• I would make your data more quantitative. quantititative date is much more powerful. Instead of using old versus new can you convert to year of manufacture? Instead of season can you use month of year or julian date of manufacture. It is obviously not the date but some seasonal shift that would affect the product? Do you think this is…[Read more]

  • I worked on a project that had a plastic part inside a metal cylinder.  The requirement that we measured was how far the plastic part was recessed inside the metal cylinder.  Because these two materials have different coeffecients of thermal expansion we had to condition the assembly overnight at 65-70 degrees F.   Conditioning the part and tak…[Read more]

  • Bode replied to the topic How 6 sigma reduces cost? in the forum General 12 years, 5 months ago

    Several good points have been made in the other posts. I want to take a slightly different approach.  Since you are new to Six Sigma, I’m not sure if you were asking specifically about six sigma or if you were speaking generally and therefore a brief discussion of Lean (which is often paired with six sigma) might be appropriate. 
    Lean is about e…[Read more]

  • ASQ has some good basic histogram tools.  Check out  This page contains links to a histogram worksheet and an excel tool that makes histogram from your data.

  • Bode replied to the topic VOC Input for ISIXSIGMA in the forum General 13 years, 3 months ago

    I’m not a computer expert but I think I have done what I can to address this issue for my machine.  I have limited options to change security settings, equipment, etc.  If I were working this issue for ISIXSIGMA and only one customer was having this problem, then changing the format to address the problem would not be a high priority.  
    The que…[Read more]

  • Bode replied to the topic Indirect Costs in the forum General 13 years, 10 months ago

    We have done several events (Projects and Kaizens)  to reduce indirect expenditures.  We have identified events the wrong way i.e. drive by Kaizens (someone thinks this indirect expenditure is a problem) and the right way using a VSA approach.  My recommendation is to follow a VSA like approach.  Do a SIPOC, high level map, Pareto chart the expe…[Read more]

  • If you decide to do 5S, I would also do standard work.  It is hard to agree on what equipment is necessary, if you haven’t agreed on how to do the job.
    If you want so show some dramatic improvements, you should also consider, using some theory of constraints tools and focus on improving your bottle necks.

  • Bode replied to the topic Project Identification Process in the forum General 14 years ago

    Depending on the size of your organization and the level of commitment of your senior managers, you should consider value stream analysis and Hoshin Planning (Hoshin kanri) as ways to identify improvement opportunities.  A value stream analysis takes a high level look at a number of processes that transform raw materials into final products to i…[Read more]

  • Teresa,
    I agree wtih Ron and Brandon.  Just some additional clarification.  The business case defines the costs and benefits.  The benefits can be other than monetary savings.  Improvements in cycle time, morale, safety, etc. are legitimate benefits that are frequently hard to convert to monetary savings.  The business case allows manag…[Read more]

  • I’m an engineer and not a statistician too so I’m interested in the results.   The ANOVA test is a way to compare multiple sets of data.  It is essentially a two sample t test for more than 2 sets of data.  Minitab has an ANOVA test.
    Let me know how this works out.

  • Bode replied to the topic Rolled Throughput Yield in the forum General 14 years, 10 months ago

    I would not count the error if the person that made the error caught and corrected the error without shutting down the line.  If someone else caught or corrected the error or the line had to be stopped, then I would count this as an error with the appropriate reduction in FPY.

  • I am curious about your statement “very lowest possible human error rates are 5000 dpmo or 0.5%” 
    What is your source?  What is the context?  This is the first time I have heard this statistic.  Does this imply that only through machines can we obtain a 6 Sigma process?

  • The first 2 years we were doing Lean Six Sigma (LSS), every person (BB, GB and Team member) that fully participated on an RIE or Project got a $75 award.  My organization sets aside a set amount each year for all performance awards and the $75 came from this pot of money.  Now we are not rewarding anyone just for participation.  Doing LSS is part…[Read more]

  • Bode replied to the topic Measurement Error in the forum General 15 years, 3 months ago

    Are you now charging the customer for all of the product delivered i.e. 200 instead of 100? If not, why not?  This seems like an improvement that has a significant ROI.  Your percent scrap is down and you have a much clearer picture of what is happening in your process.
    I work for the federal government so don’t take this personnally but this s…[Read more]

  • Bode replied to the topic Multiple indicators in the forum General 15 years, 3 months ago

    All four indicators for the last 2 periods have been moving in the positive direction and 6 of the 8 points are outside the range established for the previous 5 periods.  As I stated these are not statistical trends but the indicators moved in a positive direction for each of the indicators during each of the last 2 periods.  I have done I-MR c…[Read more]

  • Bode replied to the topic seis sigma in the forum General 15 years, 3 months ago

    No estoy seguro que entienda su pregunta. Pues, si esté no contesta su pregunta, por favor dígame. Seis Sigma es sobre reducir al minimo la variación de un proceso. Sigma representa una desviación estándar. Un proceso de Seis Sigma tiene una poquita variación y debe rentar solamente
    3.4 defectos por millón de ítemes.
    Seis Sigma sigue un proc…[Read more]

  • Bode replied to the topic Charting Calibration Data in the forum General 15 years, 4 months ago

    It depends on the purpose of the data.  If you are tracking one particular type of gages then I would opt for the absolute value.  This gives you the best information about one particular gage.  If I were evaluating the calibration service/process for a variety of gages, then I would use the deviation from standard divided by the range i.e. for yo…[Read more]

  • Bode replied to the topic Project Savings Hurdle in the forum General 15 years, 5 months ago

     This is some information from a theory of constraint class: “IF 100 PEOPLE WORKED TO IMPROVE 100 DIFFERENT PROCESSES, THE 1 PERSON WORKING ON THE CONSTRAINT PROCESS WOULD SAVE THE ORGANIZATION THOUSANDS MORE THAN ALL THE OTHER 99 PEOPLE COMBINED!” This does not change my earlier post, it just indicates that the savings may not always be re…[Read more]

  • Bode replied to the topic Project Savings Hurdle in the forum General 15 years, 5 months ago

    We don’t have a specific $ threshhold.  Which project to tackle is based on which project gives the greatest Return on Investment.   The return could be financial or non-financial:  safety, morale, etc.  If a project is justified based on $ then we shoot for a 3 to 1 annual ROI.  A typical project costs 50-100K, so we try to identify projec…[Read more]

  • Bode replied to the topic p value = 0 ? in the forum General 15 years, 8 months ago

    You sound like one of my instructors.  Have you ever been to Suffolk, VA?

  • Load More