….
    Hope this helps…
    Best Regards,
    Bob J

    0
    #115081

    Bob J
    Participant

    Tim,
    Nicely summarized….
    Best Regards,
    Bob J

    0
    #115041

    Bob J
    Participant

    BB without MBB,
    I guess it depends on the needs of your six sigma deployment in your company…
    The role of the MBB in my organization is that of a trainer and BB mentor…  This is important to us to ensure that we effectively manage the program as it fits our business needs….
    If you have access to resources (internal or external) that can effectively enable you to manage this aspect of your program, it should not matter if they are MBB or not…. 
    Hope this helps…
    Best Regards,
    Bob J

    0
    #114887

    Bob J
    Participant

    Mauricio,
    You are certainly welcome……
    Best Regards,
    Bob J

    0
    #114857

    Bob J
    Participant

    Mauricio,
    Thanks!  That helps me understand your situation….
    You should run the R&R on one color at a time….   Also, since I assume you have *many* colors it might be worth focusing your efforts on the ones which you think might be a problem….
    Given your answer I need to revise my earlier response to you since I jumped to a number of assumptions without understanding your situation….   The real answer is “it depends” and this is why…
    You need to collect your samples in such a way as to fully capture the full variability present.  If the main variability exists within a single batch such as positional (top of vat vs bottom or center vs edge) or time (lighter when first mixed and darkens later) then you want to sample accordingly….  If the main variability is batch to batch then you will need to sample multiple batches… 
    Sorry for the confusion but I hope this helps…
    Best Regards,
    Bob J

    0
    #114823

    Bob J
    Participant

    Mauricio,
    The purpose of the R&R is to evaluate the variance in the measurement process….  Given this, the samples should contain the full range of measures you would expect to see when the measurement system is in operation…  These samples should be hand selected so that you have the full range… 
    Batching is not a player here unless you need to go to different batches to select your samples so that you can fully exercise the measurement range….
    I’m not sure what you mean by different references so can’t comment on that for you….
    Hope this helps….
    Best Regards,
    Bob J

    0
    #114822

    Bob J
    Participant

    PM,
    I would have to say the ability to deliver full functionality on time and on budget….
    Hope this helps….
    Best Regards,
    Bob J

    0
    #60462

    Bob J
    Participant

    GDS,
    Thanks!  Thought so but sometimes it’s worth mentioning the less flashy (nonstats) stuff that helps projects stay on track and promote clarity…  Sometimes they get lost in the shuffle….
    Have a great weekend!
    Best Regards,
    Bob J

    0
    #60458

    Bob J
    Participant

    Ramesh,
    You don’t seem to be getting a lot of response so I’ll give it a shot…  My apologies for the length of the post…
    I recommend that you first consider what “improvement” means for your area…  What is important to your customers…  For example, I would assume that for an emergency process one thing that would be very important is the amount of time it takes from patient arrival to triage.  Another important consideration is accuracy of triage assessment.  Etc.  I like to approach this using a tool called an affinity diagram in a session composed of a broad cross section of folks that are familiar with the dept.
    In Six Sigma these are called CTQ’s….  Critical to Quality…  These are very important because they become the “keep the eye on the ball” focus that keeps your eventual project aligned with the overall business drivers…  Once you have a good idea of your CTQ’s, I like to gather data that shows how we are doing on each of these…  This assessment should be kept as simple as possible…  This data should then be reviewed with your organizational leaders with a focus on coming away with which CTQ you should be working on that will have the greatest impact on the operation…  It’s important here to come away with a defined ranking and not a “everythings important” response.  You want the one thing you should be working on…  Not a huge laundry list…
    With this understanding you should be able to come up with a project charter and get it signed off.  There are lots of threads and templates available in the forum for this if you need….  This is important so that you have a solid understanding of the scope and focus of your project to help prevent “scope creep” later…
    With this in hand I like to “walk” the process…  I find this works better than starting off with a team consensus of the process because sometimes things are not exactly what everyone thinks they are…  In your case (assuming you are working to improve time to triage) I would sit in the ER in a quiet out of the way corner and watch (and note) what steps occur when a patient arrives…  If time is your metric, note how long they wait before someone initially sees them and then how long they wait before being seen by triage….  Observe the flow and use this observation to help you define your primary process steps.  Be sure to capture the full range (assuming this is in your project scope) of patient situations (hurry cases to minor problems).
    Once you’ve defined your process and noted the key activities involved, this is where (finally) you can do your stakeholder analysis.  Ask yourself (or your team) to list everyone who either is directly involved with the process you just defined as well as anyone who might have an interest in how efficient that process is.  That list will be your stakeholders…
    What I just described for you is Define and part of Measure in the DMAIC process…
    Hope this helps…..
    Best Regards,
    Bob J
     

    0
    #60457

    Bob J
    Participant

    Jackie,
    I’m not sure if GDS used one but we require operational definitions of the measurement process as part of measure.  This ensures that everyone involved fully understands exactly what is being measured and how it is bounded.  For example, an operational definition for cycle time in this situation could be:
    Cycle time is defined as the length of time between the receipt of the request as indicated by the date/time stamp on the record of the previous process step completion and the completion the process step as indicated by the date/time stamp on the record.
    We’ve found this to be of benefit to us….
    Hope this helps…
    Best Regards,
    Bob J 

    0
    #58351

    Bob J
    Participant

    Rich,
    You’ll need to post your email for him to respond….
    Best Regards,
    Bob J

    0
    #58350

    Bob J
    Participant

    JNV,
    I understand your confusion because you are getting such mixed signals….  In my company, anything that directly impacts the bottom line is considered a hard savings….  Since the interest savings meet this criteria, they would be considered hard savings… 
    The problem is in the definition of hard and soft savings…  Our companies must define it differently… 
    Best Regards,
    Bob J

    0
    #58349

    Bob J
    Participant

    Mike,
    Nicely done….  Really cuts to the meat of the issue….
    Best Regards,
    Bob J

    0
    #114594

    Bob J
    Participant

    Ray,
    You may want to post your question on the healthcare forum….  They may be able to give you a response to your question…
    Best Regards,
    Bob J

    0
    #114592

    Bob J
    Participant

    Jm,
    Thanks!  That helps…..  It sounds like you need a mix of “tool/technique” books as well as at least one good change management reference.  You also will need at least one good “bridge” reference that shows how many of the tool/techniques apply in a heavily transactional (paper/information flow) environment.
    Breyfogle is a great “one stop” reference for many of the DMAIC tools…  Good reference stuff and worth having on the bookshelf…. 
    For a good overall DFSS book I like Design for Six Sigma by Kai Yang and Basem El-Haik.  Heavily manufacturing product focused but still worth having…
    For a “lighter” overview of statistics I would add The Cartoon Guide to Statistics by Larry Gonick and Woollcott Smith.  Good stuff and much less daunting than some of the other texts… 
    Lean Six Sigma for Service is a good book to show examples of how Six Sigma applies to the transactional world….  
    For a good overview of lean tools/techniques I like Lean Production Simplified by Pascal Dennis but be aware that it is heavily manufacturing focused and will require significant interpretation…
    For a good change management overview I would add Leading Change by John Kotter.  Good practical stuff on rollout and dealing with resistant cultures….
    Hope this helps….
    Best Regards,
    Bob J
     

    0
    #114576

    Bob J
    Participant

    Jm,
    Unfortunately, without a business and cultural context it’s hard to comment…  If you could provide some background on the target audience for this material as well as the efforts (projects) you are trying to enable it might help us comment appropriately…
    Best Regards,
    Bob J

    0
    #114372

    Bob J
    Participant

    Farrah,
    If you are looking for the resulting weighted average for the two call centers it would be (859147*37+554146*49)/(859147+554146) = 41.7 seconds….
    Best Regards,
    Bob J

    0
    #114370

    Bob J
    Participant

    Jimbo,
    I can’t cite my reference but I was taught to use a minimum sample size that gave me np>5….
    Best Regards,
    Bob J

    0
    #114233

    Bob J
    Participant

    Saranya,
    The Sigma Calculator at the top right of the page will calculate the value for you….
    Hope this helps….
    Best Regards,
    Bob J

    0
    #114229

    Bob J
    Participant

    Wally,
    Thanks!  Much appreciated!
    Best Regards,
    Bob J

    0
    #114224

    Bob J
    Participant

    Wally,
    Never mind…. 
    Mike cited a good reference in another related thread…. https://www.isixsigma.com/forum/showmessage.asp?messageID=63228
    thanks anyway….
    Best Regards,
    Bob J

    0
    #114223

    Bob J
    Participant

    Hi Wally,
    Nice summary….  Would you mind citing your source(s)?
    Best Regards,
    Bob J

    0
    #63778

    Bob J
    Participant

    Gchin,
    No problem….  Glad to help….
    Best Regards,
    Bob J

    0
    #114086

    Bob J
    Participant

    sree,
    I assume that you mean the ASQ BB Certification test….  You can find detail on the tested body of knowlege and application requirements at http://www.asq.org/cert/types/sixsigma/index.html
    Hope this helps….
    Best Regards,
    Bob J

    0
    #114028

    Bob J
    Participant

    Hi Ruth!
    I noticed that you haven’t gotten a reply yet so will give it my 2 cents….
    The GR&R will give you a measure of how much variation is associated with the measurement process…  This is related to the gage precision….  To do this properly you need to run the classic GR&R using one measurement system (gage) and at least 2 people.  Since you have already completed one gage GR&R you should run the same evaluation on the second gage to be able to accurately contrast their performance….
    The second area is that of accuracy…  A GR&R will not help you here…  To accurately assess accuracy you need to look at how the measurement system (gage) performs against a known standard.  Most cal labs (inhouse or outhouse) should be able to verify the accuracy of the gage(s) in question…. 
    Since you are planning to contrast the accuracy and precision of both gages you also will want to use similar evaluation processes (people, samples etc) for both.
    Hope this helps…..
    Best Regards,
    Bob J

    0
    #58303

    Bob J
    Participant

    JNV,
    Sure….  Any tax savings should directly reflect on the bottom line and so would be classed as “hard” savings….
    Best Regards,
    Bob J

    0
    #113585

    Bob J
    Participant

    Valerie,
    Quite the contrary….  A good business professional who understands the concepts and tools of quality can be very effective in driving real sustainable improvement in the business. 
    It’s all a question of “head set”….  Sometimes it’s actually easier to teach business professionals the concepts and tools of quality than it is to help engineering professionals reach a viable understanding of how the overall business works…
    In Malcolm Gladwell’s book The Tipping Point, he discusses the importance of people who function as “connectors” in driving cultural change and acceptance….  These people understand both cultures and are able to bridge between the two by interpreting and communicating how one “fits” into the other.  These are people who (for Six Sigma) can say, “This is how Six Sigma works in our particular part of the business”.   These folks are particulary valuable when you move to the “Big Q” quality that reaches across the entire business as opposed to doing a strictly manufacturing (Little q) quality improvement focus….
    Hope this helps….
    Best Regards,
    Bob J

    0
    #113487

    Bob J
    Participant

    Darth,
    Good to see you up and about….  Your insight is always appreciated…
    Have a great weekend!
    Best Regards,
    Bob J

    0
    #113484

    Bob J
    Participant

    Ripley,
    Depends a lot on the characteristics of the data… 
    If you are tracking actual times (5.321 days for example) then it is continuous…. 
    If you count entire days (5 days) then it is discrete…. 
    If you have a very large sample, as Darth pointed out, it’s really discrete but can be treated as continuous without too much error…
    Hope this helps…
    Best Regards,
    Bob J

    0
    #113472

    Bob J
    Participant

    Hi Ripley,
    Since it’s count data it’s discrete…
    Have a good weekend!
    Best Regards,
    Bob J

    0
    #56524

    Bob J
    Participant

    Ganesh,
    My experience with GR&R is using 3 people for the study with the focus on understanding the amount and sources of “noise” in a measurement system. 
    I’m not sure if I can comment further without knowing more about what your are trying to measure and why you thought training would help…  If you could provide a little more detail, it would help me to better understand what you are trying to do….
    Best Regards,
    Bob J
    Bob J
     

    0
    #113236

    Bob J
    Participant

    Brian,
    Glad to help…  Have a good weekend!
    Best Regards,
    Bob J 

    0
    #113227

    Bob J
    Participant

    Hi Bryan,
    This link should help     http://faculty.vassar.edu/lowry/vsord.html
    Best Regards,
    Bob J

    0
    #56519

    Bob J
    Participant

    Spyros,
    No problem….  Glad to help….  I would offer my opinion that the “New to Six Sigma” is as good an overview as I’ve seen anywhere…
    Good luck!
    Best Regards,
    Bob J

    0
    #56517

    Bob J
    Participant

    Spyros,
    A good place to start is the “New to Six Sigma?” link on the left…  Lots of good overview and comparison stuff….
    Best Regards,
    Bob J

    0
    #113119

    Bob J
    Participant

    Joseph,
    For the basics, I like The Cartoon Guide to Statistics by Larry Gonick and Woollcott Smith….
    Best Regards,
    Bob J

    0
    #63762

    Bob J
    Participant

    gchin,
    Using Six Sigma for a simple project is a bit of overkill but if the intent is to explore the (DFSS) structure and tools I would recommend using the Design for Six Sigma, A Roadmap for Product Development book by Kai Yang and Basem El-Haik.  It gives you a good overview of the DMADV process and tools….
    Best Regards,
    Bob J

    0
    #63761

    Bob J
    Participant

    Hi John,
    You will need to gather 25 subgroups after you revise the process before you can re-calculate the control limits.
    Best Regards,
    Bob J

    0
    #112104

    Bob J
    Participant

    Bach,
    We ran a series of projects around optimizing a captive plating line…  In one of the projects we changed over a Cyanide Cad process to a vastly more environmentally friendly ZnCo process…  This new process required a series of titrations to determine various component and contaminate levels…. 
    The titration process was destructive to the sample…  We also qualified our parts and optimized our process using a salt spray technique where sample parts were placed in a corrosive environment that degraded the finish.  The critical characteristic was how long the part was able to withstand the accelerated life environment without showing rust.  This was also a destructive test…  In both cases, the best we could do for the R&R was to try and run a group of similar samples…
    Hope this helps….
    Best Regards,
    Bob J
     
     

    0
    #112102

    Bob J
    Participant

    Riaz,
    The test simulation software that QCI offers may help…  It has similar questions to those found on the test and times your responses. 
    It also provides a results summary showing those areas in the BOK where your responses are weakest…. 
    Hope this helps…
    Best Regards,
    Bob J

    0
    #111923

    Bob J
    Participant

    Holly,
    My company decided not to bonus a percentage of the savings because we felt it drove a lot of the wrong behavior.  It makes it hard for a team or process owner to support the effort when someone else gets the bonus… 
    We also found that all projects are not created equal in potential savings and the amount of the savings did not necessarily correlate to the level of effort by the BB or the BB capability
    Hope this helps…
    Best Regards,
    Bob J

    0
    #111687

    Bob J
    Participant

    Nanni,
    Thanks!  I have roughtly a dozen reliability engineering/modeling references and it’s always the first book I reach for when a question comes up… 
    Good stuff!
    Best Regards,
    Bob J

    0
    #111684

    Bob J
    Participant

    Arun,
    Another good “basics” book is Practical Reliability Engineering by Patrick O’Connor.  Good coverage of the material and clearly (minimum jargon) written…
    Hope this helps….
    Best Regards,
    Bob J

    0
    #111683

    Bob J
    Participant

    Sumit,
    It’s due to sample to sample variation with is impacting the variation in the crossed study…  See mjones’s excellent post…
    Best Regards,
    Bob J

    0
    #111617

    Bob J
    Participant

    Sumit,
    Note that both answers (Marc and myself) hinge on whether the sample testing is destructive in nature.  If the same exact sample can be retested over and over again, then Marc is right and a crossed GR&R is your best choice.  If the tested sample is destroyed by the test process and another sample must be used for additional test, then a nested GR&R is your best choice…
    Hope this helps….
    Best Regards,
    Bob J
     

    0
    #111472

    Bob J
    Participant

    Doug,
    You should be able to use either Mood’s Median or Kruskal-Wallis tests….  Neither require HOV but do need the samples to be independent and the population distributions to have the same shape…
    Best Regards,
    Bob J
     

    0
    #111471

    Bob J
    Participant

    Sumit,
    Since the analysis is destructive in nature you should use the GR&R (Nested) for your study…
    Best Regards,
    Bob J

    0
    #111183

    Bob J
    Participant

    Jaime,
    Plot the data on a control chart….  If you see out of control data points then the process is unstable…
    Best Regards,
    Bob J

    0
    #111112

    Bob J
    Participant

    Shawn,
    It is bad… 
    The fact that you have a process that is not in control means exactly that there are factors that are significantly impacting your outputs that are unknown, unmonitored and unmanaged.  That being the case you can never count on any consistant performance level or drive any significant improvement….
    Best Regards,
    Bob J

    0
    #110987

    Bob J
    Participant

    bb,
    Actually the boxplot whiskers each represent a quartile of the data…  Outliers are tagged as those values which exceeds the box by more than 1.5 the interquartile range…. 
    Best Regards,
    Bob J

    0
    #110985

    Bob J
    Participant

    Chelle,
    You actually should do both….  The GR&R will tell you how much variation exists in your measurement system and the calibration will tell you whether the output correlates to some known standard…
    Best Regards,
    Bob J

    0
    #110982

    Bob J
    Participant

    Chelle,
    That is correct if you are working with means (t test) and you want to determine whether or not you want to assume equal variances when running the test. 
    Otherwise, you can use the F test as a stand alone to assess two populations to see if there is a significant difference in the respective variances.
    Hope this helps….
    Best Regards,
    Bob J

    0
    #110920

    Bob J
    Participant

    Les,
    The greatest impact of downtime on the business would be at the primary bottleneck on your value stream… 
    I would start there…
    Best Regards,
    Bob J

    0
    #110915

    Bob J
    Participant

    Chelle,
    You should use the F test when you have variable data and want to compare the variance (or associated standard deviation) between sample sets….
    You should use the t test when you have variable data and want to compare the means of two sample sets…
    Practically, the F test is applicable when you are trying to reduce variation and the t test is applicable when you are working to shift (or stabilize) the mean.
    Hope this helps…
    Best Regards,
    Bob J

    0
    #110914

    Bob J
    Participant

    Brian,
    Check this out….  It may help for your second question (assuming DMAIC)….  https://www.isixsigma.com/library/content/c020617a.asp
    As for your first question, the statistical software is primarily supporting the higher end tools (GR&R, Hypothesis Testing, DOE etc) so you see most of the application in the Analyze phase with a smaller number of applications in the other phases…
    Hope this helps…
    Best Regards,
    Bob J

    0
    #110863

    Bob J
    Participant

    Harpo,
    We’ve built it into the training material rather than have a manual.  We’ve found it a fun and hands on way to cover the application of some of the tools..
    The use of the “statapult” is relatively straight forward….  I think a manual did come with ours (NCMR) but we didn’t use it… 
    Hope this helps….
    Best Regards,
    Bob J

    0
    #110859

    Bob J
    Participant

    Harbo,
    Yes…  We use them for both GB and BB training….
    Best Regards,
    Bob J

    0
    #110833

    Bob J
    Participant

    Dean,
    I suppose you could set it up as a DOE (3 factor, 2 Levels) with output as your response… 
    The question I would ask myself would be why would you want to use a DOE instead of the specialized tuning software that is commonly available?
    Hope this helps….
    Best Regards,
    Bob J

    0
    #110326

    Bob J
    Participant

    Pat,
    The logic use of the Weibull distribution to model a data set is like the use of any other distribution in that it depends on the overall data characteristics and the fit.  No golden rule here….  
    If you have Minitab14 use Stat<Quality Tools<Individual Distribution Identification and look for the best fit (lowest AD)..  If it's Weibull, go with it….
    Best Regards,
    Bob J 

    0
Viewing 100 posts - 1 through 100 (of 476 total)